JP4211514B2 - Exhaust gas purification device for internal combustion engine - Google Patents

Description

【０１３９】
ここで、ｗＳＴは機関定常運転時に算出されたＮＯ_Ｘ浄化率ＥＦＦ又は補正値ＥＦＦＣの重み付け係数，ｗＮＳＴは機関過渡運転時に算出されたＮＯ_Ｘ浄化率ＥＦＦ又は補正値ＥＦＦＣの重み付け係数をそれぞれ表している。即ち、本発明による第１実施例では、ＮＯ_Ｘ浄化率ＥＦＦ又は補正値ＥＦＦＣが算出されたときに機関定常運転が行われているか機関過渡運転が行われているかに応じて重み付けられた加重平均値ＥＦＦＡＶが算出される。この場合、機関定常運転時に算出されたＮＯ_Ｘ浄化率ＥＦＦ又は補正値ＥＦＦＣのほうがより信頼できることを考えると、重み付け係数ｗＳＴは重み付け係数ｗＮＳＴよりも大きくされている。
【０１４０】
続くステップ２２４では、ＮＯ_Ｘ浄化率平均値ＥＦＦＡＶの算出回数を表す平均値算出回数カウンタｎＡＶが１だけインクリメントされる（ｎＡＶ＝ｎＡＶ＋１）。続くステップ２２５では、係数ｋｍｉ（ｉ＝１，２，３，４）が次式に基づいてそれぞれ算出される。
【０１４１】
ｋｍ１＝ｋｍ１＋ＱＦ
ｋｍ２＝ｋｍ２＋ＱＦ^２
ｋｍ３＝ｋｍ３＋ＱＦ・ＥＦＦＡＶ
ｋｍ４＝ｋｍ４＋ＥＦＦＡＶ
本発明による第１実施例では、ＮＯ_Ｘ浄化率平均値ＥＦＦＡＶが算出される毎に係数ｋｍｉが更新され、従ってＮＯ_Ｘ浄化率平均値ＥＦＦＡＶが算出される毎に一次式ｍが決定されているということになる。
【０１４２】
続くステップ２２６では、給油判定ルーチンが実行される。この給油判定ルーチンでは図３０を参照して後述するように、給油が行われたか否か、より正確にはイオウ濃度の異なる燃料が給油されたか否かが判定される。続くステップ２２７では、現在の燃料消費量ＱＦがＱＦＰ２として記憶され、積算値ＳＳＴ，ＳＮＳＴ、カウンタｎＳＴ，ｎＮＳＴ，ｎＥＦＦがそれぞれクリアされる。
【０１４３】
これに対し、ｎＥＦＦ＜ｎＬ２のときにはステップ２２２からステップ２２７にジャンプする。即ち、前回のＮＯ_Ｘ浄化率平均値算出時期から今回のＮＯ_Ｘ浄化率平均値算出時期までの期間内にＮＯ_Ｘ浄化率ＥＦＦ又は補正値ＥＦＦＣが算出された回数が下限値ｎＬ２よりも少ないときには、この期間におけるＮＯ_Ｘ浄化率平均値ＥＦＦＡＶが算出されない。なお、単位時間当たりにＮＯ_Ｘ浄化率ＥＦＦ又は補正値ＥＦＦＣが算出される回数はＮＯ_Ｘ処理が行われる時間間隔ｔＮに応じて定まるので、浄化率算出回数カウンタｎＥＦＦが下限値ｎＬ２よりも大きい場合もあれば小さい場合もある。
【０１４４】
次に、ステップ２３６（図２０）で実行される流入ＮＯ_Ｘ濃度補正値ＣＮＩＣの算出ルーチンを図２３を参照して説明する。図２３を参照すると、まずステップ２７０では、現在の機関運転状態即ち要求負荷Ｌ及び機関回転数Ｎに基づいて図７のマップから流入ＮＯ_Ｘ濃度ＣＮＩが算出される。続くステップ２７１では遅れ時間ＤＬＹが算出される。続くステップ２７２では、次式（１）に基づいて流入ＮＯ_Ｘ濃度補正値ＣＮＩＣが算出される。
【０１４５】
【数２】

【０１４６】
続くステップ２７３では、ステップ２７１で算出された流入ＮＯ_Ｘ濃度ＣＮＩがＣＮＩＰとして記憶される。
【０１４７】
図１３を参照して上述したように、流入ＮＯ_Ｘ濃度ＣＮＩ及び流出ＮＯ_Ｘ濃度ＣＮＯはほぼ同時期に検出される。しかしながら、排気ガス部分がそれについての流入ＮＯ_Ｘ濃度ＣＮＩが算出されてからＮＯ_Ｘセンサ５１に到達するまでに遅れ時間ＤＬＹだけ必要であることを考えると、或る時点で検出された流出ＮＯ_Ｘ濃度ＣＮＯと、この或る時点から遅れ時間ＤＬＹだけ前に検出された流入ＮＯ_Ｘ濃度ＣＮＩとに基づいてＮＯ_Ｘ浄化率ＥＦＦを算出すべきである。
【０１４８】
そこで本発明による第１実施例では、この遅れ時間ＤＬＹを求め（ステップ２７１）、現時点から遅れ時間ＤＬＹだけ前の流入ＮＯ_Ｘ濃度ＣＮＩを算出するようにしている。この算出結果が本発明による第１実施例における流入ＮＯ_Ｘ濃度補正値ＣＮＩＣである。
【０１４９】
遅れ時間ＤＬＹは機関運転状態例えば吸入空気量Ｇａの関数として予め求められており、図２４に示されるマップの形でＲＯＭ４２内に予め記憶されている。ステップ２７１では図２４のマップから遅れ時間ＤＬＹが算出される。
【０１５０】
一方、図２５に示されるように、前回即ち時間ｔ（ｊ−１）に検出された流入ＮＯ_Ｘ濃度をＣＮＩ（ｊ−１）、今回即ち時間ｔ（ｊ）に検出された流入ＮＯ_Ｘ濃度をＣＮＩ（ｊ）とすると、時間ｔと流入ＮＯ_Ｘ濃度ＣＮＩとの関係は次式（２）で表される。
【０１５１】
【数３】

【０１５２】
なお、ｊは流入ＮＯ_Ｘ濃度ＣＮＩの検出回数を表している（ｊ＝１，２，…）。この式（２）において、時間ｔにｔ−ＤＬＹを代入し、ＣＮＩ（ｊ−１）を前回検出された流入ＮＯ_Ｘ濃度ＣＮＩＰに書き換えると、上述した式（１）が得られる。
【０１５３】
なお、今回検出された流入ＮＯ_Ｘ濃度ＣＮＩと、遅れ時間ＤＬＹだけ先の流出ＮＯ_Ｘ濃度ＣＮＯとに基づいてＮＯ_Ｘ浄化率を算出するようにしてもよい。即ち、この場合には、流出ＮＯ_Ｘ濃度ＣＮＯが遅れ時間に基づいて補正されるということになる。
【０１５４】
次に、ステップ２３７（図２０）で実行される流出ＮＯ_Ｘ濃度補正値ＣＮＯＣの算出ルーチンを図２６を参照して説明する。図２６を参照すると、まずステップ２８０では、ＮＯ_Ｘセンサ５１により検出された流出ＮＯ_Ｘ濃度ＣＮＯが読み込まれる。続くステップ２８１では次式に基づいて流出ＮＯ_Ｘ濃度補正値ＣＮＯＣが算出される。
【０１５５】
【数４】

【０１５６】
ここで、ｔＮ，ｋｔＮは上述したＮＯ_Ｘ処理の時間間隔及び補正量である。
【０１５７】
図２から図５までを参照して説明したように、本発明による実施例では、ＮＯ_Ｘ吸収剤２３内に流入する排気ガスの空燃比ＡＦＮがリーンに保持されているときの流出ＮＯ_Ｘ濃度ＣＮＯが許容最大濃度を越えないように、かつＮＯ_Ｘ処理による燃料消費率が許容最大値を越えないように、補正量ｋｔＮでもって基本時間間隔ｔＮＢが補正される（ｔＮ＝ｔＮＢ＋ｋｔＮ）。ここで、補正量ｋｔＮが機関運転状態に応じて変動することを考えると、（ｔＮＢ＋ｋｔＮ）／ｔＮＢ（＝ｔＮ／ｔＮＢ）は機関運転状態の定常運転状態に対するずれ量を表しており、検出された流出ＮＯ_Ｘ濃度ＣＮＯをこの（ｔＮＢ＋ｋｔＮ）／ｔＮＢで割り算した結果は基準運転状態即ち定常運転時における流出ＮＯ_Ｘ濃度に換算した値を表している。
【０１５８】
そこで本発明による第１実施例では、ステップ２８０で検出された流出ＮＯ_Ｘ濃度ＣＮＯを（ｔＮＢ＋ｋｔＮ）／ｔＮＢで割り算することにより、定常運転時における流出ＮＯ_Ｘ濃度に換算した値を算出するようにしている。この算出結果が本発明による第１実施例における流出ＮＯ_Ｘ濃度補正値ＣＮＯＣである。このようにすると、流出ＮＯ_Ｘ濃度が機関運転状態の影響を受けず、従ってＮＯ_Ｘ浄化率又は一次式ｍが機関運転状態の影響を受けない。
【０１５９】
従って、一般的にいうと、機関定常運転時のためのＮＯ_Ｘ処理の基本時間間隔ｔＮＢを設定し、機関運転状態に基づいて基本時間間隔ｔＮＢを補正量ｋｔＮでもって補正し、補正量ｋｔＮに基づいて流出ＮＯ_Ｘ量ＣＮＯを補正しているということになる。
【０１６０】
次に、ステップ２４５（図１８）で実行されるＮＯ_Ｘ浄化率補正値ＥＦＦＣの算出ルーチンを図２７を参照して説明する。図２７を参照すると、まずステップ２９０では、ＮＯ_Ｘ浄化率ＥＦＦが次式に基づいて算出される。
【０１６１】
ＥＦＦ＝（ＣＮＩＡＶ−ＣＮＯＡＶ）／ＣＮＩＡＶ
続くステップ２９１では図２８（Ａ）のマップから第１の補正係数ｋＥ１が算出され、続くステップ２９２では図２８（Ｂ）のマップから第２の補正係数ｋＥ２が算出され、続くステップ２９３では図２８（Ｃ）のマップから第３の補正係数ｋＥ３が算出される。続くステップ２９４では次式に基づいてＮＯ_Ｘ浄化率補正値ＥＦＦＣが算出される。
【０１６２】
ＥＦＦＣ＝ＥＦＦ・ｋＥ１・ｋＥ２・ｋＥ３
第１及び第２の補正係数ｋＥ１，ｋＥ２はステップ２９０で算出されたＮＯ_Ｘ浄化率ＥＦＦを、ＮＯ_Ｘ吸収剤２３の状態が予め定められた基準状態のときのＮＯ_Ｘ浄化率に換算するためのものである。この基準状態はどのように定めてもよいが、本発明による第１実施例では、ＮＯ_Ｘ吸収剤２３の温度ＴＣが基準温度ＴＣＲであり、ＮＯ_Ｘ吸収剤２３における排気ガスの空間速度を表す吸入空気量Ｇａが基準量ＧａＲであるときを基準状態としている。
【０１６３】
第１の補正係数ｋＥ１は図２８（Ａ）に示されるように、ＮＯ_Ｘ吸収剤２３の温度ＴＣが低いときにはＮＯ_Ｘ吸収剤２３の温度ＴＣが高くなるにつれて小さくなり、ＮＯ_Ｘ吸収剤２３の温度ＴＣが高いときにはＮＯ_Ｘ吸収剤２３の温度ＴＣが高くなるにつれて大きくなり、ＮＯ_Ｘ吸収剤２３の温度ＴＣが基準温度ＴＣＲのときに１．０となる。
【０１６４】
また、第２の補正係数ｋＥ２は図２８（Ｂ）に示されるように、ＮＯ_Ｘ吸収剤２３における排気ガスの空間速度を表す吸入空気量Ｇａが多くなるにつれて大きくなり、吸入空気量Ｇａが基準量ＧａＲのときに１．０となる。
【０１６５】
このようにすると、ＮＯ_Ｘ浄化率がＮＯ_Ｘ吸収剤２３の状態の影響を受けず、従って一次式ｍがＮＯ_Ｘ吸収剤２３の状態の影響を受けない。
【０１６６】
一方、第３の補正係数ｋＥ３はＮＯ_Ｘ浄化率に対するＮＯ_Ｘ吸収剤２３の熱劣化度合いの影響を除去するためのものである。即ち、第３の補正係数ｋＥ３はＮＯ_Ｘ吸収剤２３の熱劣化度合いを表す熱劣化カウンタｎＴＣＨがゼロのときに１．０であり、熱劣化カウンタｎＴＣＨが大きくなるにつれて大きくなる。
【０１６７】
これら補正係数ｋＥ１，ｋＥ２，ｋＥ３はそれぞれ予め実験により求められており、図２８（Ａ），（Ｂ），（Ｃ）に示されるマップの形で予めＲＯＭ４２内に記憶されている。
【０１６８】
熱劣化カウンタｎＴＣＨは例えば図２９に示される熱劣化カウンタｎＴＣＨの算出ルーチンにより算出することができる。このルーチンは予め定められた設定時間毎の割り込みによって実行される。
【０１６９】
図２９を参照すると、まずステップ３００では、ＮＯ_Ｘ吸収剤２３の温度ＴＣが予め定められた設定温度ＴＣＨよりも高いか否かが判別される。ＴＣ≦ＴＣＨのときには処理サイクルを終了し、ＴＣ＞ＴＣＨのときには次いでステップ３０１に進み、熱劣化カウンタｎＴＣＨが１だけインクリメントされる（ｎＴＣＨ＝ｎＴＣＨ＋１）。
【０１７０】
次に、ステップ２２６（図１７）で実行される給油判定ルーチンを図３０を参照して説明する。図３０を参照すると、まずステップ３１０では、給油フラグＸＦＣがリセットされているか否かが判別される。この給油フラグは給油、より正確にはイオウ濃度の異なる燃料の給油が行われたと判定されたときにセットされ（ＸＦＣ＝１）、それ以外はリセットされる（ＸＦＣ＝０）ものである。給油フラグＸＦＣがリセットされているときには次いでステップ３１１に進み、ＮＯ_Ｘ浄化率平均値ＥＦＦＡＶの一次式ｍに対する偏差Ｒが算出される。この偏差Ｒは例えば相関係数で表すことができる。
【０１７１】
続くステップ３１２では、今回の処理サイクルにおける偏差Ｒと前回の処理サイクルにおける偏差ＲＰとの差が予め定められたしきい値Ｒ１よりも大きいか否かが判別される。Ｒ−ＲＰ≦Ｒ１のときには次いでステップ３１３に進み、今回の処理サイクルにおける偏差ＲがＲＰとして記憶される。これに対し、Ｒ−ＲＰ＞Ｒ１のときには次いでステップ３１４に進み、給油フラグＸＦＣがセットされる（ＸＦＣ＝１）。即ち、Ｒ−ＲＰ＞Ｒ１のときには給油が行われたと判断される。
【０１７２】
図８を参照して上述したように、給油が行われない限り、燃料消費量ＱＦとＮＯ_Ｘ浄化率平均値ＥＦＦＡＶとの関係を表す一次式ｍは変わらない。このことは、給油前と給油後とで、燃料消費量ＱＦとＮＯ_Ｘ浄化率平均値ＥＦＦＡＶとの関係を表す一次式ｍが異なることを意味している。即ち、図３１に示されるように、給油前には燃料消費量ＱＦとＮＯ_Ｘ浄化率平均値ＥＦＦＡＶとの関係を表す一次式はｍ１であり、給油後にはｍ２になる。
【０１７３】
このように、本来、二つの一次式ｍ１，ｍ２で表されるべき燃料消費量ＱＦとＮＯ_Ｘ浄化率平均値ＥＦＦＡＶとの関係を一つの一次式ｍで表そうとすると、偏差Ｒが大きくなる。そこで本発明による第１実施例では、偏差Ｒがしきい値Ｒ１よりも大きくなったときに給油が行われたと判断しているのである。
【０１７４】
続くステップ３１５では、係数ｋｍｉ及び平均値算出回数カウンタｎＡＶがクリアされる。即ち、給油が行われたと判断されたときには、給油前のＮＯ_Ｘ浄化率ＥＦＦ又は平均値ＥＦＦＡＶに基づく一次式ｍの決定が禁止される。
【０１７５】
続くステップ３１６では、高イオウ濃度フラグＸＦＳＨがセットされているか否かが判別される。この高イオウ濃度フラグＸＦＳＨは燃料が高イオウ濃度燃料であると判断されたときにセットされ（ＸＦＳＨ＝１）、低イオウ濃度燃料又は標準燃料であると判断されたときにリセットされる（ＸＦＳＨ＝０）ものであり、図３２から図３４までの処理開始時量ＱＳＳの設定ルーチンにおいてセット又はリセットされる。
【０１７６】
高イオウ濃度フラグＸＦＳＨがセットされているときには次いでステップ３１７に進み、低イオウ濃度判定フラグＸＦＣＬがセットされる（ＸＦＣＬ＝１）。この低イオウ濃度判定フラグＸＦＣＬは燃料が低イオウ濃度燃料又は標準燃料に切り替えられたと判断されたときにセットされ（ＸＦＣＬ＝１）、それ以外はリセットされる（ＸＦＣＬ＝０）ものである。即ち、燃料が高イオウ濃度燃料である（ＸＦＳＨ＝１）ときに給油が行われたということは、燃料が低イオウ濃度燃料又は標準燃料に切り替えられたということであり、このとき低イオウ濃度判定フラグＸＦＣＬがセットされる。
【０１７７】
これに対し、高イオウ濃度フラグＸＦＳＨがリセットされているときには次いでステップ３１８に進み、高イオウ濃度判定フラグＸＦＣＨがセットされる（ＸＦＣＨ＝１）。この高イオウ濃度判定フラグＸＦＣＨは燃料が高イオウ濃度燃料に切り替えられたと判断されたときにセットされ（ＸＦＣＨ＝１）、それ以外はリセットされる（ＸＦＣＨ＝０）ものである。即ち、燃料が低イオウ濃度燃料又は標準燃料である（ＸＦＳＨ＝０）ときに給油が行われたということは、燃料が高イオウ濃度燃料に切り替えられたということであり、このとき高イオウ濃度判定フラグＸＦＣＨがセットされる。
【０１７８】
一方、ステップ３１０において、給油フラグＸＦＣがセットされると（ＸＦＣ＝１）、処理サイクルが完了される。後述するように、この給油フラグＸＦＣは図３２から図３４までに示される処理開始時量ＱＳＳの設定ルーチンにおいてリセットされる。従って、給油フラグＸＦＣが一旦セットされると、処理開始時量ＱＳＳの設定又はＳＯ_Ｘ処理が行われるまで給油判定が行われず、処理開始時量ＱＳＳの設定又はＳＯ_Ｘ処理が行われると、再び給油判定が行われることになる。
【０１７９】
次に、ステップ２０４（図１５）で実行される処理開始時量ＱＳＳの設定ルーチンを図３２から図３４までを参照して説明する。図３２から図３４までを参照すると、まずステップ３２０では、蓄積可能最大量ＱＳＭが算出される（ＱＳＭ＝ΔＱＦ１・ＣＳＭ）。続くステップ３２１では、高イオウ濃度判定フラグＸＦＣＨがリセットされているか否かが判別される。高イオウ濃度判定フラグＸＦＣＨがリセットされているときには次いでステップ３２２に進み、低イオウ濃度判定フラグＸＦＣＬがリセットされているか否かが判別される。低イオウ濃度判定フラグＸＦＣＬがリセットされているときには次いでステップ３２３に進む。高イオウ濃度判定フラグＸＦＣＨ又は低イオウ濃度判定フラグＸＦＣＬがセットされるのは給油が行われたと判定されたときであるので（図３０参照）、給油が行われたと判定されていないときにステップ３２３に進むことになる。
【０１８０】
ステップ３２３では、ＮＯ_Ｘ浄化率平均値ＥＦＦＡＶの算出回数を表す平均値算出回数カウンタｎＡＶ（図１７参照）が予め定められた下限値ｎＬ１以上か否かが判別される。ｎＡＶ≧ｎＬ１のときには次いでステップ３２４に進み、ＳＯ_Ｘ処理を開始すべきときのＮＯ_Ｘ浄化率推定値ＥＦＦＥが次式（３）に基づいて算出される。
【０１８１】
【数５】

【０１８２】
続くステップ３２５では、ＮＯ_Ｘ浄化率推定値ＥＦＦＥが境界浄化率ＥＦＦＢよりも高いか否かが判別される。ＥＦＦＥ＞ＥＦＦＢのときには次いでステップ３２６に進み、図９のマップから蓄積ＳＯ_Ｘ量推定値ＱＳＥが算出される。続くステップ３２７では、処理開始時量ＱＳＳがこの蓄積ＳＯ_Ｘ量推定値ＱＳＥに設定される。続くステップ３２８では、高イオウ濃度フラグＸＦＳＨがセットされる（ＸＦＳＨ＝１）。即ち、この場合には燃料が高イオウ濃度燃料であると判断される。次いでステップ３３５に進む。
【０１８３】
これに対し、ＥＦＦＥ≦ＥＦＦＢのときにはステップ３２５からステップ３２９に進み、基準濃度ＣＳＲを用いて処理開始時量ＱＳＳが設定される（ＱＳＳ＝ΔＱＦ１・ＣＳＲ）。即ち、処理開始時量ＱＳＳがＮＯ_Ｘ浄化率平均値ＥＦＦＡＶ又は蓄積ＳＯ_Ｘ量推定値ＱＳＥとは無関係な量に設定される。続くステップ３３０では、高イオウ濃度フラグＸＦＳＨがリセットされる（ＸＦＳＨ＝０）。即ち、この場合には燃料が低イオウ濃度燃料又は標準燃料であると判断される。次いでステップ３３５に進む。
【０１８４】
一方、ｎＡＶ＜ｎＬ１のときにはステップ３２３からステップ３３１に進み、流出ＮＯ_Ｘ濃度補正値ＣＮＯＣが許容最小値ＣＮＯＬよりも小さいときの回数を表す少量回数カウンタｎＮＬ（図２０参照）が予め定められた上限値ｎＵよりも小さいか否かが判別される。ｎＮＬ＜ｎＵのときには次いでステップ３３２に進み、処理開始時量ＱＳＳが蓄積可能最大量ＱＳＭに設定される。次いでステップ３２８に進み、高イオウ濃度フラグＸＦＳＨがセットされる。
【０１８５】
これに対し、ｎＮＬ≧ｎＵのときには次いでステップ３２９に進んで処理開始時量ＱＳＳが設定され（ＱＳＳ＝ΔＱＦ１・ＣＳＲ），続くステップ３３０では高イオウ濃度フラグＸＦＳＨがリセットされる（ＸＦＳＨ＝０）。
【０１８６】
このように、平均値算出回数カウンタｎＡＶにより表されるＮＯ_Ｘ浄化率平均値ＥＦＦＡＶの算出回数が多いときには、ＮＯ_Ｘ浄化率推定値ＥＦＦＥ及び蓄積ＳＯ_Ｘ量推定値ＱＳＥに基づいて処理開始時量ＱＳＳが設定され、ＮＯ_Ｘ浄化率平均値ＥＦＦＡＶの算出回数が少ないときには、燃料中のイオウ濃度をＣＳＲ又はＣＳＭと仮定して処理開始時量ＱＳＳが設定される。
【０１８７】
従って、一般的に言うと、検出された流出ＮＯ_Ｘ量ＣＮＯ又は補正値ＣＮＯｃが予め定められた許容最小量ＣＮＯＬよりも少ないときの回数ｎＮＬが上限値ｎＵを越えたときには、処理開始時量ＱＳＳを算出された蓄積ＳＯ_Ｘ量ＱＳＥとは無関係な量に設定しているということになる。
【０１８８】
その上で、少量回数カウンタｎＮＬが小さいときには高イオウ濃度燃料であると判断され、少量回数カウンタｎＮＬが大きいときには燃料が低イオウ濃度燃料又は標準燃料であると判断される。
【０１８９】
一方、高イオウ濃度判定フラグＸＦＣＨがセットされているとき、即ち高イオウ濃度燃料が給油されたと判断されたときにはステップ３２１からステップ３３３に進み、平均値算出回数カウンタｎＡＶが下限値ｎＬ１以上か否かが判別される。ｎＡＶ≧ｎＬ１のときにはステップ３３４に進み、ＳＯ_Ｘ処理を開始すべきときのＮＯ_Ｘ浄化率推定値ＥＦＦＥが上述の式（３）に基づいて算出される。次いでステップ３２６，３２７，３２８に順次進む。これに対し、ｎＡＶ＜ｎＬ１のときには、ステップ３３２，３２８に順次進む。
【０１９０】
一方、低イオウ濃度判定フラグＸＦＣＬがセットされているとき、即ち低イオウ濃度燃料又は標準燃料が給油されたと判断されたときにはステップ３２２からステップ３２９，３３０に順次進む。
【０１９１】
ステップ３３５では、設定された処理開始時量ＱＳＳが蓄積可能最大量ＱＳＭよりも多いか否かが判別される。ＱＳＭ＞ＱＳＭのときには次いでステップ３３６に進み、処理開始時量ＱＳＳを蓄積可能最大量ＱＳＭに設定した後にステップ３３７に進む。これに対し、ＱＳＳ≦ＱＳＭのときにはステップ３３７にジャンプする。
【０１９２】
ステップ３３７では、係数ｋｍｉ（ｉ＝１，２，３，４）、平均値算出回数カウンタｎＡＶ、及び少量回数カウンタｎＮＬがクリアされ、給油フラグＸＦＣ、高イオウ濃度判定フラグＸＦＣＨ、及び低イオウ濃度判定フラグＸＦＣＬがリセットされる。
【０１９３】
このように本発明による第１実施例では、図３５に示されるように、燃料消費量ＱＦが第２の設定量ｑｆ２だけ増大する毎にＮＯ_Ｘ浄化率平均値ＥＦＦＡＶが算出される。なお、浄化率算出回数カウンタｎＥＦＦで表されるＮＯ_Ｘ浄化率ＥＦＦの算出回数が少ないときには、図３５において矢印Ｗで示されるようにＮＯ_Ｘ浄化率平均値ＥＦＦＡＶは算出されない。その上で、ＮＯ_Ｘ浄化率平均値ＥＦＦＡＶと燃料消費量ＱＦとの関係を表す一次式ｍがこれらＮＯ_Ｘ浄化率平均値ＥＦＦＡＶに基づき決定される。次いで、先のＳＯ_Ｘ処理が完了してから（図３５の矢印Ｘ１参照）、燃料消費量ＱＦが第１の設定量ｑｆ１だけ増大すると、このときのＮＯ_Ｘ浄化率推定値ＥＦＦＥが一次式ｍを用いて算出される。
【０１９４】
次に、本発明による第２実施例を説明する。
【０１９５】
図３６に示されるように、ＮＯ_Ｘ処理が行われると、ＮＯ_Ｘセンサ５１により検出される流出ＮＯ_Ｘ濃度ＣＮＯはまず、急激に減少して下向きのピークＰＫｍ１に達し、次いで増大して今度は上向きのピークＰＫＭに達し、次いで再び減少して下向きのピークＰＫｍ２に達した後に、徐々に増大する。
【０１９６】
流出ＮＯ_Ｘ濃度ＣＮＯがなぜこのような挙動をとるかは必ずしも明らかではないが、次の理由によるものと考えられている。即ち、ＮＯ_Ｘ吸収剤２３内に流入する排気ガスの空燃比ＡＦＮがリッチに切り替えられると、上述したように、ＮＯ_Ｘ吸収剤２３内に蓄えられているＮＯ_ＸがＮＯ_Ｘ吸収剤２３から例えばＮＯ_２の形で放出される。ＮＯ_Ｘ処理が開始された直後はこの放出されたＮＯ_２が直ちに還元され、斯くして下向きのピークＰＫｍ１が発生する。ところが、時間の経過と共にＮＯ_Ｘの還元反応が生じにくくなり、ＮＯ_Ｘ吸収剤２３から放出されるけれども還元されないＮＯ_Ｘの量が増大し、斯くして上向きのピークＰＫＭが発生する。更に時間が経過すると、ＮＯ_Ｘ吸収剤２３から放出されるＮＯ_Ｘの量が次第に減少し、しかしながらＮＯ_Ｘ吸収剤２３内に流入するＮＯ_ＸのうちＮＯ_Ｘ吸収剤２３内に吸収されることなくＮＯ_Ｘ吸収剤２３を通過するＮＯ_Ｘの量が少しずつ増大し、斯くして下向きのピークＰＫｍ２が発生する。
【０１９７】
上向きのピークＰＫＭにおける流出ＮＯ_Ｘ濃度ＣＮＯを極大値ＣＰＫと称すると、この極大値ＣＰＫはＮＯ_Ｘ処理によりＮＯ_Ｘ吸収剤２３から放出されるＮＯ_Ｘの量に依存し、従ってＮＯ_Ｘ処理が開始されるときのＮＯ_Ｘ吸収剤２３内の蓄積ＮＯ_Ｘ量に依存する。即ち、蓄積ＮＯ_Ｘ量が多いときには極大値ＣＰＫは大きくなり、蓄積ＮＯ_Ｘ量が少ないときには極大値ＣＰＫは小さくなる。一方、ＮＯ_Ｘ吸収剤２３内の蓄積ＳＯ_Ｘ量が多くなるにつれて、ＮＯ_Ｘ吸収剤２３内に蓄えられ得るＮＯ_Ｘの量が少なくなる。
【０１９８】
従って、蓄積ＳＯ_Ｘ量が多くなるにつれて極大値ＣＰＫが小さくなり、燃料消費量ＱＦが多くなるにつれて極大値ＣＰＫが小さくなることがわかる。
【０１９９】
そこで本発明による第２実施例では、極大値ＣＰＫと燃料消費量ＱＦとの関係を表す関係式を求め、ＳＯ_Ｘ処理を開始すべきと判断されたときの極大値ＣＰＫの推定値ＣＰＫＥをこの関係式を用いて算出するようにしている。
【０２００】
具体的には、ＮＯ_Ｘ処理が行われる毎に極大値ＣＰＫが算出される。その上で、図３７に示されるように、平均値算出時期が到来する毎に、即ち燃料消費量ＱＦが第２の設定量ｑｆ２だけ増大する毎に極大値平均値ＣＰＫＡＶが算出され、極大値平均値ＣＰＫＡＶと燃料消費量ＱＦとの関係を表す一次式ｍがこれら極大値平均値ＣＰＫＡＶに基づき決定される。次いで、先のＳＯ_Ｘ処理が完了してから（図３７の矢印Ｘ１参照）、燃料消費量ＱＦが第１の設定量ｑｆ１だけ増大すると、このときの極大値推定値ＣＰＫＥが一次式ｍを用いて算出される。
【０２０１】
更に、極大値ＣＰＫと蓄積ＳＯ_Ｘ量推定値ＱＳＥとの関係は予め実験により求められており、図３８に示されるマップの形で予めＲＯＭ４２内に記憶されている。従って、ＳＯ_Ｘ処理を行うべきと判断されたときの極大値推定値ＣＰＫＥが算出されれば、このときの蓄積ＳＯ_Ｘ量推定値ＱＳＥが図３８のマップを用いて求められる。
【０２０２】
従って、一般的に表現すると、ＮＯ_Ｘ処理が行われたときに生ずる流出ＮＯ_Ｘ量の極大値を検出し、検出された流出ＮＯ_Ｘ量の極大値に基づいてＮＯ_Ｘ吸収剤の蓄積ＳＯ_Ｘ量を求めているということになる。
【０２０３】
図３９は本発明による第２実施例の係数ｋｍｉの算出ルーチンを示している。本発明による第２実施例でも図１５に示されるＳＯ_Ｘ制御ルーチンが実行され、この係数ｋｍｉの算出ルーチンはＳＯ_Ｘ制御ルーチンのステップ２０３で実行される。
【０２０４】
図３９を参照すると、ステップ４００では現在の燃料消費量ＱＦと、先の平均値算出時期における燃料消費量ＱＦＰ２との差が第２の設定量ｑｆ２以上か否かが判別される。ＱＦ−ＱＦＰ２＜ｑｆ２のとき、即ち先の極大値平均値ＣＰＫＡＶが算出されてから燃料が第２の設定量ｑｆ２だけ消費されていないときには次いでステップ４０１に進み、ＮＯ_Ｘ処理が行われているか否かが判別される。ＮＯ_Ｘ処理が行われていないときには処理サイクルを終了し、ＮＯ_Ｘ処理が行われているときには次いでステップ４０２に進み、流出ＮＯ_Ｘ濃度ＣＮＯの極大値ＣＰＫが算出される。なお、極大値ＣＰＫを機関運転状態又はＮＯ_Ｘ触媒２３の状態に基づいて補正することもできる。続くステップ４０３では、極大値ＣＰＫの積算値ＳＣＰＫが算出される（ＳＣＰＫ＝ＳＣＰＫ＋ＣＰＫ）。続くステップ４０４では、極大値ＣＰＫの検出回数を表す検出回数カウンタｎＣＰＫが１だけインクリメントされる（ｎＣＰＫ＝ｎＣＰＫ＋１）。
【０２０５】
ＱＦ−ＱＦＰ２≧ｑｆ２になるとステップ４００からステップ４０５に進み、極大値平均値ＣＰＫＡＶが算出される（ＣＰＫＡＶ＝ＳＣＰＫ／ｎＣＰＫ）。続くステップ４０６では、極大値平均値ＣＰＫＡＶの算出回数を表す平均値算出回数カウンタｎＡＶが１だけインクリメントされる（ｎＡＶ＝ｎＡＶ＋１）。続くステップ４０７では、係数ｋｍｉ（ｉ＝１，２，３，４）が次式に基づいてそれぞれ算出される。
【０２０６】
ｋｍ１＝ｋｍ１＋ＱＦ
ｋｍ２＝ｋｍ２＋ＱＦ^２
ｋｍ３＝ｋｍ３＋ＱＦ・ＣＰＫ
ｋｍ４＝ｋｍ４＋ＣＰＫ
続くステップ４０８では、現在の燃料消費量ＱＦがＱＦＰ２として記憶され、積算値ＳＣＰＫ及びカウンタｎＣＰＫがそれぞれクリアされる。
【０２０７】
図４０は本発明による第２実施例の処理開始時量ＱＳＳの設定ルーチンを示している。このルーチンはＳＯ_Ｘ制御ルーチンのステップ２０４で実行される。
【０２０８】
図４０を参照すると、ステップ４１０では極大値推定値ＣＰＫＥが次式に基づいて算出される。
【０２０９】
【数６】

【０２１０】
続くステップ４１１では、図３８のマップから蓄積ＳＯ_Ｘ量推定値ＱＳＥが算出される。続くステップ４１２では、処理開始時量ＱＳＳがこの蓄積ＳＯ_Ｘ量推定値ＱＳＥに設定される。続くステップ４１３では、係数ｋｍｉ（ｉ＝１，２，３，４）及び平均値算出回数カウンタｎＡＶがクリアされる。
【０２１１】
本発明による第２実施例のその他の構成及び作用は上述の実施例と同様であるので説明を省略する。
【０２１２】
次に、本発明による第３実施例を説明する。
【０２１３】
上述したように、ＮＯ_Ｘ処理が行われると、ＮＯ_Ｘ吸収剤２３内に蓄えられているＮＯ_Ｘが還元されることなくＮＯ_Ｘ吸収剤２３から流出する。このときＮＯ_Ｘ吸収剤２３から流出するＮＯ_Ｘの積算量ＳＱＬは図４１に示されるように、流出ＮＯ_Ｘ濃度ＣＮＯに下向きのピークＰＫｍ１が発生してから、下向きのピークＰＫｍ２が発生するまでの間の、流出ＮＯ_Ｘ濃度ＣＮＯの積算値の形で求めることができる。
【０２１４】
この流出ＮＯ_Ｘ積算量ＳＱＬは上述の極大値ＣＰＫと同様に、ＮＯ_Ｘ処理が開始されるときのＮＯ_Ｘ吸収剤２３内の蓄積ＮＯ_Ｘ量に依存する。即ち、蓄積ＳＯ_Ｘ量が多くなるにつれて流出ＮＯ_Ｘ積算量ＳＱＬが少なくなり、燃料消費量ＱＦが多くなるにつれて流出ＮＯ_Ｘ積算量ＳＱＬが少なくなる。
【０２１５】
そこで本発明による第３実施例では、流出ＮＯ_Ｘ積算量ＳＱＬと燃料消費量ＱＦとの関係を表す関係式を求め、ＳＯ_Ｘ処理を開始すべきと判断されたときの流出ＮＯ_Ｘ積算量ＳＱＬの推定値ＳＱＬＥをこの関係式を用いて算出するようにしている。
【０２１６】
具体的には、ＮＯ_Ｘ処理が行われる毎に流出ＮＯ_Ｘ積算量ＳＱＬが算出される。その上で、図４２に示されるように、平均値算出時期が到来する毎に、即ち燃料消費量ＱＦが第２の設定量ｑｆ２だけ増大する毎に流出ＮＯ_Ｘ積算量平均値ＳＱＬＡＶが算出され、流出ＮＯ_Ｘ積算量平均値ＳＱＬＡＶと燃料消費量ＱＦとの関係を表す一次式ｍがこれら流出ＮＯ_Ｘ積算量平均値ＳＱＬＡＶに基づき決定される。次いで、先のＳＯ_Ｘ処理が完了してから（図４２の矢印Ｘ１参照）、燃料消費量ＱＦが第１の設定量ｑｆ１だけ増大すると、このときの流出ＮＯ_Ｘ積算量推定値ＳＱＬＥが一次式ｍを用いて算出される。
【０２１７】
更に、流出ＮＯ_Ｘ積算量ＳＱＬと蓄積ＳＯ_Ｘ量推定値ＱＳＥとの関係は予め実験により求められており、図４３に示されるマップの形で予めＲＯＭ４２内に記憶されている。従って、ＳＯ_Ｘ処理を行うべきと判断されたときの流出ＮＯ_Ｘ積算量推定値ＳＱＬＥが算出されれば、このときの蓄積ＳＯ_Ｘ量推定値ＱＳＥが図４３のマップを用いて求められる。
【０２１８】
従って、一般的に表現すると、ＮＯ_Ｘ処理が行われたときの流出ＮＯ_Ｘ量積算値を検出し、検出された流出ＮＯ_Ｘ量積算値に基づいてＮＯ_Ｘ吸収剤の蓄積ＳＯ_Ｘ量を求めているということになる。
【０２１９】
図４４は本発明による第３実施例の係数ｋｍｉの算出ルーチンを示している。本発明による第３実施例でも図１５に示されるＳＯ_Ｘ制御ルーチンが実行され、この係数ｋｍｉの算出ルーチンはＳＯ_Ｘ制御ルーチンのステップ２０３で実行される。
【０２２０】
図４４を参照すると、ステップ４２０では現在の燃料消費量ＱＦと、先の平均値算出時期における燃料消費量ＱＦＰ２との差が第２の設定量ｑｆ２以上か否かが判別される。ＱＦ−ＱＦＰ２＜ｑｆ２のとき、即ち先の極大値平均値ＣＰＫＡＶが算出されてから燃料が第２の設定量ｑｆ２だけ消費されていないときには次いでステップ４２１に進み、ＮＯ_Ｘ処理が行われているか否かが判別される。ＮＯ_Ｘ処理が行われていないときには処理サイクルを終了し、ＮＯ_Ｘ処理が行われているときには次いでステップ４２２に進み、流出ＮＯ_Ｘ量積算値ＳＱＬが算出される。なお、流出ＮＯ_Ｘ量積算値ＳＱＬを機関運転状態又はＮＯ_Ｘ触媒２３の状態に基づいて補正することもできる。続くステップ４２３では、流出ＮＯ_Ｘ量積算値ＳＱＬの積算値ＳＳＱＬが算出される（ＳＳＱＬ＝ＳＳＱＬ＋ＳＱＬ）。続くステップ４２４では、流出ＮＯ_Ｘ量積算値ＳＱＬの検出回数を表す検出回数カウンタｎＳＱＬが１だけインクリメントされる（ｎＳＱＬ＝ｎＳＱＬ＋１）。
【０２２１】
ＱＦ−ＱＦＰ２≧ｑｆ２になるとステップ４２０からステップ４２５に進み、流出ＮＯ_Ｘ量積算値平均値ＳＱＬＡＶが算出される（ＳＱＬＡＶ＝ＳＳＱＬ／ｎＳＱＬ）。続くステップ４２６では、流出ＮＯ_Ｘ量積算値平均値ＳＱＬＡＶの算出回数を表す平均値算出回数カウンタｎＡＶが１だけインクリメントされる（ｎＡＶ＝ｎＡＶ＋１）。続くステップ４２７では、係数ｋｍｉ（ｉ＝１，２，３，４）が次式に基づいてそれぞれ算出される。
【０２２２】
ｋｍ１＝ｋｍ１＋ＱＦ
ｋｍ２＝ｋｍ２＋ＱＦ^２
ｋｍ３＝ｋｍ３＋ＱＦ・ＳＱＬ
ｋｍ４＝ｋｍ４＋ＳＱＬ
続くステップ４２８では、現在の燃料消費量ＱＦがＱＦＰ２として記憶され、積算値ＳＳＱＬ及びカウンタｎＳＱＬがそれぞれクリアされる。
【０２２３】
図４５は本発明による第３実施例の処理開始時量ＱＳＳの設定ルーチンを示している。このルーチンはＳＯ_Ｘ制御ルーチンのステップ２０４で実行される。
【０２２４】
図４５を参照すると、ステップ４３０では流出ＮＯ_Ｘ積算量推定値ＳＱＬＥが次式に基づいて算出される。
【０２２５】
【数７】

【０２２６】
続くステップ４３１では、図４３のマップから蓄積ＳＯ_Ｘ量推定値ＱＳＥが算出される。続くステップ４３２では、処理開始時量ＱＳＳがこの蓄積ＳＯ_Ｘ量推定値ＱＳＥに設定される。続くステップ４３３では、係数ｋｍｉ（ｉ＝１，２，３，４）及び平均値算出回数カウンタｎＡＶがクリアされる。
【０２２７】
本発明による第３実施例のその他の構成及び作用は上述の実施例と同様であるので説明を省略する。
【０２２８】
次に、本発明による第４実施例を説明する。
【０２２９】
図２から図５までを参照して上述したように、本発明による実施例では、ＮＯ_Ｘ吸収剤２３内に流入する排気ガスの空燃比ＡＦＮがリーンに保持されているときの流出ＮＯ_Ｘ濃度ＣＮＯが許容最大濃度を越えないように、かつＮＯ_Ｘ処理による燃料消費率が許容最大値を越えないように、補正量ｋｔＮでもって基本時間間隔ｔＮＢを補正することにより、ＮＯ_Ｘ処理の時間間隔ｔＮが算出される（ｔＮ＝ｔＮＢ＋ｋｔＮ）。
【０２３０】
蓄積ＳＯ_Ｘ量が少ないときには、ＮＯ_Ｘ吸収剤２３内に流入したＮＯ_ＸのうちＮＯ_Ｘ吸収剤２３内に蓄えられるＮＯ_Ｘの割合が高いので、流出ＮＯ_Ｘ濃度ＣＮＯが許容最大濃度に達するまでの時間が比較的長くなっている。その結果、図４６（Ａ）に示されるようにＮＯ_Ｘ処理の時間間隔ｔＮは比較的長くなっている。これに対し、蓄積ＳＯ_Ｘ量が多くなると、ＮＯ_Ｘ吸収剤２３内に蓄えられるＮＯ_Ｘの割合が低下するので、流出ＮＯ_Ｘ濃度ＣＮＯが許容最大濃度に達するまでの時間が短くなり、図４６（Ｂ）に示されるように時間間隔ｔＮが短くなる。
【０２３１】
本発明による実施例では上述したように、蓄積ＳＯ_Ｘ量がほぼゼロのときに最適になるように基本時間間隔ｔＮＢが設定されており、蓄積ＳＯ_Ｘ量が多くなるにつれて時間間隔ｔＮが短くなる。従って、蓄積ＳＯ_Ｘ量が多くなるにつれて時間間隔比ｔＮＢ／ｔＮが大きくなり、燃料消費量ＱＦが多くなるにつれて時間間隔比ｔＮＢ／ｔＮが大きくなる。
【０２３２】
ところが上述したように、時間間隔ｔＮは機関運転状態によっても変動し得る。しかしながら、定常運転時における時間間隔ｔＮ又は時間間隔比ｔＮＢ／ｔＮは機関運転状態の影響を受けず、機関定常運転時における実際の時間間隔ｔＮの基本時間間隔ｔＮＢに対するずれは蓄積ＳＯ_Ｘ量によるものである。
【０２３３】
そこで本発明による第４実施例では、定常運転時にＮＯ_Ｘ処理が行われたときの時間間隔比、即ち定常時時間間隔比ＲＳＴ（＝ｔＮＢ／（ｔＮＢ＋ｋｔＮ））と、燃料消費量ＱＦとの関係を表す関係式を求め、ＳＯ_Ｘ処理を開始すべきと判断されたときの定常時時間間隔比ＲＳＴの推定値ＲＳＴＥをこの関係式を用いて算出するようにしている。
【０２３４】
具体的には、定常運転時にＮＯ_Ｘ処理が行われる毎に定常時時間間隔比ＲＳＴが算出される。その上で、図４７に示されるように、平均値算出時期が到来する毎に、即ち燃料消費量ＱＦが第２の設定量ｑｆ２だけ増大する毎に定常時時間間隔比平均値ＲＳＴＡＶが算出され、定常時時間間隔比平均値ＲＳＴＡＶと燃料消費量ＱＦとの関係を表す一次式ｍがこれら定常時時間間隔比平均値ＲＳＴＡＶに基づき決定される。次いで、先のＳＯ_Ｘ処理が完了してから（図４７の矢印Ｘ１参照）、燃料消費量ＱＦが第１の設定量ｑｆ１だけ増大すると、このときの定常時時間間隔比推定値ＲＳＴＥが一次式ｍを用いて算出される。
【０２３５】
更に、定常時時間間隔比ＲＳＴと蓄積ＳＯ_Ｘ量推定値ＱＳＥとの関係は予め実験により求められており、図４８に示されるマップの形で予めＲＯＭ４２内に記憶されている。従って、ＳＯ_Ｘ処理を行うべきと判断されたときの定常時時間間隔比推定値ＲＳＴＥが算出されれば、このときの蓄積ＳＯ_Ｘ量推定値ＱＳＥが図４８のマップを用いて求められる。
【０２３６】
従って、一般的に表現すると、ＮＯ_Ｘ吸収剤２３内に流入する排気ガスの空燃比がリーンに保持されているときの流出ＮＯ_Ｘ量が予め定められた許容最大量を越えないようにＮＯ_Ｘ処理が行われる時間間隔を設定し、制御された時間間隔に基づいてＮＯ_Ｘ吸収剤の蓄積イオウ量を求めているということになる。
【０２３７】
図４９は本発明による第４実施例の係数ｋｍｉの算出ルーチンを示している。本発明による第４実施例でも図１５に示されるＳＯ_Ｘ制御ルーチンが実行され、この係数ｋｍｉの算出ルーチンはＳＯ_Ｘ制御ルーチンのステップ２０３で実行される。
【０２３８】
図４９を参照すると、ステップ４４０では現在の燃料消費量ＱＦと、先の平均値算出時期における燃料消費量ＱＦＰ２との差が第２の設定量ｑｆ２以上か否かが判別される。ＱＦ−ＱＦＰ２＜ｑｆ２のとき、即ち先の定常時時間間隔比平均値ＲＳＴＡＶが算出されてから燃料が第２の設定量ｑｆ２だけ消費されていないときには次いでステップ４４１に進み、定常運転時でかつＮＯ_Ｘ処理が行われているか否かが判別される。定常運転時でないか又はＮＯ_Ｘ処理が行われていないときには処理サイクルを終了し、定常運転時でかつＮＯ_Ｘ処理が行われているときには次いでステップ４４２に進み、現在の機関運転状態が属する領域が決定される。続くステップ４４３では、定常時時間間隔比ＲＳＴが算出される。
【０２３９】
ＲＳＴ＝ｔＮＢ／（ｔＮＢ＋ｋｔＮ）
続くステップ４４４では、定常時時間間隔比ＲＳＴの積算値ＳＲＳＴが算出される（ＳＲＳＴ＝ＳＲＳＴ＋ＲＳＴ）。続くステップ４４５では、定常時時間間隔比ＲＳＴの検出回数を表す検出回数カウンタｎＲＳＴが１だけインクリメントされる（ｎＲＳＴ＝ｎＲＳＴ＋１）。
【０２４０】
ＱＦ−ＱＦＰ２≧ｑｆ２になるとステップ４４０からステップ４４６に進み、定常時時間間隔比平均値ＲＳＴＡＶが算出される（ＲＳＴＡＶ＝ＳＲＳＴ／ｎＲＳＴ）。続くステップ４４７では、定常時時間間隔比ＲＳＴＡＶの算出回数を表す平均値算出回数カウンタｎＡＶが１だけインクリメントされる（ｎＡＶ＝ｎＡＶ＋１）。続くステップ４４８では、係数ｋｍｉ（ｉ＝１，２，３，４）が次式に基づいてそれぞれ算出される。
【０２４１】
ｋｍ１＝ｋｍ１＋ＱＦ
ｋｍ２＝ｋｍ２＋ＱＦ^２
ｋｍ３＝ｋｍ３＋ＱＦ・ＲＳＴ
ｋｍ４＝ｋｍ４＋ＲＳＴ
続くステップ４４９では、現在の燃料消費量ＱＦがＱＦＰ２として記憶され、積算値ＳＲＳＴ及びカウンタｎＲＳＴがそれぞれクリアされる。
【０２４２】
図５０は本発明による第４実施例の処理開始時量ＱＳＳの設定ルーチンを示している。このルーチンはＳＯ_Ｘ制御ルーチンのステップ２０４で実行される。
【０２４３】
図５０を参照すると、ステップ４５０では定常時時間間隔比推定値ＲＳＴＥが次式に基づいて算出される。
【０２４４】
【数８】

【０２４５】
続くステップ４５１では、図４８のマップから蓄積ＳＯ_Ｘ量推定値ＱＳＥが算出される。続くステップ４５２では、処理開始時量ＱＳＳがこの蓄積ＳＯ_Ｘ量推定値ＱＳＥに設定される。続くステップ４５３では、係数ｋｍｉ（ｉ＝１，２，３，４）及び平均値算出回数カウンタｎＡＶがクリアされる。
【０２４６】
本発明による第４実施例のその他の構成及び作用は上述の実施例と同様であるので説明を省略する。
【０２４７】
次に、本発明による第５実施例を説明する。
【０２４８】
図４６を参照して説明したように、蓄積ＳＯ_Ｘ量に応じてＮＯ_Ｘ処理の時間間隔ｔＮが変動する。従って、単位期間当たりに還元剤供給弁３１から供給される還元剤の総量も蓄積ＳＯ_Ｘ量に応じて変動することになる。即ち、蓄積ＳＯ_Ｘ量が多くなるにつれて単位期間当たりの還元剤供給量ＱＲは多くなり、燃料消費量ＱＦが多くなるにつれて単位期間当たりの還元剤供給量ＱＲは多くなる。
【０２４９】
そこで本発明による第５実施例では、単位期間当たりの還元剤供給量ＱＲと燃料消費量ＱＦとの関係を表す関係式を求め、ＳＯ_Ｘ処理を開始すべきと判断されたときの単位期間当たりの還元剤供給量ＱＲの推定値ＱＲＥをこの関係式を用いて算出するようにしている。
【０２５０】
具体的には、図５１に示されるように、供給量算出時期が到来する毎に、即ち例えば燃料消費量ＱＦが第２の設定量ｑｆ２だけ増大する毎に還元剤供給量の積算値、即ち単位期間当たりの還元剤供給量ＱＲが算出され、単位期間当たりの還元剤供給量ＱＲと燃料消費量ＱＦとの関係を表す一次式ｍがこれら単位期間当たりの還元剤供給量ＱＲに基づき決定される。次いで、先のＳＯ_Ｘ処理が完了してから（図５１の矢印Ｘ１参照）、燃料消費量ＱＦが第１の設定量ｑｆ１だけ増大すると、このときの単位期間当たりの還元剤供給量の推定値ＱＲＥが一次式ｍを用いて算出される。
【０２５１】
更に、単位期間当たりの還元剤供給量ＱＲと蓄積ＳＯ_Ｘ量推定値ＱＳＥとの関係は予め実験により求められており、図５２に示されるマップの形で予めＲＯＭ４２内に記憶されている。従って、ＳＯ_Ｘ処理を行うべきと判断されたときの単位期間当たりの還元剤供給量の推定値ＱＲＥが算出されれば、このときの蓄積ＳＯ_Ｘ量推定値ＱＳＥが図５２のマップを用いて求められる。
【０２５２】
従って、一般的に表現すると、単位期間当たりにＮＯ_Ｘ処理により消費された燃料又は還元剤の量を求め、単位期間当たりにＮＯ_Ｘ処理により消費された燃料又は還元剤の量に基づいてＮＯ_Ｘ吸収剤の蓄積ＳＯ_Ｘ量を求めているということになる。
【０２５３】
図５３は本発明による第５実施例の係数ｋｍｉの算出ルーチンを示している。本発明による第５実施例でも図１５に示されるＳＯ_Ｘ制御ルーチンが実行され、この係数ｋｍｉの算出ルーチンはＳＯ_Ｘ制御ルーチンのステップ２０３で実行される。
【０２５４】
図５３を参照すると、ステップ４６０では現在の燃料消費量ＱＦと、先の還元剤量算出時期における燃料消費量ＱＦＰ２との差が第２の設定量ｑｆ２以上か否かが判別される。ＱＦ−ＱＦＰ２＜ｑｆ２のとき、即ち先の単位当たりの還元剤供給量ＱＲが算出されてから燃料が第２の設定量ｑｆ２だけ消費されていないときには次いでステップ４６１に進み、ＮＯ_Ｘ処理が行われているか否かが判別される。ＮＯ_Ｘ処理が行われていないときには処理サイクルを終了し、ＮＯ_Ｘ処理が行われているときには次いでステップ４６２に進み、今回のＮＯ_Ｘ処理において還元剤供給弁３１から供給される還元剤量ｑｒが算出される。続くステップ４６３では、還元剤量ｑｒの積算値Ｓｑｒが算出される（Ｓｑｒ＝Ｓｑｒ＋ｑｒ）。
【０２５５】
ＱＦ−ＱＦＰ２≧ｑｆ２になるとステップ４６０からステップ４６４に進み、積算値Ｓｑｒが単位期間当たりの還元剤供給量ＱＲとされる。続くステップ４６５では、単位当たりの還元剤供給量ＱＲの算出回数を表す算出回数カウンタｎＱＲが１だけインクリメントされる（ｎＱＲ＝ｎＱＲ＋１）。続くステップ４６６では、係数ｋｍｉ（ｉ＝１，２，３，４）が次式に基づいてそれぞれ算出される。
【０２５６】
ｋｍ１＝ｋｍ１＋ＱＦ
ｋｍ２＝ｋｍ２＋ＱＦ^２
ｋｍ３＝ｋｍ３＋ＱＦ・ＱＲ
ｋｍ４＝ｋｍ４＋ＱＲ
続くステップ４６７では、現在の燃料消費量ＱＦがＱＦＰ２として記憶され、積算値Ｓｑｒ及びカウンタｎＱＲがそれぞれクリアされる。
【０２５７】
図５４は本発明による第５実施例の処理開始時量ＱＳＳの設定ルーチンを示している。このルーチンはＳＯ_Ｘ制御ルーチンのステップ２０４で実行される。
【０２５８】
図５４を参照すると、ステップ４７０では単位期間当たりの還元剤供給量の推定値ＱＲＥが次式に基づいて算出される。
【０２５９】
【数９】

【０２６０】
続くステップ４７１では、図５２のマップから蓄積ＳＯ_Ｘ量推定値ＱＳＥが算出される。続くステップ４７２では、処理開始時量ＱＳＳがこの蓄積ＳＯ_Ｘ量推定値ＱＳＥに設定される。続くステップ４７３では、係数ｋｍｉ（ｉ＝１，２，３，４）及び算出回数カウンタｎＱＲがクリアされる。
【０２６１】
本発明による第５実施例のその他の構成及び作用は上述の実施例と同様であるので説明を省略する。
【０２６２】
【発明の効果】
ＮＯ_Ｘ吸収剤内に蓄えられているイオウの量を正確に求めることができる。
【図面の簡単な説明】
【図１】内燃機関の全体図である。
【図２】ＮＯ_Ｘ処理を説明するためのタイムチャートである。
【図３】ＮＯ_Ｘ処理の基本時間間隔ｔＮＢ及び補正量ｋｔＮを示す線図である。
【図４】ＮＯ_Ｘ制御ルーチンを示すフローチャートである。
【図５】補正量ｋｔＮの更新ルーチンを示すフローチャートである。
【図６】本発明による実施例のＳＯ_Ｘ処理を説明するためのタイムチャートである。
【図７】流入ＮＯ_Ｘ濃度ＣＮＩを示す線図である。
【図８】一次式ｍを説明するための線図である。
【図９】蓄積ＳＯ_Ｘ量推定値ＱＳＥとＮＯ_Ｘ浄化率ＥＦＦとの関係を示す線図である。
【図１０】蓄積ＳＯ_Ｘ量推定値ＱＳＥの可能最大量ＱＳＥＭを示す線図である。
【図１１】ＮＯ_Ｘ浄化率平均値ＥＦＦＡＶ等の算出タイミングを説明するためのタイムチャートである。
【図１２】ＮＯ_Ｘ濃度検出期間ＤＰを説明するためのタイムチャートである。
【図１３】ＮＯ_Ｘ浄化率ＥＦＦ等の算出タイミングを説明するためのタイムチャートである。
【図１４】蓄積ＳＯ_Ｘ量推定値ＱＳＥ等の算出タイミングを説明するためのタイムチャートである。
【図１５】ＳＯ_Ｘ制御ルーチンを示すフローチャートである。
【図１６】ＳＯ_Ｘ減少量ｑｓｒを示す線図である。
【図１７】係数ｋｍｉの算出ルーチンを示すフローチャートである。
【図１８】ＮＯ_Ｘ浄化率算出ルーチンを示すフローチャートである。
【図１９】ＮＯ_Ｘ浄化率算出ルーチンを示すフローチャートである。
【図２０】ＮＯ_Ｘ浄化率算出ルーチンを示すフローチャートである。
【図２１】蓄積ＮＯ_Ｘ量推定値ＱＮＥの算出ルーチンを示すフローチャートである。
【図２２】ＮＯ_Ｘ増加量ｑｎａを示す線図である。
【図２３】流入ＮＯ_Ｘ濃度補正値ＣＮＩＣの算出ルーチンを示すフローチャートである。
【図２４】遅れ時間ＤＬＹを示す線図である。
【図２５】流入ＮＯ_Ｘ濃度補正値ＣＮＩＣの算出方法を説明するためのタイムチャートである。
【図２６】流出ＮＯ_Ｘ濃度補正値ＣＮＯＣの算出ルーチンを示すフローチャートである。
【図２７】ＮＯ_Ｘ浄化率補正値ＥＦＦＣの算出ルーチンを示すフローチャートである。
【図２８】補正係数ｋＥ１，ｋＥ２，ｋＥ３を示す線図である。
【図２９】熱劣化カウンタｎＴＣＨの算出ルーチンを示すフローチャートである。
【図３０】給油判定ルーチンを示すフローチャートである。
【図３１】給油判定方法を説明するための図である。
【図３２】処理開始時量ＱＳＳの設定ルーチンを示すフローチャートである。
【図３３】処理開始時量ＱＳＳの設定ルーチンを示すフローチャートである。
【図３４】処理開始時量ＱＳＳの設定ルーチンを示すフローチャートである。
【図３５】本発明による第１実施例を説明するための図である。
【図３６】本発明による第２実施例を説明するための図である。
【図３７】本発明による第２実施例を説明するための図である。
【図３８】蓄積ＳＯ_Ｘ量推定値ＱＳＥと極大値ＣＰＫとの関係を示す線図である。
【図３９】本発明による第２実施例の係数ｋｍｉの算出ルーチンを示すフローチャートである。
【図４０】本発明による第２実施例の処理開始時量ＱＳＳの設定ルーチンを示すフローチャートである。
【図４１】本発明による第３実施例を説明するための図である。
【図４２】本発明による第３実施例を説明するための図である。
【図４３】蓄積ＳＯ_Ｘ量推定値ＱＳＥと流出ＮＯ_Ｘ積算量ＳＱＬとの関係を示す線図である。
【図４４】本発明による第３実施例の係数ｋｍｉの算出ルーチンを示すフローチャートである。
【図４５】本発明による第３実施例の処理開始時量ＱＳＳの設定ルーチンを示すフローチャートである。
【図４６】本発明による第４実施例を説明するための図である。
【図４７】本発明による第４実施例を説明するための図である。
【図４８】蓄積ＳＯ_Ｘ量推定値ＱＳＥと定常時時間間隔比ＲＳＴとの関係を示す線図である。
【図４９】本発明による第４実施例の係数ｋｍｉの算出ルーチンを示すフローチャートである。
【図５０】本発明による第４実施例の処理開始時量ＱＳＳの設定ルーチンを示すフローチャートである。
【図５１】本発明による第５実施例を説明するための図である。
【図５２】蓄積ＳＯ_Ｘ量推定値ＱＳＥと還元剤供給量ＱＲとの関係を示す線図である。
【図５３】本発明による第５実施例の係数ｋｍｉの算出ルーチンを示すフローチャートである。
【図５４】本発明による第５実施例の処理開始時量ＱＳＳの設定ルーチンを示すフローチャートである。
【符号の説明】
１…機関本体
２０ａ，２０ｂ…排気管
２３…ＮＯ_Ｘ吸収剤
３１…還元剤供給弁
５１…ＮＯ_Ｘセンサ[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an exhaust emission control device for an internal combustion engine.
[0002]
[Prior art]
NO in the exhaust gas flowing into the exhaust passage of the internal combustion engine where the combustion is performed under a lean air-fuel ratio when the air-fuel ratio of the exhaust gas flowing in is lean_XNO when storing the reducing agent in the exhaust gas when the air-fuel ratio of the exhaust gas flowing in decreases_XNO is being reduced and stored_XNO decreases in quantity_XPlace the absorbent, NO_XThe amount of change per unit time in the amount of accumulated sulfur, which is the amount of sulfur stored in the absorbent, is determined as NO._XAbsorbent temperature, intake air volume, and NO_XThere is known an internal combustion engine which is calculated based on the air-fuel ratio of exhaust gas flowing into the absorbent and calculates the accumulated sulfur amount by integrating the amount of change (see Patent Document 1).
[0003]
[Patent Document 1]
JP 2001-74727 A
[Patent Document 2]
Japanese Patent Laid-Open No. 11-229858
[Patent Document 3]
JP 2000-274229 A
[Patent Document 4]
JP 2002-266628 A
[Patent Document 5]
JP 2002-221028 A
[Patent Document 6]
JP 2000-104536 A
[0004]
[Problems to be solved by the invention]
NO when the amount of accumulated sulfur is small_XNO that can be stored in the absorbent_XThe amount of accumulated sulfur increases, but NO increases._XNO that can be stored in the absorbent_XAs a result, the amount of NO decreases_XNO in exhaust gas flowing out from absorbent_XThe amount of spillage NO_XThe amount can be large. In this way, outflow NO_XThe amount is an indicator of the amount of accumulated sulfur.
[0005]
However, in the above-described internal combustion engine, the outflow NO._XThere is a problem that the amount of accumulated sulfur cannot always be accurately calculated because no amount is taken into consideration.
[0006]
Therefore, the object of the present invention is NO._XAn object of the present invention is to provide an exhaust emission control device for an internal combustion engine that can accurately determine the amount of sulfur stored in an absorbent.
[0007]
[Means for Solving the Problems]
  In order to solve the above-mentioned problem, according to the first invention, the exhaust gas flowing into the exhaust passage of the internal combustion engine in which combustion is performed under the lean air-fuel ratio flows when the air-fuel ratio of the exhaust gas flowing in is lean. NO inside_XNO when storing the reducing agent in the exhaust gas when the air-fuel ratio of the exhaust gas flowing in decreases_XNO is being reduced and stored_XNO decreases in quantity_XIn an internal combustion engine with an absorbent, NO_XNO in exhaust gas flowing out from absorbent_XThe amount of spillage NO_XNO to detect quantity_XNO placed in the exhaust passage downstream of the absorbent_XSensor and NO_XThe amount of accumulated sulfur, which is the amount of sulfur stored in the absorbent, is converted to the outflow NO._XMeans for determining based on quantityExhaust gas purification device, NO _X A means for performing sulfur treatment for reducing the amount of accumulated sulfur in the absorbent and a means for judging whether or not to perform sulfur treatment are further provided. _X Determine the amount of accumulated sulfur in the absorbent and set the amount at the start of treatment to the obtained amount of accumulated sulfur. _X The sulfur treatment required to reduce the accumulated sulfur amount of the absorbent from the amount at the start of the treatment to a predetermined lower limit amount or less is performed, and the obtained NO _X When the accumulated sulfur amount of the absorbent is larger than the predetermined boundary amount, the processing start amount is set to the obtained accumulated sulfur amount and NO. _X Sulfur treatment necessary to reduce the accumulated sulfur amount of the absorbent from the amount at the start of the treatment to the lower limit amount or less is performed, and the obtained NO _X When the amount of accumulated sulfur in the absorbent is smaller than the boundary amount, NO is set by setting the amount at the start of processing to an amount unrelated to the obtained accumulated sulfur amount. _X To perform sulfur treatment necessary to reduce the amount of accumulated sulfur in the absorbent from the amount at the start of treatment to the lower limit amount or less.is doing.
[0010]
  Also,2According to the second invention1In the second invention, when it is assumed that the fuel is a fuel having a reference concentration of sulfur, the amount is irrelevant to the obtained accumulated sulfur amount._XThe maximum amount of sulfur that can be stored in the absorbent is set.
[0011]
  Also,3According to the second invention1In the second invention, spilled NO_XThe quantity is detected several times at intervals of each other, and the spillage NO._XNO based on quantity_XThe amount of accumulated sulfur in the absorbent is determined, and the detected outflow NO_XWhen the number of times when the amount is less than the predetermined allowable minimum amount exceeds the upper limit number, the processing start amount is set to an amount irrelevant to the obtained accumulated sulfur amount and NO._XSulfur treatment necessary to reduce the accumulated sulfur amount of the absorbent from the amount at the start of the treatment to the lower limit amount or less is performed.
[0012]
  Also,4According to the second inventionNO in the exhaust gas flowing into the exhaust passage of the internal combustion engine where the combustion is performed under the lean air-fuel ratio when the air-fuel ratio of the exhaust gas flowing in is lean _X NO when storing the reducing agent in the exhaust gas when the air-fuel ratio of the exhaust gas flowing in decreases _X NO is being reduced and stored _X NO decreases in quantity _X In an internal combustion engine with an absorbent, NO _X NO in exhaust gas flowing out from absorbent _X The amount of spillage NO _X NO to detect quantity _X NO placed in the exhaust passage downstream of the absorbent _X Sensor and NO _X The amount of accumulated sulfur, which is the amount of sulfur stored in the absorbent, is converted into the outflow NO. _X NOx purification device comprising means for obtaining based on quantity, NO _X A means for performing sulfur treatment for reducing the amount of accumulated sulfur in the absorbent and a means for judging whether or not to perform sulfur treatment are further provided. _X Determine the amount of accumulated sulfur in the absorbent and set the amount at the start of treatment to the obtained amount of accumulated sulfur. _X The sulfur treatment necessary to reduce the amount of sulfur accumulated in the absorbent from the amount at the start of the treatment to a predetermined lower limit or less is performed,When the set processing start amount is assumed that the fuel is the fuel having the highest sulfur concentration among the fuels available in the market, NO._XWhen the amount of sulfur that can be stored in the absorbent is greater than the maximum amount that can be stored, the amount at the start of processing is set to the maximum amount that can be stored and NO._XSulfur treatment necessary to reduce the accumulated sulfur amount of the absorbent from the amount at the start of the treatment to the lower limit amount or less is performed.
[0014]
  Also,5According to the second invention1 or 4In the second aspect, it is determined that the sulfur treatment should be performed when the fuel is consumed by the first predetermined amount after the previous sulfur treatment is completed.
[0015]
  Also,6According to the second inventionNO in the exhaust gas flowing into the exhaust passage of the internal combustion engine where the combustion is performed under the lean air-fuel ratio when the air-fuel ratio of the exhaust gas flowing in is lean _X NO when storing the reducing agent in the exhaust gas when the air-fuel ratio of the exhaust gas flowing in decreases _X NO is being reduced and stored _X NO decreases in quantity _X In an internal combustion engine with an absorbent, NO _X NO in exhaust gas flowing out from absorbent _X The amount of spillage NO _X NO to detect quantity _X NO placed in the exhaust passage downstream of the absorbent _X Sensor and NO _X The amount of accumulated sulfur, which is the amount of sulfur stored in the absorbent, is converted into the outflow NO. _X An exhaust emission control device comprising means for obtaining based on a quantity,NO_XNO in the exhaust gas flowing into the absorbent_XInflow NO_XMeans for determining the amount and inflow NO_XVolume and spill NO_XNO based on quantity_XAbsorbent NO_XMeans for determining the purification rate and the determined NO_XNO based on purification rate_XAnd a means for determining the amount of accumulated sulfur in the absorbent.
[0016]
  Also,7According to the second invention6In the second invention, NO_XAbsorbent NO_XThe purification rate is obtained several times at intervals of each other, and these obtained NOs are obtained._XA relational expression representing the relation between the purification rate and the fuel consumption is obtained, and the NO at the judgment time is determined from the fuel consumption at the judgment time and the relational expression._XThe purification rate is obtained, and NO at the determined judgment time._XBased on the purification rate, the amount of accumulated sulfur at the determination time is obtained.
[0017]
  Also,8According to the second invention7In the second invention, the determined NO_XThe relationship between the purification rate and the fuel consumption is expressed by a linear expression, and the linear expression is obtained by the method of least squares.
[0018]
  Also,9According to the second invention7In the second invention, NO_XAbsorbent NO_XThe purification rate is obtained several times at intervals of each other, and the NO obtained during the period from the previous average value calculation time to the current average value calculation time every time the average value calculation time arrives._XThe average value of the purification rate is calculated, and the calculated NO_XA relational expression representing the relation between the average value of the purification rate and the fuel consumption is obtained.
[0019]
  Also,10According to the second invention9In the second invention, NO_XNO stored in the absorbent_XNO_XNO stored in the absorbent_XNO to reduce the amount of_XNO temporarily switching the air-fuel ratio of the exhaust gas flowing into the absorbent to rich_XMeans for processing and NO_XOutflow NO when the air-fuel ratio of the exhaust gas flowing into the absorbent is kept lean_XNO so that the amount does not exceed the predetermined maximum allowable amount_XMeans for setting a time interval at which processing is performed, and NO_XInflow NO when the air-fuel ratio of the exhaust gas flowing into the absorbent is lean_XVolume and spill NO_XWhile calculating the amount, these inflow NO_XVolume and spill NO_XNO based on quantity_XAbsorbent NO_XThe purification rate is calculated, and NO is determined during the period from the previous average value calculation time to the current average value calculation time._XWhen the number of times the purification rate has been obtained is less than the lower limit number, NO during that period_XThe average value of the purification rate is not calculated.
[0020]
  Also,11According to the second invention9In the second invention, NO calculated every time the average value calculation time comes_XThe average value of the purification rate is NO_XNO weighted according to whether steady engine operation or transient engine operation is performed when the purification rate is determined_XIt is a weighted average value of the purification rate.
[0021]
  Also,12According to the second invention9In the second invention, it is determined that the next average value calculation time has come when the fuel is consumed by a predetermined second set amount from the previous average value calculation time.
[0022]
  Also,13According to the second invention7In the second invention, NO_XMeans for performing sulfur treatment to reduce the amount of accumulated sulfur in the absorbent, and the determined NO_XWhen the accumulated sulfur amount of the absorbent is larger than a predetermined boundary amount, it is determined that the fuel is a high-sulfur concentration fuel, and the obtained NO_XAnd a means for determining that the fuel is a low-sulfur concentration fuel when the amount of accumulated sulfur in the absorbent is less than the boundary amount, and when the fuel is determined to be a high-sulfur concentration fuel, the processing start amount is determined. NO is set to the obtained accumulated sulfur amount._XWhen the sulfur treatment necessary to reduce the amount of accumulated sulfur in the absorbent from the amount at the start of the treatment to a predetermined lower limit or less is performed, and it is determined that the fuel is a low-sulfur concentration fuel, the amount at the start of treatment NO is set to an amount irrelevant to the obtained accumulated sulfur amount._XSulfur treatment necessary to reduce the accumulated sulfur amount of the absorbent from the amount at the start of the treatment to the lower limit amount or less is performed.
[0023]
  Also,14According to the second invention13In the second invention, the determined NO_XMeans for obtaining a deviation of the purification rate from the relational expression is further provided, and when the obtained deviation changes beyond a predetermined threshold when it is determined that the fuel is a high-sulfur concentration fuel, the fuel Is changed to a low-sulfur concentration fuel, and when it is determined that the fuel is a low-sulfur concentration fuel, the fuel is changed to a high-sulfur concentration fuel when the obtained deviation changes beyond the threshold value. Judging that it was done.
[0024]
  Also,15According to the second invention6In the second invention, NO_XOutflow NO set during a period when the air-fuel ratio of the exhaust gas flowing into the absorbent is lean_XOutflow NO during volume detection period_XWhile calculating the average value of the amount, the outflow NO_XInflow NO during volume detection period_XCalculate the average amount of these inflow NO_XAverage volume and spill NO_XThe spill NO based on the average amount_XNO in quantity detection period_XThe purification rate is calculated.
[0025]
  Also,16According to the second invention6In the second invention, inflow NO_XVolume and spill NO_XThe amount of exhaust gas is required at almost the same time, and the exhaust gas part is inflow NO._XNO after quantity is requested_XMeans for determining the delay time required to reach the sensor, and inflow NO with the delay time_XVolume or spill NO_XAnd a means for correcting the amount, the corrected inflow NO._XVolume or spill NO_XNO based on quantity_XThe purification rate is calculated.
[0026]
  Also,17According to the second invention6In the second invention, the determined NO_XNO purification rate_XNO when the absorbent is in a predetermined reference state_XA means for converting to a purification rate is further provided, and the converted NO_XNO based on purification rate_XThe amount of accumulated sulfur in the absorbent is calculated.
[0027]
  Also,18According to the second invention6In the second invention, NO_XMeans for determining the degree of thermal degradation of the absorbent and the determined NO_XThe obtained NO based on the degree of thermal deterioration of the absorbent._XMeans for correcting the purification rate, and the corrected NO._XNO based on purification rate_XThe amount of accumulated sulfur in the absorbent is calculated.
[0028]
  Also,19According to the second invention6In the second aspect of the invention, the apparatus further comprises means for detecting an engine operating state, and the inflow NO is based on the detected engine operating state._XThe amount is calculated.
[0030]
  Also,20According to the second inventionNO in the exhaust gas flowing into the exhaust passage of the internal combustion engine where the combustion is performed under the lean air-fuel ratio when the air-fuel ratio of the exhaust gas flowing in is lean _X NO when storing the reducing agent in the exhaust gas when the air-fuel ratio of the exhaust gas flowing in decreases _X NO is being reduced and stored _X NO decreases in quantity _X In an internal combustion engine with an absorbent, NO _X NO in exhaust gas flowing out from absorbent _X The amount of spillage NO _X NO to detect quantity _X NO placed in the exhaust passage downstream of the absorbent _X Sensor and NO _X The amount of accumulated sulfur, which is the amount of sulfur stored in the absorbent, is converted into the outflow NO. _X An exhaust emission control device comprising means for obtaining based on a quantity,NO_XNO stored in the absorbent_XNO_XNO stored in the absorbent_XNO to reduce the amount of_XNO temporarily switching the air-fuel ratio of the exhaust gas flowing into the absorbent to rich_XMeans for processing and NO_XOutflow NO generated when processing is performed_XMeans for detecting the maximum value of the quantity, and the detected outflow NO._XNO based on local maximum_XThe amount of accumulated sulfur in the absorbent is calculated.
[0031]
  Also,21According to the second inventionNO in the exhaust gas flowing into the exhaust passage of the internal combustion engine where the combustion is performed under the lean air-fuel ratio when the air-fuel ratio of the exhaust gas flowing in is lean _X NO when storing the reducing agent in the exhaust gas when the air-fuel ratio of the exhaust gas flowing in decreases _X NO is being reduced and stored _X NO decreases in quantity _X In an internal combustion engine with an absorbent, NO _X NO in exhaust gas flowing out from absorbent _X The amount of spillage NO _X NO to detect quantity _X NO placed in the exhaust passage downstream of the absorbent _X Sensor and NO _X The amount of accumulated sulfur, which is the amount of sulfur stored in the absorbent, is converted into the outflow NO. _X An exhaust emission control device comprising means for obtaining based on a quantity,NO_XNO stored in the absorbent_XNO_XNO stored in the absorbent_XNO to reduce the amount of_XNO temporarily switching the air-fuel ratio of the exhaust gas flowing into the absorbent to rich_XMeans for processing and NO_XNO when processing is done_XNO flowing out of the absorbent_XOutflow NO that is the accumulated amount of_XMeans for obtaining the integrated amount of the quantity, and the obtained outflow NO._XNO based on the integrated amount_XThe amount of accumulated sulfur in the absorbent is calculated.
[0032]
  Also,22According to the second inventionNO in the exhaust gas flowing into the exhaust passage of the internal combustion engine where the combustion is performed under the lean air-fuel ratio when the air-fuel ratio of the exhaust gas flowing in is lean _X NO when storing the reducing agent in the exhaust gas when the air-fuel ratio of the exhaust gas flowing in decreases _X NO is being reduced and stored _X NO decreases in quantity _X In an internal combustion engine with an absorbent, NO _X NO in exhaust gas flowing out from absorbent _X The amount of spillage NO _X NO to detect quantity _X NO placed in the exhaust passage downstream of the absorbent _X Sensor and NO _X The amount of accumulated sulfur, which is the amount of sulfur stored in the absorbent, is converted into the outflow NO. _X An exhaust emission control device comprising means for obtaining based on a quantity,NO_XNO stored in the absorbent_XNO_XNO stored in the absorbent_XNO to reduce the amount of_XNO temporarily switching the air-fuel ratio of the exhaust gas flowing into the absorbent to rich_XMeans for processing and NO_XOutflow NO when the air-fuel ratio of the exhaust gas flowing into the absorbent is kept lean_XNO so that the amount does not exceed the predetermined maximum allowable amount_XMeans for setting a time interval for processing, and NO based on the set time interval_XAnd a means for determining the amount of accumulated sulfur in the absorbent.
[0034]
In this specification, the ratio of the air supplied into the exhaust passage upstream of a certain position of the exhaust passage, the combustion chamber, and the intake passage and the reducing agent such as hydrocarbon HC and carbon monoxide CO This is called the air-fuel ratio of the exhaust gas at the position.
[0035]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 shows a case where the present invention is applied to a compression ignition type internal combustion engine. The present invention can also be applied to a spark ignition type internal combustion engine.
[0036]
Referring to FIG. 1, 1 is an engine body, 2 is a cylinder block, 3 is a cylinder head, 4 is a piston, 5 is a combustion chamber, 6 is an electrically controlled fuel injection valve for directly injecting fuel into the cylinder, 7 Is an intake valve, 8 is an intake port, 9 is an exhaust valve, and 10 is an exhaust port. The intake port 8 is connected to a surge tank 12 via a corresponding intake branch pipe 11, and the surge tank 12 is connected to an outlet of a compressor 15 of an exhaust turbocharger 14 via an intake duct 13. An intake pipe 13 a is connected to the inlet of the compressor 15. A throttle valve 17 driven by a step motor 16 is arranged in the intake duct 13, and a cooling device 18 for cooling the intake air flowing in the intake duct 13 is arranged around the intake duct 13 downstream of the compressor 15. The
[0037]
On the other hand, the exhaust port 10 is connected to the inlet of the exhaust turbine 21 of the exhaust turbocharger 14 via the exhaust manifold 19 and the exhaust pipe 20, and the outlet of the exhaust turbine 21 is connected to the catalytic converter 22 via the exhaust pipe 20a. In the catalytic converter 22, a particulate filter 22a for collecting particulates in the exhaust gas is accommodated, and NO on the particulate filter 22a as will be described later._XAn absorbent 23 is supported. The catalytic converter 22 is connected to the exhaust pipe 20b.
[0038]
Still referring to FIG. 1, the exhaust manifold 19 and the surge tank 12 are connected to each other via an exhaust gas recirculation (hereinafter referred to as EGR) passage 24, and an electrically controlled EGR control valve 25 is provided in the EGR passage 24. Be placed. A cooling device 26 for cooling the EGR gas flowing in the EGR passage 24 is disposed around the EGR passage 24 upstream of the EGR control valve 25.
[0039]
On the other hand, each fuel injection valve 6 is connected to a fuel reservoir, so-called common rail 27, through a fuel supply pipe 6a. Fuel is supplied into the common rail 27 from an electrically controlled fuel pump 28 with variable discharge amount, and the fuel supplied into the common rail 27 is supplied to the fuel injection valve 6 via each fuel supply pipe 6a. A fuel pressure sensor 29 for detecting the fuel pressure in the common rail 27 is attached to the common rail 27, and a fuel pump 28 is configured so that the fuel pressure in the common rail 27 becomes the target fuel pressure based on the output signal of the fuel pressure sensor 29. The discharge amount is controlled. In FIG. 1, reference numeral 30 denotes a fuel tank.
[0040]
An electric control type reducing agent supply valve 31 for supplying a reducing agent into the exhaust manifold 19 is attached to the exhaust manifold 19. In the embodiment according to the present invention, fuel is used as a reducing agent, and the reducing agent supply valve 31 is connected to the common rail 27 through a reducing agent supply pipe 31a.
[0041]
The electronic control unit 40 is composed of a digital computer, and a ROM (Read Only Memory) 42, a RAM (Random Access Memory) 43, and a B-RAM (Backup RAM) always connected to a power source are connected to each other by a bidirectional bus 41. 43a, a CPU (microprocessor) 44, an input port 45 and an output port 46. An air flow meter 49 for detecting the intake air mass flow rate is attached to the intake pipe 13a, and a water temperature sensor 50 for detecting the engine cooling water temperature THW is attached to the engine body 1. NO_XIn the exhaust pipe 20b downstream of the absorbent 23, NO_XNO in exhaust gas flowing out from absorbent 23_XConcentration or spill NO_XNO for detecting concentration CNO_XSensor 51 and NO_XAn exhaust temperature sensor 52 for detecting the temperature of the exhaust gas flowing out from the absorbent 23 is attached. The temperature of the exhaust gas detected by the exhaust temperature sensor 52 is NO._XThe temperature TC of the absorbent 23 is shown. Fuel pressure sensor 29, air flow meter 49, water temperature sensor 50, NO_XOutput signals of the sensor 51 and the exhaust temperature sensor 52 are input to the input port 45 via the corresponding AD converters 47, respectively.
[0042]
Further, a load sensor 53 that generates an output voltage proportional to the amount of depression of the accelerator pedal is connected to an accelerator pedal (not shown), and the output voltage of the load sensor 53 is connected to the input port 45 via a corresponding AD converter 47. Entered. Here, the amount of depression of the accelerator pedal represents the required load L. Further, a crank angle sensor 54 that generates an output pulse every time the crankshaft rotates, for example, 30 ° is connected to the input port 45. The CPU 44 calculates the engine speed N based on this output pulse.
[0043]
On the other hand, the output port 46 is connected to the fuel injection valve 6, the throttle valve driving step motor 16, the EGR control valve 25, the fuel pump 28, and the reducing agent supply valve 31 through corresponding drive circuits 48.
[0044]
The particulate filter 22a is made of a porous material such as cordierite, and NO on both side surfaces of the partition walls and the inner wall surfaces of the pores._XAbsorbents 23 are respectively carried. This NO_XThe absorbent 23 has, for example, alumina as a carrier. On this carrier, for example, potassium K, sodium Na, lithium Li, alkali metal such as cesium Cs, alkaline earth such as barium Ba, calcium Ca, lanthanum La, yttrium Y, etc. At least one selected from such rare earths and a noble metal such as platinum Pt, palladium Pd, rhodium Rh, and iridium Ir are supported.
[0045]
NO_XThe absorbent is NO when the average air-fuel ratio of the inflowing exhaust gas is lean_XNO when storing the reducing agent in the exhaust gas when the air-fuel ratio of the exhaust gas flowing in decreases_XNO is being reduced and stored_XAccumulation and reduction action to reduce the amount of.
[0046]
NO_XThe detailed mechanism of the accumulation and reduction action of the absorbent is not completely clarified. However, the mechanism currently considered can be briefly described as follows, taking as an example the case where platinum Pt and barium Ba are supported on a support.
[0047]
That is, NO_XWhen the air-fuel ratio of the exhaust gas flowing into the absorbent becomes considerably leaner than the stoichiometric air-fuel ratio, the oxygen concentration in the inflowing exhaust gas greatly increases, and oxygen O₂Is O₂ ⁻Or O^2-It adheres to the surface of platinum Pt. On the other hand, NO in the inflowing exhaust gas adheres to the surface of platinum Pt, and O on the surface of platinum Pt.₂ ⁻Or O^2-Reacts with NO₂(NO + O₂→ NO₂+ O^*Where O^*Is active oxygen). Then the generated NO₂Part of the NO is being oxidized further on platinum Pt_XNitrate ion NO while being absorbed in the absorbent and combined with barium oxide BaO₃ ⁻NO in the form of_XDiffuses in the absorbent. In this way NO_XIs NO_XStored in absorbent.
[0048]
In contrast, NO_XWhen the air-fuel ratio of the exhaust gas flowing into the absorbent becomes rich or stoichiometric, the oxygen concentration in the exhaust gas decreases and NO₂Production amount decreases and the reaction proceeds in the reverse direction (NO₃ ⁻→ NO + 2O^*) And thus NO_XNitrate ion NO in absorbent₃ ⁻NO in the form of NO_XReleased from the absorbent. This released NO_XIf the exhaust gas contains a reducing agent, that is, HC or CO, it can be reduced by reacting with the HC and CO. In this way, NO on the surface of platinum Pt._XNO when no longer exists_XNO from absorbent to next_XIs released and reduced, NO_XNO stored in the absorbent_XThe amount of is gradually reduced.
[0049]
NO without forming nitrate_XStore NO_XNO without releasing_XIt is also considered possible to reduce In addition, active oxygen O^*If you pay attention to, NO_XAbsorbent is NO_XWith the accumulation and release of oxygen^*It can also be regarded as an active oxygen generating catalyst that generates
[0050]
The internal combustion engine shown in FIG. 1 continues to burn under a lean air-fuel ratio, and therefore NO._XThe air-fuel ratio of the exhaust gas flowing into the absorbent 23 is kept lean. As a result, NO in the exhaust gas_XIs NO_XIt is stored in the absorbent 23.
[0051]
However, over time, NO_XNO stored in absorbent 23_XNO accumulated_XThe amount increases gradually. Therefore, in an embodiment according to the present invention, NO_XNO stored in absorbent 23_XNO_XAccumulated NO in absorbent 23_XIn order to decrease the amount, the reducing agent is temporarily supplied from the reducing agent supply valve 31 and NO._XNO temporarily switching the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 to rich_XProcessing is performed. NO_XIn order to temporarily switch the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 to be rich, for example, additional fuel may be injected from the fuel injection valve 6 into the expansion stroke or the exhaust stroke, or the combustion chamber The air-fuel ratio of the air-fuel mixture burned in the engine 5 may be temporarily switched to rich.
[0052]
For example, NO_XNO accumulated in absorbent 23_XDetermine the amount of this accumulated NO_XNO every time the amount exceeds a certain amount_XProcessing can also be performed. However, in the embodiment according to the present invention, as indicated by the arrow RN in FIG._XThe processing is repeatedly performed at a time interval tN. This time interval tN is obtained by correcting the basic time interval tNB with the correction amount ktN (tN = tNB + ktN). The basic time interval tNB and the correction amount ktN can be obtained as follows.
[0053]
That is, as shown in FIG._XWhen the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is kept lean, the outflow NO_XConcentration CNO increases little by little and NO_XIt becomes ZCNO immediately before the processing is started. This ZCNO is immediately before NO_XIn terms of concentration, if the time interval tN is short, the immediately preceding NO_XWhen the concentration ZCNO is low and the time interval tN is long, immediately before NO_XThe concentration ZCNO increases. NO immediately before_XThe concentration ZCNO is preferably as low as possible, but if the time interval tN is shortened to shorten the time interval tN, the combustion consumption rate increases. Also, immediately before NO_XThe concentration ZCNO is NO, as will be described later._XSO stored in the absorbent 23_XThe amount of accumulated SO_XIt can vary depending on the amount.
[0054]
Therefore, in the embodiment according to the present invention, the engine is in steady operation and the accumulated SO_XWhen the amount is a reference amount, for example, almost zero, NO_XOutflow NO when the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is kept lean_XConcentration CNO does not exceed the allowable maximum concentration, and NO_XThe basic time interval tNB is set in advance so that the fuel consumption rate by the processing does not exceed the allowable maximum value.
[0055]
In addition, NO during engine transient operation_XOutflow NO when the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is kept lean_XConcentration CNO does not exceed the allowable maximum concentration, and NO_XThe correction amount ktN is updated so that the fuel consumption rate by the process does not exceed the allowable maximum value, and the basic time interval tNB is corrected with the updated correction amount ktN.
[0056]
In the embodiment according to the present invention, as shown in FIG. 3 (A), the engine operating region determined by the required load L and the engine speed N is divided into nine regions, and the basic time interval tNB for each region. Is set. Further, the basic time interval tNB is obtained in advance by experiments, and is stored in advance in the ROM 42 in the form of a map shown in FIG. On the other hand, as shown in FIG. 3B, the correction amount ktN is also set for each region set in the same manner as the region for the basic time interval tNB, and B in the form of the map shown in FIG. -Stored in the RAM 43a.
[0057]
FIG. 4 shows the NO of the embodiment according to the present invention._XA control routine is shown. This routine is executed by interruption every predetermined time.
[0058]
Referring to FIG. 4, first, in step 100, NO_XIt is determined whether or not the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is lean. When the air-fuel ratio AFN of the exhaust gas is not lean, the processing cycle is terminated, and when the air-fuel ratio AFN of the exhaust gas is lean, the routine proceeds to step 101 where the region to which the current engine operating state belongs is determined. In the subsequent step 102, the time interval tN is calculated based on the basic time interval tNB and the correction amount ktN set for the region determined in step 101 (tN = tNB + ktN). In the following step 103, the previous NO_XIt is determined whether time tN has elapsed since the process. Previous NO_XWhen the time tN has not elapsed since the processing, the processing cycle is terminated and the previous NO_XWhen time tN has elapsed from the processing, the routine proceeds to step 104 and NO_XProcessing is performed. Specifically, NO_XThe reducing agent is temporarily supplied from the reducing agent supply valve 31 so that the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 becomes rich.
[0059]
FIG. 5 shows a correction amount ktN update routine according to the embodiment of the present invention. This routine is executed by interruption every predetermined time.
[0060]
Referring to FIG. 5, first, in step 110, NO_XIt is determined whether or not processing has started. NO_XWhen the process has not been started, the routine proceeds to step 111 where the outflow NO at this time_XThe concentration CNO is stored as ZCNO. The processing cycle is then terminated.
[0061]
Then NO_XWhen the process is started, the process proceeds from step 110 to step 112. Therefore, ZCNO is the immediately preceding NO_XRepresents the concentration. In step 112, immediately before NO_XConcentration ZCNO is the aforementioned outflow NO_XIt is determined whether or not the concentration is higher than CNH corresponding to the allowable maximum concentration of CNO. Next, when ZCNO> CNH, the routine proceeds to step 113, where the region to which the current engine operating state belongs is determined, and at the subsequent step 114, the correction amount ktN of this region is decreased by, for example, a constant value Δk. In this way, NO_XOutflow NO when the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is kept lean_XThe time interval tN is set or the basic time interval tN is corrected so that the concentration CNO does not exceed the allowable maximum concentration.
[0062]
On the other hand, when ZCNO ≦ CNH, the routine proceeds to step 115 where the immediately preceding NO_XIt is determined whether or not the concentration ZCNO is smaller than CNL corresponding to the allowable maximum value of the fuel consumption rate described above. When ZCNO <CNL, the routine proceeds to step 116 where the region to which the current engine operating state belongs is determined. In the following step 117, the correction amount ktN in this area is increased by, for example, a constant value Δk. In this way, NO_XThe time interval tN is set or the basic time interval tN is corrected so that the fuel consumption rate by the processing does not exceed the allowable maximum value.
[0063]
On the other hand, when ZCNO ≧ CNL, that is, when CNL ≦ ZCNO ≦ CNH, the processing cycle is terminated without updating the correction amount ktN. In this case, immediately before NO_XIt can also be said that the correction amount ktN is updated so that the concentration ZCNO is maintained within the range from CNL to CNH.
[0064]
In this way NO_XProcessing is performed and the result is NO_XNO that can be stored in the absorbent 23_XThe amount of is increased.
[0065]
However, sulfur content in the exhaust gas is SO._XIs included in the form of NO_XNO in the absorbent 23_XNot only SO_XCan also be stored. This SO_XNO_XThe accumulation mechanism in the absorbent 23 is NO._XThis is considered to be the same as the accumulation mechanism. That is, when the case where platinum Pt and barium Ba are supported on a carrier is simply described as an example, NO_XWhen the air-fuel ratio of the exhaust gas flowing into the absorbent 23 is lean, as described above, the oxygen O₂Is O₂ ⁻Or O^2-Attached to the surface of platinum Pt in the form of SO₂Adheres to the surface of platinum Pt and O on the surface of platinum Pt.₂ ⁻Or O^2-Reacts with SO₃It becomes. The generated SO₃NO is being oxidized on platinum Pt_XWhile being absorbed into the absorbent 23 and bonded to barium oxide BaO, sulfate ions SO₄ ⁻NO in the form of_XIt diffuses into the absorbent 23. This sulfate ion SO₄ ⁻Then barium ion Ba⁺Combined with sulfate BaSO₄Is generated.
[0066]
This sulfate BaSO₄Is difficult to decompose, NO_XEven if the air-fuel ratio of the exhaust gas flowing into the absorbent 23 is simply rich, NO_XSulfate BaSO in absorbent 23₄The amount of does not decrease. For this reason, as time passes, NO_XSulfate BaSO in absorbent 23₄The amount of NO increases, resulting in NO_XNO that can be stored in the absorbent 23_XThe amount of will decrease.
[0067]
However, NO_XWhile maintaining the temperature of the absorbent 23 at, for example, 600 ° C. or higher, NO_XWhen the average air-fuel ratio of the exhaust gas flowing into the absorbent 23 is made the stoichiometric air-fuel ratio or rich, NO_XSulfate BaSO in absorbent 23₄Decomposes into SO₃NO in the form of_XReleased from the absorbent 23. This released SO₃If exhaust gas contains a reducing agent, ie, HC or CO, it reacts with HC and CO to react with SO.₂To be reduced. In this way NO_XIn the absorbent 23, sulfate BaSO₄SO stored in the form of_XThe amount of NO decreases gradually, at this time NO_XAbsorbent 23 to SO_XIs SO₃It will not leak out.
[0068]
Therefore, in an embodiment according to the present invention, NO_XAccumulated SO in the absorbent 23_XNO to reduce the amount_XWhile maintaining the temperature of the absorbent 23 at the sulfate decomposition temperature, for example, 600 ° C. or higher, the reducing agent is supplied from the reducing agent supply valve 31 and NO._XSO that temporarily switches the average air-fuel ratio of the exhaust gas flowing into the absorbent 23 to slightly rich_XProcessing is performed. NO_XIn order to temporarily change the average air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 to be slightly slightly rich, for example, additional fuel may be injected from the fuel injection valve 6 into the expansion stroke or the exhaust stroke. The air-fuel ratio of the air-fuel mixture burned in the combustion chamber 5 may be temporarily switched to rich.
[0069]
For example, NO_XAccumulated SO of absorbent 23_XDetermine the amount of this accumulated SO_XEvery time the amount exceeds a certain amount, SO_XProcessing can also be performed. However, in the embodiment according to the present invention, as shown in FIG._XAfter the processing is completed (arrow X1), when the fuel consumption amount QF increases by the first set amount qf1, the next SO_XProcessing is started (arrow X2). Or the previous SO_XWhen the fuel consumption QF when the processing is completed is QFP1, when the fuel consumption QF increases from QFP1 by qf1, the next SO_XProcessing begins. The fuel consumption amount QF is the sum of the amount of fuel injected from the fuel injection valve 6 and the amount of reducing agent or fuel supplied from the reducing agent supply valve 31.
[0070]
SO_XThere are various ways to determine how long the process will take place, for example SO_XProcessing can also be performed for a certain period of time. However, in embodiments according to the present invention, the accumulated SO_XSO required to reduce the amount from the processing start amount QSS to a predetermined lower limit value QSL or less_XProcessing is performed.
[0071]
Specifically, the previous SO_XWhen the fuel consumption amount QF increases by the first set amount qf1 after the processing is completed, first, at this time, the accumulated SO_XThe amount is calculated, and the processing start time amount QSS is the accumulated SO_XSet to quantity, SO_XProcessing is started. SO_XWhen the process is started, as shown in FIG._XAccumulated SO representing quantity_XThe quantity representative value QS is successively subtracted from the processing start time quantity QSS, and then stored SO_XWhen the quantity representative value QS is less than or equal to the lower limit value QSL, SO_XThe process is completed and the result is NO_XThe average air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is returned to lean (arrow X3).
[0072]
Therefore, in general terms, SO_XWhen it is determined that processing should be performed, the accumulated SO at this time_XAlong with obtaining the quantity, the processing start time quantity QSS is obtained from the obtained accumulated SO._XSet to amount NO_XAccumulated SO of absorbent_XSO required to reduce the amount from the processing start amount QSS to the lower limit value QSL or less_XThis means that processing is performed.
[0073]
Here, in the embodiment according to the present invention, the lower limit value QSL is set to almost zero, so that the accumulated SO_XSO required to reduce the amount by approximately the amount at the start of processing_XProcessing will be performed.
[0074]
Although not shown in FIG. 6, the SO_XNO when processing is not performed (from arrow X1 to X2)_XThe process is repeated, so NO_XThe air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is repeatedly rich temporarily.
[0075]
NO_XAccumulated SO in the absorbent 23_XIt is difficult to determine the quantity directly. Thus, in each embodiment according to the present invention, the accumulated SO_XThe amount is estimated. Next, the accumulated SO of the first embodiment according to the present invention._XA method for calculating the quantity estimated value QSE will be described.
[0076]
In a first embodiment according to the present invention, SO_XMore precisely when processing is not being done_XNO when the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is lean_XNO in absorbent 23_XThe purification rate EFF is calculated a plurality of times at time intervals, and this NO_XAccumulated SO based on purification rate EFF_XA quantity estimate QSE is calculated. Where NO_XNO in the exhaust gas flowing into the absorbent 23_XInflow NO as concentration_XIf the concentration is CNI, NO_XNO in absorbent 23_XThe purification rate EFF is expressed by the following equation.
[0077]
EFF = (CNI-CNO) / CNI
Outflow NO_XThe concentration CNO is NO as described above._XIt is detected by the sensor 51. On the other hand, inflow NO_XIn order to obtain the concentration CNI, for example, NO in the exhaust pipe 20a_XA sensor can be attached, but in the first embodiment according to the present invention, inflow NO._XThe concentration CNI is obtained based on the engine operating state. That is, outflow NO_XThe concentration CNI is obtained in advance by experiments as a function of the required load L and the engine speed N, and is stored in advance in the ROM 42 in the form of a map shown in FIG.
[0078]
FIG. 8 shows a plurality of calculated NOs._XAn example of the result of plotting the purification rate EFF against the fuel consumption QF is shown. In FIG. 8, EFF (i), EFF (i + 1), and EFF (i + 2) are NO when the fuel consumption QF is QF (i), QF (i + 1), and QF (i + 2), respectively._XRepresents the purification rate EFF, i, i + 1, i + 2 are NO_XThe number of purification rate calculations is represented (i = 1, 2,...).
[0079]
As can be seen from FIG. 8, NO increases as the fuel consumption QF increases._XThe purification rate EFF decreases, fuel consumption QF and NO_XThe relationship with the purification rate EFF can be generally expressed by the primary expression m.
[0080]
Here, the slope of the primary equation m is NO per unit fuel consumption._XIt represents the amount of reduction in the purification rate EFF. As will be described later, the accumulated SO_XNO as the amount increases_XThe purification rate EFF decreases, NO_XReduction amount of purification rate EFF is accumulated SO_XIt represents the amount of increase. On the other hand, accumulated SO_XThe amount of increase per unit time is NO_XSO flowing into the absorbent 23 per unit time_XDepending on the amount of this inflow SO_XThe amount depends on the fuel consumption per unit time.
[0081]
Therefore, as long as the type of fuel, specifically the sulfur concentration in the fuel, does not change, the accumulated SO per unit fuel consumption_XThe amount of increase is kept almost constant, NO per unit fuel consumption_XThe reduction amount of the purification rate EFF is also maintained constant. As a result, fuel consumption QF and NO_XThe primary expression m representing the relationship with the purification rate EFF does not change unless the fuel type changes, that is, unless fueling is performed.
[0082]
Then, if the primary equation m is obtained, NO at any time_XThe purification rate EFF can be estimated. That is, in the example shown in FIG. 8, NO when the fuel consumption QF is f._XIt can be estimated that the purification rate EFF is e.
[0083]
Therefore, in the first embodiment according to the present invention, the fuel consumption QF and NO_XA linear expression m representing the relationship with the purification rate EFF is obtained. Specifically, NO_XThe purification rate EFF is calculated at least twice, and the linear expression m is determined by, for example, the least square method.
[0084]
On the other hand, NO_XThe purification rate EFF is the accumulated SO as described above._XDepends on the amount, specifically accumulated SO_XIt becomes smaller as the amount increases. Accumulated SO_XQuantity and NO_XThe relationship with the purification rate EFF is NO_XIt can vary depending on the type of absorbent 23, but NO_XIf the type of the absorbent 23 is determined, this relationship can be obtained in advance by experiments. Figure 9 shows accumulated SO_XQuantity estimate QSE and NO_XThis is an example showing the relationship with the purification rate EFF, which is obtained in advance by experiments and stored in the ROM. In this example, NO_XAccumulated SO when the purification rate EFF is e_XThe quantity estimate QSE is s.
[0085]
Therefore, in the example shown in FIGS. 8 and 9, the accumulated SO when the fuel consumption QF is f._XIt can be seen that the quantity estimated value QSE is s.
[0086]
In a first embodiment according to the present invention, SO_XWhen it is determined that processing should be performed, that is, the previous SO_XWhen the fuel consumption amount QF increases by the first set amount qf1 after the processing is completed, the fuel consumption amount QF and NO_XA primary expression m representing the relationship with the purification rate EFF is obtained, and the NO at this time is calculated using the primary expression m from the fuel consumption QF (= QFP1 + qf1) at this time._XThe purification rate EFF is calculated, and the calculated NO_XAccumulated SO at this time using the map of FIG. 9 from the purification rate_XA quantity estimate QSE is calculated. Next, the processing start time amount QSS is stored in this accumulated SO._XThe quantity estimated value QSE is set.
[0087]
Therefore, generally speaking, spill NO_XConcentration CNO or NO_XAccumulated SO based on purification rate EFF_XIt means that we are looking for quantity. Or NO_XA relational expression representing the relationship between the purification rate and the fuel consumption is obtained, and the NO at the judgment time is determined from the fuel consumption and the relational expression at the judgment time._XObtain the purification rate, NO at the judgment time_XAccumulated SO at the time of judgment based on the purification rate_XIt can also be said that the amount is requested. In this case, the previous SO_XWhen fuel is consumed by a predetermined first set amount after processing is completed, or SO_XThe time when it is determined that the processing should be performed can be considered as the determination time.
[0088]
However, fuel consumption QF and NO as in the primary equation m_XEven without obtaining a relational expression representing the relationship with the purification rate EFF, SO_XOutflow NO at this time when it is determined that processing should be performed_XConcentration CNO is detected and NO at this time_XIf the purification rate EFF is calculated, the accumulated SO at this time using the map of FIG._XA quantity estimation value QSE can be calculated. However, as described above, the relational expression such as the primary expression m is calculated NO several times._XIt is determined based on the purification rate EFF, and this relational expression m does not change as long as the fuel is the same. Therefore, it is better to use the relational expression m to store SO._XThe quantity can be determined more accurately.
[0089]
NO_XPurification rate EFF and accumulated SO_XThe quantity estimate QSE may include an error. That is, as shown in FIG. 10 (A), the calculated NO_XEven if the purification rate EFF is e, for example, the actual accumulated SO_XThe quantity is not necessarily s, but is in the range of QSEM to QSEm. In this case, store QSEM SO_XThe quantity estimated value QSE can be considered as the maximum quantity that can be taken as a reference, that is, the maximum possible quantity, and QSEm is stored as SO._XThe quantity estimated value QSE can be considered as the minimum quantity that can be taken as a reference.
[0090]
Therefore, SO_XNO when it is determined that processing should be performed_XWhen the purification start rate QSS is set to s because the purification rate EFF is e._XIf the quantity is the maximum possible quantity QSEM, the accumulated SO_XSO required to reduce the amount below the lower limit_XEven after processing, NO_XSO in the absorbent 23_XWill remain.
[0091]
So to prevent this, SO_XIt is also possible to obtain the maximum possible amount QSEM when it is determined that the processing should be performed, and to set the processing start time amount QSS to the maximum possible amount QSEM. Here, the maximum possible amount QSEM is NO_XAs a function of the purification rate EFF, for example, it can be stored in the ROM 42 in advance in the form of a map indicated by a solid line in FIG. In FIG. 10B, the broken line indicates the accumulated SO shown in FIG._XThe quantity estimated value QSE is represented.
[0092]
By the way, in the first embodiment according to the present invention, SO_XAccumulated SO when it is determined that processing should be performed_XThe estimated quantity QSE is the accumulated SO when a certain amount of fuel is consumed._XIt generally represents the amount of increase in the amount, so the accumulated SO per unit fuel consumption_XThe amount of increase, that is, the sulfur concentration in the fuel.
[0093]
Then, SO_XAccumulated SO when it is determined that processing should be performed_XWhen the estimated amount QSE is larger than a predetermined boundary amount, the fuel is a high-sulfur concentration fuel, and SO_XAccumulated SO when it is determined that processing should be performed_XWhen the estimated amount QSE is smaller than the boundary amount, it can be determined that the fuel is a low-sulfur concentration fuel or a standard fuel. In the first embodiment according to the present invention, NO_XThe purification rate EFF is accumulated SO_XSince it corresponds to the quantity estimate QSE, SO_XNO when it is determined that processing should be performed_XWhen the purification rate EFF is lower than the boundary purification rate EFFB corresponding to the boundary amount described above, the fuel is a high-sulfur concentration fuel, and the SO_XNO when it is determined that processing should be performed_XWhen the purification rate EFF is higher than the boundary purification rate EFFB, it is determined that the fuel is a low-sulfur concentration fuel or a standard fuel.
[0094]
In addition, when it is determined that the fuel is a low-sulfur concentration fuel or a standard fuel, the processing start amount QSS is set to SO._XAccumulated SO when it is determined that processing should be performed_XThe amount is set to an amount unrelated to the amount estimated value QSE. This irrelevant amount may be determined in any way, but in the first embodiment according to the present invention, NO is assumed when the fuel is assumed to be a fuel having a sulfur concentration of the reference concentration CSR._XMaximum SO that can be stored in the absorbent 23_XSet to quantity. Further, the reference concentration CSR may be determined in any way, but in the first embodiment according to the present invention, the reference concentration CSR is set to a standard sulfur concentration in fuel, for example, 50 ppm. That is, when the fuel whose sulfur concentration is the reference concentration CSR is consumed by the first set amount qf1, NO_XMaximum SO that can be stored in the absorbent 23_XThe quantity is qf1 · CSR, and the processing start time quantity QSS is set to this qf1 · CSR.
[0095]
This is done for the following reason. That is, in the case of fuel with low sulfur concentration, the accumulated SO_XAmount is NO_XThe effect on the purification rate EFF is small, and thus the accumulated SO calculated in this case_XThe accuracy of the quantity estimated value QSE is not necessarily high. Therefore, when it is determined that the fuel is a low-sulfur concentration fuel or a standard fuel, the processing start time amount QSS is calculated as the accumulated SO._XInstead of setting the quantity estimated value QSE, the process start time quantity QSS is set to an amount determined according to the known sulfur concentration, assuming that a fuel having a known sulfur concentration is used. .
[0096]
On the other hand, when it is determined that the fuel is a high-sulfur concentration fuel, as described above, the processing start time amount QSS is equal to the SO._XAccumulated SO when it is determined that processing should be performed_XThe quantity estimated value QSE is set.
[0097]
Therefore, generally speaking, SO_XAccumulated SO when it is determined that processing should be performed_XThe amount is calculated and the calculated accumulated SO_XWhen the amount is larger than the boundary amount, the processing start amount is calculated as the calculated accumulated SO._XWhen the amount is small, the processing start amount is set to the calculated accumulated SO._XThis means that it is set to an amount unrelated to the amount.
[0098]
On the other hand, when the sulfur concentration of the fuel having the highest sulfur concentration among the fuels available on the market is expressed by CSM, NO is emitted when the fuel is consumed by the first set amount qf1._XSO that can be stored in the absorbent 23_XThe quantity is at most qf1 · CSM, SO_XActual storage SO when it is determined that processing should be performed_XThe quantity does not exceed qf1 · CSM. That is, NO is assumed when the fuel is assumed to be the fuel with the highest sulfur concentration among the fuels available on the market._XWhen the maximum sulfur amount QSM that can be stored in the absorbent is stored, the process start time amount QSS should not exceed the maximum amount QSM that can be stored.
[0099]
Therefore, in the first embodiment according to the present invention, when the processing start time amount QSS exceeds the storable maximum amount QSM, the processing start time amount QSS is set to the storable maximum amount QSM. In this way, the processing start time amount QSS is actually stored in the SO._XIt is possible to prevent a large deviation from the quantity estimated value QSE.
[0100]
By the way, as mentioned above, the primary expression m (see FIG. 8) is NO._XPurification rate EFF or spill NO_XIt is determined based on the concentration CNO. Next, a method for determining the primary expression m will be described schematically first.
[0101]
As shown by the arrow RN in FIG._XWhen processing is not being performed, the time interval is NO._XThe process is repeated and NO_XNO when processing is not performed_XThe air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is lean. In the first embodiment according to the present invention, NO NO is thus obtained._XNO during a period when the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is lean._XThe concentration detection period DP is set, and this NO_XNO flowing into concentration detection period DP_XConcentration CNI and spill NO_XThe concentration CNO is detected.
[0102]
NO_XConcentration detection period DP is NO_XAs long as the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is lean, it may be determined in any way, but in the first embodiment according to the present invention, as shown in FIG._XNO accumulated in absorbent 23_XThe amount of time from when the estimated value QNE exceeds the lower limit QNm until it exceeds the upper limit QNM is NO_XThe concentration detection period DP is set.
[0103]
NO_XSpilled NO when processing is done_XAlthough the concentration CNO varies greatly, NO_XIf a certain amount of time has passed since the process was performed,_XThe concentration CNO is stable. On the other hand, NO_XAccumulated NO as the elapsed time from the processing increases_XThe quantity estimate QNE gradually increases, so the accumulated NO_XQuantity estimate QNE is NO_XIt represents the elapsed time since the process was performed. Therefore, in the first embodiment according to the present invention, the accumulated NO_XNO based on the quantity estimate QNE_XThe density detection period DP is set.
[0104]
NO as shown by arrow Y1 in FIG._XWhen the concentration detection period DP is started, inflow NO_XConcentration CNI and spill NO_XThe concentration CNO is detected almost at the same time. Specifically, inflow NO_XThe concentration CNI is calculated using the map of FIG._XConcentration CNO is NO_XDetection is based on the output of the sensor 51. Next, these detected inflow NOs_XConcentration CNI and spill NO_XCorrection values CNIC and CNOC for the concentration CNO are calculated.
[0105]
Such detection of CNI and CNO and calculation of CNIC and CNOC are repeatedly performed with a slight time interval Δt. Next, as shown by arrow Y2 in FIG._XWhen the concentration detection period DP is completed, NO_XInflow NO in concentration detection period DP_XAverage value CNIAV and outflow NO of concentration correction value CNIC_XAn average value CNOAV of the density correction value CNOC is calculated, and then NO based on these average values CIAV, CNOAV_XA purification rate EFF is calculated. In this case, inflow NO in a certain period_XConcentration and spill NO_XNO using the average concentration_XSince the purification rate EFF is calculated, NO_XThe purification rate EFF can be obtained with high accuracy. Then NO_XA correction value EFFC for the purification rate EFF is calculated.
[0106]
That is, as can be seen from FIG._XEvery time a lean operation is performed after the processing is performed, NO_XPurification rate EFF or NO_XA purification rate correction value EFFC is calculated.
[0107]
Then the previous NO_XAfter the purification rate average value EFFAV is calculated (see arrow Z1 in FIG. 11), when the fuel consumption amount QF increases by the second set amount qf2 (see arrow Z2 in FIG. 11), the fuel consumption amount QF becomes the second value. NO calculated during the period when the set amount qf2 is consumed_XNO which is the average value of the purification rate correction value EFFC_XA purification rate average value EFFAV is calculated.
[0108]
That is, as shown in FIG. 14, every time the fuel is consumed by the second set amount qf2, NO_XA purification rate average value EFFAV is calculated. The second set amount qf2 is stored even if the fuel is consumed by qf2._XLittle increase in volume, so NO_XThe amount (<< qf1) at which the purification rate hardly changes is set.
[0109]
Then the previous SO_XWhen the processing is completed (see arrow X1 in FIG. 14), the fuel is consumed by the first set amount qf1 (see arrow X2 in FIG. 14)._XA primary equation m representing the relationship between the purification rate average value EFFAV and the fuel consumption QF is determined._XThe purification rate estimated value EFFE is calculated using the linear expression m. Then this NO_XAccumulated SO using the map of FIG. 9 from the estimated purification rate EFFE_XA quantity estimate QSE is calculated.
[0110]
Therefore, generally speaking, NO_XInflow NO when the air-fuel ratio of the exhaust gas flowing into the absorbent 23 is lean_XVolume and spill NO_XThe amount of inflow NO._XVolume and spill NO_XNO based on quantity_XAbsorbent NO_XThe purification rate is obtained several times at intervals of each other, and every time the average value calculation time arrives, NO determined within the period from the previous average value calculation time to the current average value calculation time_XThe average value EFFAV of the purification rate is calculated, and these NO_XIt means that the relational expression m representing the relation between the purification rate average value EFFAV and the fuel consumption is obtained. In addition, it is determined that the next average value calculation time has come when the fuel is consumed by a predetermined second set amount qf2 from the previous average value calculation time.
[0111]
Next, the first embodiment of the present invention will be described in detail with reference to FIGS. 15 to 35 with reference to FIGS. 6 and 11 to 14.
[0112]
FIG. 15 shows the SO according to the first embodiment of the present invention._XA control routine is shown. This routine is executed by interruption every predetermined set time Δt.
[0113]
Referring to FIG. 15, first, at step 200, SO is selected._XIt is determined whether or not the processing flag XSOX is set. This SO_XProcessing flag XSOX is SO_XIt is set when processing is to be executed (XSOX = 1), otherwise it is reset (XSOX = 0). SO_XWhen the processing flag XSOX is reset, the routine proceeds to step 201 where a fuel consumption difference ΔQF (= QF−QFP1) is calculated. Where QF is the current fuel consumption and QFP1 is the previous SO_XThe fuel consumption at the time when the processing is completed is shown. In the following step 202, it is determined whether or not the fuel consumption difference ΔQF is equal to or larger than the first set amount qf1. When ΔQF <qf1, that is, the previous SO_XIf the fuel is not consumed by the first set amount qf1 after the processing is completed, the routine proceeds to step 203, where a calculation routine for a coefficient kmi (i = 1, 2, 3, 4) for specifying the primary expression m is executed. Is done. The calculation routine for the coefficient kmi is shown in FIG.
[0114]
Next, when ΔQF ≧ qf1, the routine proceeds from step 202 to step 204, where a routine for setting the processing start time amount QSS is executed. A routine for setting the processing start time amount QSS is shown in FIGS.
[0115]
In the following step 205, the accumulated SO_XThe quantity representative value QS is set to the processing start time quantity QSS set in step 204. In the next step 206, SO_XA processing flag XSOX is set (XSOX = 1).
[0116]
SO_XWhen the processing flag XSOX is set, the routine proceeds from step 200 to step 207, where SO_XProcessing is performed. Specifically, NO_XWhile the temperature of the absorbent 23 is maintained above the sulfate decomposition temperature, NO_XThe reducing agent is temporarily supplied from the reducing agent supply valve 31 so that the average air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 becomes slightly rich.
[0117]
That is, as shown in FIG._XWhen the fuel consumption amount QF increases by the first set amount qf1 after the processing is completed, the next SO_XProcessing begins.
[0118]
In the subsequent step 208, NO from the previous processing cycle to the current processing cycle is determined._XAccumulated SO of absorbent 23_XA quantity reduction amount qsr is calculated. This SO_XFor example, as shown in FIG. 16, the reduction amount qsr is obtained in advance as a function of the engine operating state, for example, the required load L and the engine speed N, and is stored in the ROM 42 in advance in the form of a map shown in FIG. Yes. In step 208, the SO shown in FIG._XA decrease amount qsr is calculated. In the following step 209, the accumulated SO_XQuantity representative value QS is SO_XIt is decreased by the decrease amount qsr (QS = QS−qsr). In the following step 210, the accumulated SO_XIt is determined whether or not the quantity representative value QS is equal to or lower than the lower limit value QSL. When QS> QSL, the processing cycle is terminated and SO_XProcessing continues. Next, when QS ≦ QSL, the routine proceeds from step 210 to step 211, where SO_XThe processing flag XSOX is reset (XSOX = 0). Next, the routine proceeds to step 212, where the fuel consumption amount QF at this time is stored as QFP1.
[0119]
That is, as shown in FIG._XWhen the quantity representative value QS is less than or equal to the lower limit value QSL, SO_XProcessing is completed.
[0120]
Next, the coefficient kmi calculation routine executed in step 203 (FIG. 15) will be described with reference to FIG. Referring to FIG. 17, first, in step 220, it is determined whether or not the difference between the current fuel consumption amount QF and the fuel consumption amount QFP2 at the previous average value calculation timing is equal to or greater than a second set amount qf2. When QF-QFP2 <qf2, that is, the previous NO_XWhen the fuel is not consumed by the second set amount qf2 after the purification rate average value EFFAV is calculated, the routine then proceeds to step 221 where NO_XA purification rate calculation routine is executed. This NO_XThe purification rate calculation routine is shown in FIGS.
[0121]
Referring to FIGS. 18 to 20, first, at step 230, it is judged if the detection period flag XDP is set. This detection period flag XDP is NO_XIt is set during the density detection period DP (XDP = 1), and is reset otherwise (XDP = 0). When the detection period flag XDP is reset, that is, currently NO_XIf it is not during the concentration detection period DP, the process proceeds to step 231 and NO._XAbsorbent 23 and NO_XIt is determined whether or not the sensor 51 is activated. For example, when the temperature of the exhaust gas detected by the exhaust temperature sensor 52 is a certain value or more, NO_XAbsorbent 23 and NO_XIt can be determined that the sensor 51 is activated. NO_XAbsorbent 23 and NO_XWhen it is determined that the sensor 51 is activated, the routine proceeds to step 232 where NO_XIt is determined whether or not the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is lean. For example, when no reducing agent is supplied from the reducing agent supply valve 31, NO_XIt can be determined that the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is lean. NO_XWhen it is determined that the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is lean, the routine proceeds to step 233, where the accumulated NO_XIt is determined whether or not the quantity estimated value QNE is larger than the lower limit value QNm.
[0122]
Accumulated NO_XThe amount estimated value QNE is, for example, the accumulated NO shown in FIG._XIt can be calculated by a calculation routine of the quantity estimated value QNE. This calculation routine is executed by interruption every predetermined time.
[0123]
Referring to FIG. 21, first, at step 260, it is judged if the reducing agent supply action from the reducing agent supply valve 31 is stopped. When the reducing agent supply operation is stopped, the routine proceeds to step 261, where NO from the previous processing cycle to the current processing cycle is determined._XNO accumulated in absorbent 23_XA quantity increase amount qna is calculated. This NO_XFor example, as shown in FIG. 22, the increase amount qna is obtained in advance as a function of the engine operating state, for example, the required load L and the engine speed N, and is stored in the ROM 42 in advance in the form of a map shown in FIG. Yes. In step 261, NO from the map of FIG._XAn increase amount qna is calculated. In the following step 262, the accumulated NO_XQuantity estimate QNE is NO_XIt is increased by the increase amount qna (QNE = QNE + qna).
[0124]
On the other hand, when the reducing agent supply operation is being performed, the routine proceeds to step 263, where the accumulated NO._XThe quantity estimated value QNE is cleared. That is, as shown in FIG. 12, when the reducing agent supply action is started, the accumulated NO_XWhen the estimated amount QNE is cleared and the reducing agent supply action is stopped, the accumulated NO_XThe quantity estimate QNE begins to increase.
[0125]
Referring to FIG. 19 again, when QNE> QNm at step 233, the routine proceeds to step 234, where the detection period flag XDP is set (XDP = 1).
[0126]
In contrast, NO_XAbsorbent 23 or NO_XSensor 51 is not activated or NO_XWhen the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is not lean or when QNE ≦ QNm, the processing cycle is terminated.
[0127]
When the detection period flag XDP is set, the process proceeds from step 230 to step 235 and the accumulation NO_XIt is determined whether or not the quantity estimated value QNE is larger than the upper limit value QNM. When QNE ≦ QNM, the routine then proceeds to step 236 where the inflow NO_XA routine for calculating the density correction value CNIC is executed. In this CNIC calculation routine, as will be described later with reference to FIG._XConcentration CNI is detected and inflow NO_XA density correction value CNIC is calculated. In the following step 237, the outflow NO_XA routine for calculating the density correction value CNOC is executed. In this CNOC calculation routine, as will be described later with reference to FIG._XConcentration CNO is detected, outflow NO_XA density correction value CNOC is calculated.
[0128]
That is, since this routine is executed every time Δt, as shown in FIG._XConcentration CNI and spill NO_XConcentration CNO is detected, and inflow NO at every time Δt_XConcentration correction value CNIC and outflow NO_XA density correction value CNOC is calculated.
[0129]
In the following step 238, the outflow NO_XIt is determined whether or not the density correction value CNOC is smaller than the allowable minimum value CNOL. When CNOC <CNOL, the routine then proceeds to step 239, where the outflow NO_XA small number counter nNL indicating the number of times when the density correction value CNOC is smaller than the allowable minimum value CNOL is incremented by 1 (nNL = nNL + 1). Next, the routine proceeds to step 240. On the other hand, when CNOC ≧ CNOL, the routine jumps from step 238 to step 240.
[0130]
In step 240, inflow NO_XConcentration CNI and spill NO_XA detection number counter nC indicating the number of detections of the density CNO or the correction values CNIC and CNOC is incremented by 1 (nC = nC + 1). In the following step 241, inflow NO_XIntegrated value SCNI of concentration correction value CNIC and outflow NO_XThe integrated value SCNO of the density correction value CNOC is calculated (SCNI = SCNI + CNIO, SCNO = SCNO + CNOC).
[0131]
Then accumulated NO_XWhen the amount estimated value QNE exceeds the upper limit value QNM, the process proceeds from step 235 to step 242 and the detection period flag XDP is reset (XDP = 0).
[0132]
In the subsequent step 243, it is determined whether or not the detection number counter nC is greater than or equal to a predetermined lower limit value nDP. When nC ≧ nDP, the routine proceeds to step 244 where inflow NO_XConcentration average value CNIAV and outflow NO_XThe concentration average value CNOAV is calculated (CNIAV = SCNI / nC, CNOAV = SCNO / nC). In the following step 245, NO_XA routine for calculating the purification rate correction value EFFC is executed. NO_XIn the routine for calculating the purification rate correction value EFFC, as will be described later with reference to FIG._XPurification rate EFF and NO_XA purification rate correction value EFFC is calculated.
[0133]
That is, as shown in FIG. 11 to FIG._XIf the quantity estimate value QNE exceeds the upper limit value QNM, NO_XThe concentration detection period DP is completed._XConcentration average value CNIAV and outflow NO_XConcentration average values CNOAV are calculated, respectively, and NO_XPurification rate EFF and NO_XA purification rate correction value EFFC is calculated.
[0134]
In the following step 246, it is determined whether or not it is currently in steady operation. For example, based on the rate of change of the engine speed N, it can be determined whether or not the engine is in steady operation. When it is determined that the vehicle is in steady operation, the process proceeds to step 247, where NO_XAn integrated value SST of the purification rate correction value EFFC calculated during the steady operation is calculated (SST = SST + EFFC). In the following step 248, NO during steady operation is obtained._XA counter nST representing the number of times of calculation of the purification rate correction value EFFC is incremented by 1 (nST = nST + 1). Next, the routine proceeds to step 251. On the other hand, when it is determined that the current state is the transient operation, the routine proceeds to step 249, where NO_XAn integrated value SNST of the purification rate correction value EFFC calculated during the transient operation is calculated (SRST = SNTP + EFFC). In the following step 250, NO during the transient operation_XA counter nNST indicating the number of calculations of the purification rate correction value EFFC is incremented by 1 (nNST = nNST + 1). Next, the routine proceeds to step 251. In step 251, NO_XA purification rate calculation number counter nEFF indicating the number of times of calculation of the purification rate EFF or the correction value EFFC is incremented by 1 (nEFF = nEFF + 1). Next, the routine proceeds to step 252.
[0135]
In step 252, the detection number counter nC and the integrated values SCNI and SCNO are cleared.
[0136]
On the other hand, when nC <nDP, the routine jumps from step 243 to step 252. That is, one NO_XInflow NO during concentration detection period DP_XConcentration CNI and spill NO_XWhen the number of times the concentration CNO is detected is less than the lower limit nDP, NO_XPurification rate EFF and NO_XThe purification rate correction value EFFC is not calculated. This is because inflow NO_XConcentration CNI and spill NO_XWhen the number of detections of concentration CNO decreases, NO_XPurification rate EFF and NO_XThis is because it becomes difficult to accurately calculate the purification rate correction value EFFC. Here, the counter nC is NO_XConsidering that it represents the duration of the concentration detection period DP, NO_XWhen the duration of the concentration detection period DP is shorter than the set time, NO_XPurification rate EFF and NO_XIt can also be said that calculation of the purification rate correction value EFFC is prohibited. NO_XThe duration of the concentration detection period DP is the accumulated NO_XSince the amount estimated value QNE may fluctuate depending on the speed of increase, that is, the engine operating state, the detection number counter nC may be larger or smaller than the lower limit value nDP.
[0137]
Referring to FIG. 17 again, when QF−QFP2 ≧ qf2, the routine proceeds from step 220 to step 222, where it is determined whether or not the purification rate calculation number counter nEFF is greater than or equal to a predetermined lower limit value nL2. When nEFF ≧ nL2, the routine then proceeds to step 223, where NONO_XA purification rate average value EFFAV is calculated.
[0138]
[Expression 1]

[0139]
Here, wST is the NO calculated during engine steady operation._XThe weighting coefficient of the purification rate EFF or the correction value EFFC, wNST is the NO calculated during engine transient operation_XThe weighting coefficient of the purification rate EFF or the correction value EFFC is shown respectively. That is, in the first embodiment according to the present invention, NO_XWhen the purification rate EFF or the correction value EFFC is calculated, a weighted average value EFFAV weighted according to whether the engine steady operation or engine transient operation is performed is calculated. In this case, the NO calculated during steady engine operation_XConsidering that the purification rate EFF or the correction value EFFC is more reliable, the weighting coefficient wST is larger than the weighting coefficient wNST.
[0140]
In the following step 224, NO_XAn average value calculation number counter nAV indicating the calculation number of the purification rate average value EFFAV is incremented by 1 (nAV = nAV + 1). In the subsequent step 225, the coefficient kmi (i = 1, 2, 3, 4) is calculated based on the following equations, respectively.
[0141]
km1 = km1 + QF
km2 = km2 + QF²
km3 = km3 + QF ・ EFFAV
km4 = km4 + EFFAV
In the first embodiment according to the present invention, NO_XThe coefficient kmi is updated every time the purification rate average value EFFAV is calculated, and therefore NO_XThis means that the primary expression m is determined every time the purification rate average value EFFAV is calculated.
[0142]
In the following step 226, a fuel supply determination routine is executed. In this fuel supply determination routine, as will be described later with reference to FIG. 30, it is determined whether or not fuel supply has been performed, or more precisely, whether or not fuels having different sulfur concentrations have been supplied. In the following step 227, the current fuel consumption amount QF is stored as QFP2, and the integrated values SST, SNTP, counters nST, nNST, nEFF are cleared.
[0143]
On the other hand, when nEFF <nL2, the routine jumps from step 222 to step 227. That is, the previous NO_XThis time NO from the purification rate average calculation time_XNO within the period up to the average purification rate calculation time_XWhen the number of times the purification rate EFF or the correction value EFFC is calculated is less than the lower limit nL2, NO in this period_XThe purification rate average value EFFAV is not calculated. NO per unit time_XThe number of times the purification rate EFF or the correction value EFFC is calculated is NO_XSince it is determined according to the time interval tN at which the process is performed, the purification rate calculation number counter nEFF may be larger or smaller than the lower limit nL2.
[0144]
Next, the inflow NO executed in step 236 (FIG. 20)_XA routine for calculating the density correction value CNIC will be described with reference to FIG. Referring to FIG. 23, first, at step 270, based on the current engine operating state, that is, the required load L and the engine speed N, an inflow NO._XThe concentration CNI is calculated. In the following step 271, the delay time DLY is calculated. In the following step 272, the inflow NO is calculated based on the following equation (1)._XA density correction value CNIC is calculated.
[0145]
[Expression 2]

[0146]
In the following step 273, the inflow NO calculated in step 271 is obtained._XThe concentration CNI is stored as CNIP.
[0147]
As described above with reference to FIG._XConcentration CNI and spill NO_XThe concentration CNO is detected almost at the same time. However, the exhaust gas part is inflow NO for that_XNO after concentration CNI is calculated_XConsidering that only the delay time DLY is required to reach the sensor 51, the outflow NO detected at a certain time point_XConcentration CNO and inflow NO detected before a certain delay time DLY from this point in time_XNO based on concentration CNI_XThe purification rate EFF should be calculated.
[0148]
Therefore, in the first embodiment according to the present invention, this delay time DLY is obtained (step 271), and the inflow NO before the delay time DLY from the present time is obtained._XThe concentration CNI is calculated. This calculation result is the inflow NO in the first embodiment according to the present invention._XThis is the density correction value CNIC.
[0149]
The delay time DLY is obtained in advance as a function of the engine operating state, for example, the intake air amount Ga, and is stored in advance in the ROM 42 in the form of a map shown in FIG. In step 271, the delay time DLY is calculated from the map of FIG.
[0150]
On the other hand, as shown in FIG. 25, the inflow NO detected last time, that is, at time t (j−1)._XConcentration is CNI (j-1), inflow NO detected this time, that is, at time t (j)_XWhen the concentration is CNI (j), time t and inflow NO_XThe relationship with the concentration CNI is expressed by the following equation (2).
[0151]
[Equation 3]

[0152]
J is inflow NO_XThis represents the number of detections of the density CNI (j = 1, 2,. In this formula (2), t-DLY is substituted for time t, and CNI (j-1) is detected as the inflow NO detected last time._XWhen rewritten to the concentration CNIP, the above-described equation (1) is obtained.
[0153]
The inflow NO detected this time_XConcentration CNI and outflow NO ahead by delay time DLY_XNO based on concentration CNO_XThe purification rate may be calculated. That is, in this case, the outflow NO_XThe concentration CNO is corrected based on the delay time.
[0154]
Next, the outflow NO executed in step 237 (FIG. 20)_XA routine for calculating the density correction value CNOC will be described with reference to FIG. Referring to FIG. 26, first, at step 280, NO_XOutflow NO detected by sensor 51_XThe concentration CNO is read. In the following step 281, the outflow NO is calculated based on the following equation:_XA density correction value CNOC is calculated.
[0155]
[Expression 4]

[0156]
Here, tN and ktN are the above-mentioned NO._XThe processing time interval and the correction amount.
[0157]
As described with reference to FIGS. 2 to 5, in the embodiment according to the present invention, NO is used._XOutflow NO when the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is kept lean_XConcentration CNO does not exceed the allowable maximum concentration, and NO_XThe basic time interval tNB is corrected by the correction amount ktN (tN = tNB + ktN) so that the fuel consumption rate by the process does not exceed the allowable maximum value. Here, considering that the correction amount ktN varies depending on the engine operation state, (tNB + ktN) / tNB (= tN / tNB) represents the amount of deviation of the engine operation state from the steady operation state, and is detected. Outflow NO_XThe result of dividing the concentration CNO by this (tNB + ktN) / tNB is the outflow NO in the standard operation state, that is, steady operation._XIt represents the value converted to concentration.
[0158]
Therefore, in the first embodiment according to the present invention, the outflow NO detected in step 280_XBy dividing the concentration CNO by (tNB + ktN) / tNB, the outflow NO in steady operation_XThe value converted into the concentration is calculated. This calculation result is the outflow NO in the first embodiment according to the present invention._XThis is the density correction value CNOC. In this way, outflow NO_XConcentration is not affected by engine operating conditions, so NO_XThe purification rate or the primary equation m is not affected by the engine operating condition.
[0159]
Therefore, generally speaking, NO for steady engine operation_XA basic time interval tNB for processing is set, the basic time interval tNB is corrected with the correction amount ktN based on the engine operating state, and the outflow NO is based on the correction amount ktN._XThis means that the amount CNO is corrected.
[0160]
Next, the NO executed in step 245 (FIG. 18)_XA routine for calculating the purification rate correction value EFFC will be described with reference to FIG. Referring to FIG. 27, first, at step 290, NO_XThe purification rate EFF is calculated based on the following equation.
[0161]
EFF = (CNIAV−CNOAV) / CNIAV
In the subsequent step 291, the first correction coefficient kE 1 is calculated from the map of FIG. 28A, in the subsequent step 292, the second correction coefficient kE 2 is calculated from the map of FIG. A third correction coefficient kE3 is calculated from the map of (C). In the following step 294, NO is calculated based on the following equation._XA purification rate correction value EFFC is calculated.
[0162]
EFFC = EFF / kE1 / kE2 / kE3
The first and second correction coefficients kE1 and kE2 are the NO calculated in step 290._XThe purification rate EFF is NO_XNO when the state of the absorbent 23 is a predetermined reference state_XThis is for conversion into a purification rate. This reference state may be determined in any way, but in the first embodiment according to the present invention, NO_XThe temperature TC of the absorbent 23 is the reference temperature TCR, and NO_XThe reference state is when the intake air amount Ga representing the space velocity of the exhaust gas in the absorbent 23 is the reference amount GaR.
[0163]
As shown in FIG. 28A, the first correction coefficient kE1 is NO_XNO when the temperature TC of the absorbent 23 is low_XNO decreases as the temperature TC of the absorbent 23 increases, and NO_XNO when the temperature TC of the absorbent 23 is high_XNO increases as the temperature TC of the absorbent 23 increases._XIt becomes 1.0 when the temperature TC of the absorbent 23 is the reference temperature TCR.
[0164]
Further, as shown in FIG. 28B, the second correction coefficient kE2 is NO_XIt increases as the intake air amount Ga representing the space velocity of the exhaust gas in the absorbent 23 increases, and becomes 1.0 when the intake air amount Ga is the reference amount GaR.
[0165]
In this way, NO_XNO purification rate_XUnaffected by the state of the absorbent 23, so the primary m is NO_XIt is not affected by the state of the absorbent 23.
[0166]
On the other hand, the third correction coefficient kE3 is NO._XNO to purification rate_XThis is for removing the influence of the degree of thermal deterioration of the absorbent 23. That is, the third correction coefficient kE3 is NO._XThe value is 1.0 when the thermal degradation counter nTCH representing the degree of thermal degradation of the absorbent 23 is zero, and increases as the thermal degradation counter nTCH increases.
[0167]
These correction coefficients kE1, kE2, and kE3 are obtained in advance by experiments, and are stored in advance in the ROM 42 in the form of maps shown in FIGS. 28 (A), (B), and (C).
[0168]
The thermal deterioration counter nTCH can be calculated by, for example, a calculation routine for the thermal deterioration counter nTCH shown in FIG. This routine is executed by interruption every predetermined time.
[0169]
Referring to FIG. 29, first, at step 300, NO_XIt is determined whether or not the temperature TC of the absorbent 23 is higher than a preset temperature TCH. When TC ≦ TCH, the processing cycle is terminated. When TC> TCH, the routine proceeds to step 301 where the thermal deterioration counter nTCH is incremented by 1 (nTCH = nTCH + 1).
[0170]
Next, the refueling determination routine executed in step 226 (FIG. 17) will be described with reference to FIG. Referring to FIG. 30, first, at step 310, it is judged if the refueling flag XFC has been reset. This refueling flag is set when it is determined that refueling, more precisely, fuel with different sulfur concentrations has been performed (XFC = 1), and otherwise it is reset (XFC = 0). When the refueling flag XFC is reset, the routine proceeds to step 311 where NO_XA deviation R with respect to the linear expression m of the purification rate average value EFFAV is calculated. This deviation R can be expressed by a correlation coefficient, for example.
[0171]
In the following step 312, it is determined whether or not the difference between the deviation R in the current processing cycle and the deviation RP in the previous processing cycle is greater than a predetermined threshold value R1. When R−RP ≦ R1, the process proceeds to step 313, and the deviation R in the current processing cycle is stored as RP. On the other hand, when R-RP> R1, the routine proceeds to step 314 where the refueling flag XFC is set (XFC = 1). That is, when R-RP> R1, it is determined that refueling has been performed.
[0172]
As described above with reference to FIG. 8, the fuel consumption QF and NO are determined unless refueling is performed._XThe primary expression m representing the relationship with the purification rate average value EFFAV does not change. This means that before and after refueling, fuel consumption QF and NO_XThis means that the primary expression m representing the relationship with the purification rate average value EFFAV is different. That is, as shown in FIG. 31, before refueling, the fuel consumption QF and NO_XThe primary expression representing the relationship with the purification rate average value EFFAV is m1, and becomes m2 after refueling.
[0173]
Thus, the fuel consumption QF and NO that should be originally expressed by the two linear expressions m1 and m2_XIf the relationship with the purification rate average value EFFAV is expressed by one linear expression m, the deviation R increases. Therefore, in the first embodiment according to the present invention, it is determined that the refueling has been performed when the deviation R becomes larger than the threshold value R1.
[0174]
In the subsequent step 315, the coefficient kmi and the average value calculation number counter nAV are cleared. That is, when it is determined that refueling has been performed, NO before refueling_XDetermination of the primary expression m based on the purification rate EFF or the average value EFFAV is prohibited.
[0175]
In the following step 316, it is determined whether or not the high sulfur concentration flag XFSH is set. The high sulfur concentration flag XFSH is set when it is determined that the fuel is a high sulfur concentration fuel (XFSH = 1), and is reset when it is determined that the fuel is a low sulfur concentration fuel or a standard fuel (XFSH = 0), and is set or reset in the processing start time amount QSS setting routine shown in FIGS.
[0176]
When the high sulfur concentration flag XFSH is set, the routine proceeds to step 317, where the low sulfur concentration determination flag XFCL is set (XFCL = 1). The low sulfur concentration determination flag XFCL is set when it is determined that the fuel has been switched to the low sulfur concentration fuel or the standard fuel (XFCL = 1), and otherwise it is reset (XFCL = 0). That is, when the fuel is a high-sulfur concentration fuel (XFSH = 1), refueling means that the fuel has been switched to the low-sulfur concentration fuel or the standard fuel. At this time, the low-sulfur concentration determination Flag XFCL is set.
[0177]
On the other hand, when the high sulfur concentration flag XFSH is reset, the routine proceeds to step 318, where the high sulfur concentration determination flag XFCH is set (XFCH = 1). The high sulfur concentration determination flag XFCH is set when it is determined that the fuel has been switched to the high sulfur concentration fuel (XFCH = 1), and otherwise it is reset (XFCH = 0). That is, when the fuel is a low-sulfur concentration fuel or a standard fuel (XFSH = 0), it means that the fuel has been switched to a high-sulfur concentration fuel. Flag XFCH is set.
[0178]
On the other hand, when the refueling flag XFC is set in step 310 (XFC = 1), the processing cycle is completed. As will be described later, the refueling flag XFC is reset in the processing start time amount QSS setting routine shown in FIGS. Therefore, once the refueling flag XFC is set, the processing start time amount QSS is set or SOSO is set._XThe refueling determination is not performed until the processing is performed, the processing start time amount QSS is set or SO_XWhen the process is performed, the fuel supply determination is performed again.
[0179]
Next, the processing start time amount QSS setting routine executed in step 204 (FIG. 15) will be described with reference to FIGS. Referring to FIGS. 32 to 34, first, in step 320, the maximum storable amount QSM is calculated (QSM = ΔQF1 · CSM). In the following step 321, it is determined whether or not the high sulfur concentration determination flag XFCH has been reset. When the high sulfur concentration determination flag XFCH is reset, the routine proceeds to step 322, where it is determined whether or not the low sulfur concentration determination flag XFCL is reset. When the low sulfur concentration determination flag XFCL is reset, the process proceeds to step 323. The high sulfur concentration determination flag XFCH or the low sulfur concentration determination flag XFCL is set when it is determined that refueling has been performed (see FIG. 30). Therefore, when it is not determined that refueling has been performed, step 323 is performed. Will proceed to.
[0180]
In step 323, NO_XIt is determined whether or not an average value calculation number counter nAV (see FIG. 17) representing the number of calculation of the purification rate average value EFFAV is equal to or greater than a predetermined lower limit value nL1. If nAV ≧ nL1, then the routine proceeds to step 324, where SO_XNO when processing should start_XThe purification rate estimated value EFFE is calculated based on the following equation (3).
[0181]
[Equation 5]

[0182]
In the following step 325, NO_XIt is determined whether or not the purification rate estimated value EFFE is higher than the boundary purification rate EFFB. When EFFE> EFFB, the routine proceeds to step 326, where the accumulated SO is calculated from the map of FIG._XA quantity estimate QSE is calculated. In the following step 327, the processing start time amount QSS is stored in the accumulated SO._XThe quantity estimated value QSE is set. In the following step 328, the high sulfur concentration flag XFSH is set (XFSH = 1). That is, in this case, it is determined that the fuel is a high-sulfur concentration fuel. Next, the routine proceeds to step 335.
[0183]
On the other hand, when EFFE ≦ EFFB, the routine proceeds from step 325 to step 329, where the processing start time amount QSS is set using the reference concentration CSR (QSS = ΔQF1 · CSR). That is, the processing start time amount QSS is NO._XPurification rate average value EFFAV or accumulated SO_XIt is set to an amount unrelated to the amount estimated value QSE. In the subsequent step 330, the high sulfur concentration flag XFSH is reset (XFSH = 0). That is, in this case, it is determined that the fuel is a low-sulfur concentration fuel or a standard fuel. Next, the routine proceeds to step 335.
[0184]
On the other hand, when nAV <nL1, the routine proceeds from step 323 to step 331, where the outflow NO._XIt is determined whether or not a small quantity counter nNL (see FIG. 20) indicating the number of times when the density correction value CNOC is smaller than the allowable minimum value CNOL is smaller than a predetermined upper limit value nU. When nNL <nU, the routine proceeds to step 332, where the processing start time amount QSS is set to the maximum storable amount QSM. Next, the routine proceeds to step 328, where the high sulfur concentration flag XFSH is set.
[0185]
On the other hand, when nNL ≧ nU, the routine proceeds to step 329 where the processing start time amount QSS is set (QSS = ΔQF1 · CSR), and at the next step 330, the high sulfur concentration flag XFSH is reset (XFSH = 0).
[0186]
In this way, the NO value represented by the average value calculation number counter nAV_XWhen the number of calculations of the purification rate average value EFFAV is large, NO_XPurification rate estimate EFFE and accumulated SO_XThe processing start time amount QSS is set based on the amount estimated value QSE, and NO_XWhen the number of calculations of the purification rate average value EFFAV is small, the processing start time amount QSS is set assuming that the sulfur concentration in the fuel is CSR or CSM.
[0187]
Therefore, generally speaking, the detected spill NO_XWhen the number of times nNL when the amount CNO or the correction value CNOc is smaller than the predetermined allowable minimum amount CNOL exceeds the upper limit value nU, the processing start time amount QSS is calculated as the accumulated SO_XThis means that the amount is set to an amount unrelated to the amount QSE.
[0188]
In addition, it is determined that the fuel is high-sulfur concentration fuel when the small-quantity counter nNL is small, and it is determined that the fuel is low-sulfur concentration fuel or standard fuel when the small-quantity counter nNL is large.
[0189]
On the other hand, when the high sulfur concentration determination flag XFCH is set, that is, when it is determined that the high sulfur concentration fuel has been refueled, the routine proceeds from step 321 to step 333, and whether or not the average value calculation number counter nAV is greater than or equal to the lower limit value nL1. Is determined. When nAV ≧ nL1, the routine proceeds to step 334, where SO_XNO when processing should start_XThe purification rate estimated value EFFE is calculated based on the above equation (3). Subsequently, it progresses to step 326,327,328 sequentially. On the other hand, when nAV <nL1, the process proceeds to steps 332 and 328 sequentially.
[0190]
On the other hand, when the low sulfur concentration determination flag XFCL is set, that is, when it is determined that the low sulfur concentration fuel or the standard fuel is supplied, the routine proceeds from step 322 to steps 329 and 330 sequentially.
[0191]
In step 335, it is determined whether or not the set process start time amount QSS is larger than the maximum storable amount QSM. When QSM> QSM, the process proceeds to step 336. After the processing start time amount QSS is set to the maximum storable amount QSM, the process proceeds to step 337. On the other hand, when QSS ≦ QSM, the routine jumps to step 337.
[0192]
In step 337, the coefficient kmi (i = 1, 2, 3, 4), the average value calculation number counter nAV, and the small amount number counter nNL are cleared, and the fueling flag XFC, the high sulfur concentration determination flag XFCH, and the low sulfur concentration determination The flag XFCL is reset.
[0193]
Thus, in the first embodiment according to the present invention, as shown in FIG. 35, every time the fuel consumption amount QF increases by the second set amount qf2, NO_XA purification rate average value EFFAV is calculated. In addition, the NO expressed by the purification rate calculation number counter nEFF_XWhen the number of calculations of the purification rate EFF is small, as shown by the arrow W in FIG._XThe purification rate average value EFFAV is not calculated. On top of that, NO_XThe primary expression m representing the relationship between the average purification rate EFFAV and the fuel consumption QF is NO._XIt is determined based on the average purification rate EFFAV. Then the previous SO_XAfter the processing is completed (see arrow X1 in FIG. 35), if the fuel consumption amount QF increases by the first set amount qf1, NO at this time_XThe purification rate estimated value EFFE is calculated using the linear expression m.
[0194]
Next, a second embodiment according to the present invention will be described.
[0195]
As shown in FIG. 36, NO_XWhen processing is done, NO_XOutflow NO detected by sensor 51_XThe concentration CNO first decreases sharply to reach the downward peak PKm1, then increases to reach the upward peak PKM, and then decreases again to reach the downward peak PKm2, and then gradually increases.
[0196]
Outflow NO_XThe reason why the concentration CNO takes such a behavior is not necessarily clear, but it is considered to be due to the following reason. That is, NO_XWhen the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is switched to rich, as described above, NO_XNO stored in absorbent 23_XIs NO_XFor example, NO from the absorbent 23₂Is released in the form of NO_XImmediately after the treatment is started, this released NO₂Is immediately reduced, thus generating a downward peak PKm1. However, over time, NO_XNO reduction reaction is difficult to occur, NO_XNO released from absorbent 23 but not reduced_XThus increasing the upward peak PKM. If more time passes, NO_XNO released from the absorbent 23_XThe amount of NO decreases gradually, however NO_XNO flowing into the absorbent 23_XNO_XNO without being absorbed into the absorbent 23_XNO passing through the absorbent 23_XGradually increases, and thus a downward peak PKm2 is generated.
[0197]
Outflow NO at upward peak PKM_XWhen the concentration CNO is called a maximum value CPK, the maximum value CPK is NO._XNO by processing_XNO released from the absorbent 23_XDepends on the amount of NO and therefore NO_XNO when processing starts_XAccumulated NO in absorbent 23_XDepends on the amount. That is, accumulated NO_XWhen the amount is large, the maximum value CPK increases and the accumulated NO_XWhen the amount is small, the maximum value CPK becomes small. On the other hand, NO_XAccumulated SO in the absorbent 23_XNO as the amount increases_XNO that can be stored in the absorbent 23_XThe amount of.
[0198]
Therefore, the accumulated SO_XIt can be seen that the maximum value CPK decreases as the amount increases, and the maximum value CPK decreases as the fuel consumption QF increases.
[0199]
Therefore, in the second embodiment according to the present invention, a relational expression representing the relation between the maximum value CPK and the fuel consumption QF is obtained, and the SO_XAn estimated value CPKE of the maximum value CPK when it is determined that the processing should be started is calculated using this relational expression.
[0200]
Specifically, NO_XEach time processing is performed, a maximum value CPK is calculated. Then, as shown in FIG. 37, every time the average value calculation time comes, that is, every time the fuel consumption amount QF increases by the second set amount qf2, the maximum value average value CPKAV is calculated, and the maximum value is calculated. A linear expression m representing the relationship between the average value CPKAV and the fuel consumption QF is determined based on these maximum value average values CPKAV. Then the previous SO_XAfter the processing is completed (see arrow X1 in FIG. 37), when the fuel consumption amount QF increases by the first set amount qf1, the local maximum estimated value CPKE at this time is calculated using the primary expression m.
[0201]
Furthermore, maximum value CPK and accumulated SO_XThe relationship with the quantity estimated value QSE is obtained in advance by experiments, and is stored in advance in the ROM 42 in the form of a map shown in FIG. Therefore, SO_XIf the local maximum estimated value CPKE when it is determined that the processing should be performed is calculated, the accumulated SO at this time is calculated._XThe quantity estimated value QSE is obtained using the map of FIG.
[0202]
Therefore, in general terms, NO_XOutflow NO generated when processing is performed_XThe maximum value of the quantity is detected, and the detected outflow NO_XNO based on local maximum_XAccumulated SO of absorbent_XIt means that we are looking for quantity.
[0203]
FIG. 39 shows a routine for calculating the coefficient kmi of the second embodiment according to the present invention. Also in the second embodiment according to the present invention, the SO shown in FIG._XA control routine is executed, and the calculation routine of this coefficient kmi is SO_XThis is executed in step 203 of the control routine.
[0204]
Referring to FIG. 39, in step 400, it is determined whether or not the difference between the current fuel consumption amount QF and the fuel consumption amount QFP2 at the previous average value calculation timing is equal to or greater than a second set amount qf2. When QF-QFP2 <qf2, that is, when the fuel has not been consumed by the second set amount qf2 after the previous maximum average value CPKAV is calculated, the routine proceeds to step 401, where NO_XIt is determined whether or not processing is being performed. NO_XWhen the process is not performed, the process cycle is ended and NO_XWhen the process is being performed, the routine proceeds to step 402 where the outflow NO._XA maximum value CPK of the concentration CNO is calculated. Note that the maximum value CPK is determined as the engine operating state or NO._XCorrection can also be made based on the state of the catalyst 23. In the subsequent step 403, an integrated value SCPK of the maximum value CPK is calculated (SCPK = SCPK + CPK). In subsequent step 404, a detection number counter nCPK representing the number of detections of the maximum value CPK is incremented by 1 (nCPK = nCPK + 1).
[0205]
When QF−QFP2 ≧ qf2, the routine proceeds from step 400 to step 405, where the maximum value average value CPKAV is calculated (CPKAV = SCPK / nCPK). In the following step 406, an average value calculation number counter nAV indicating the number of calculation of the maximum value average value CPKAV is incremented by 1 (nAV = nAV + 1). In the subsequent step 407, the coefficient kmi (i = 1, 2, 3, 4) is calculated based on the following equations.
[0206]
km1 = km1 + QF
km2 = km2 + QF²
km3 = km3 + QF ・ CPK
km4 = km4 + CPK
In the subsequent step 408, the current fuel consumption amount QF is stored as QFP2, and the integrated value SCPK and the counter nCPK are cleared.
[0207]
FIG. 40 shows a routine for setting the processing start time amount QSS of the second embodiment according to the present invention. This routine is SO_XIt is executed in step 204 of the control routine.
[0208]
Referring to FIG. 40, in step 410, a local maximum estimated value CPKE is calculated based on the following equation.
[0209]
[Formula 6]

[0210]
In the following step 411, the accumulated SO is obtained from the map of FIG._XA quantity estimate QSE is calculated. In the following step 412, the processing start time amount QSS is stored in the accumulated SO._XThe quantity estimated value QSE is set. In the subsequent step 413, the coefficient kmi (i = 1, 2, 3, 4) and the average value calculation number counter nAV are cleared.
[0211]
Since the other structure and operation of the second embodiment according to the present invention are the same as those of the above-described embodiment, the description thereof will be omitted.
[0212]
Next, a third embodiment according to the present invention will be described.
[0213]
As mentioned above, NO_XWhen processing is done, NO_XNO stored in absorbent 23_XNO without being reduced_XIt flows out from the absorbent 23. NO at this time_XNO flowing out from the absorbent 23_XAs shown in FIG. 41, the accumulated amount SQL of_XThe outflow NO from when the downward peak PKm1 occurs to the concentration CNO until when the downward peak PKm2 occurs_XIt can be obtained in the form of an integrated value of the concentration CNO.
[0214]
This outflow NO_XThe integrated amount SQL is NO, similar to the above-mentioned maximum value CPK._XNO when processing starts_XAccumulated NO in absorbent 23_XDepends on the amount. That is, accumulated SO_XSpilled NO as the volume increases_XAs the accumulated amount SQL decreases and the fuel consumption QF increases, the outflow NO._XThe accumulated amount SQL is reduced.
[0215]
Therefore, in the third embodiment according to the present invention, the outflow NO._XA relational expression representing the relationship between the integrated amount SQL and the fuel consumption amount QF is obtained, and SO_XOutflow NO when it is determined that processing should be started_XThe estimated value SQLLE of the integrated amount SQL is calculated using this relational expression.
[0216]
Specifically, NO_XSpill NO every time processing is done_XAn integrated amount SQL is calculated. Then, as shown in FIG. 42, every time the average value calculation time comes, that is, every time the fuel consumption amount QF increases by the second set amount qf2, the outflow NO._XThe accumulated amount average value SQLAV is calculated and the outflow NO._XThe primary expression m representing the relationship between the integrated amount average value SQLAV and the fuel consumption amount QF is the outflow NO._XIt is determined based on the integrated amount average value SQLAV. Then the previous SO_XAfter the processing is completed (see arrow X1 in FIG. 42), if the fuel consumption amount QF increases by the first set amount qf1, the outflow NO at this time_XThe accumulated amount estimated value SQLLE is calculated using the primary expression m.
[0217]
Furthermore, outflow NO_XAccumulated amount SQL and accumulated SO_XThe relationship with the quantity estimated value QSE is obtained in advance by experiments, and is stored in advance in the ROM 42 in the form of a map shown in FIG. Therefore, SO_XOutflow NO when it is determined that processing should be performed_XIf the integrated amount estimated value SQLLE is calculated, the accumulated SO at this time_XThe quantity estimated value QSE is obtained using the map of FIG.
[0218]
Therefore, in general terms, NO_XOutflow NO when processing is performed_XDetects the amount integrated value and detects the outflow NO._XNO based on the integrated value_XAccumulated SO of absorbent_XIt means that we are looking for quantity.
[0219]
FIG. 44 shows the routine for calculating the coefficient kmi of the third embodiment according to the present invention. In the third embodiment according to the present invention, the SO shown in FIG._XA control routine is executed, and the calculation routine of this coefficient kmi is SO_XThis is executed in step 203 of the control routine.
[0220]
Referring to FIG. 44, in step 420, it is determined whether or not the difference between the current fuel consumption amount QF and the fuel consumption amount QFP2 at the previous average value calculation timing is equal to or greater than a second set amount qf2. When QF-QFP2 <qf2, that is, when the fuel has not been consumed by the second set amount qf2 after the previous maximum average value CPKAV is calculated, the routine proceeds to step 421, where NO_XIt is determined whether or not processing is being performed. NO_XWhen the process is not performed, the process cycle is ended and NO_XWhen the process is being performed, the routine proceeds to step 422, where the outflow NO._XA quantity integrated value SQL is calculated. Outflow NO_XThe amount integrated value SQL is set to the engine operating state or NO._XCorrection can also be made based on the state of the catalyst 23. In the following step 423, the outflow NO_XAn integrated value SSQL of the amount integrated value SQL is calculated (SSQL = SSQL + SQL). In the following step 424, the outflow NO_XA detection number counter nSQL indicating the number of detections of the integrated value SQL is incremented by 1 (nSQL = nSQL + 1).
[0221]
When QF-QFP2 ≧ qf2, the routine proceeds from step 420 to step 425, where the outflow NO_XA quantity integrated value average value SQLLAV is calculated (SQLLAV = SSQL / nSQL). In the following step 426, the outflow NO_XAn average value calculation number counter nAV indicating the number of calculation of the amount integrated value average value SQLAV is incremented by 1 (nAV = nAV + 1). In the subsequent step 427, the coefficients kmi (i = 1, 2, 3, 4) are calculated based on the following equations, respectively.
[0222]
km1 = km1 + QF
km2 = km2 + QF²
km3 = km3 + QF ・ SQL
km4 = km4 + SQL
In the following step 428, the current fuel consumption amount QF is stored as QFP2, and the integrated value SSQL and the counter nSQL are cleared.
[0223]
FIG. 45 shows a routine for setting the processing start time amount QSS of the third embodiment according to the present invention. This routine is SO_XIt is executed in step 204 of the control routine.
[0224]
Referring to FIG. 45, in step 430, the outflow NO._XThe integrated amount estimated value SQLLE is calculated based on the following equation.
[0225]
[Expression 7]

[0226]
In the subsequent step 431, the accumulated SO is obtained from the map of FIG._XA quantity estimate QSE is calculated. In the subsequent step 432, the processing start time amount QSS is stored in the accumulated SO._XThe quantity estimated value QSE is set. In the subsequent step 433, the coefficient kmi (i = 1, 2, 3, 4) and the average value calculation number counter nAV are cleared.
[0227]
The other configurations and operations of the third embodiment according to the present invention are the same as those of the above-described embodiment, so that the description thereof is omitted.
[0228]
Next, a fourth embodiment according to the present invention will be described.
[0229]
As described above with reference to FIGS. 2-5, in an embodiment according to the present invention, NO_XOutflow NO when the air-fuel ratio AFN of the exhaust gas flowing into the absorbent 23 is kept lean_XConcentration CNO does not exceed the allowable maximum concentration, and NO_XBy correcting the basic time interval tNB with the correction amount ktN so that the fuel consumption rate by the processing does not exceed the allowable maximum value, NO_XThe processing time interval tN is calculated (tN = tNB + ktN).
[0230]
Accumulated SO_XWhen the amount is small, NO_XNO flowing into the absorbent 23_XNO_XNO stored in absorbent 23_XSpill NO_XThe time until the concentration CNO reaches the maximum allowable concentration is relatively long. As a result, as shown in FIG._XThe processing time interval tN is relatively long. In contrast, accumulated SO_XWhen the amount increases, NO_XNO stored in absorbent 23_XSpillage NO._XThe time until the concentration CNO reaches the allowable maximum concentration is shortened, and the time interval tN is shortened as shown in FIG.
[0231]
In the embodiment according to the present invention, as described above, the accumulated SO_XThe basic time interval tNB is set so as to be optimal when the amount is almost zero, and the accumulated SO_XAs the amount increases, the time interval tN decreases. Therefore, the accumulated SO_XThe time interval ratio tNB / tN increases as the amount increases, and the time interval ratio tNB / tN increases as the fuel consumption QF increases.
[0232]
However, as described above, the time interval tN may vary depending on the engine operating state. However, the time interval tN or the time interval ratio tNB / tN during steady operation is not affected by the engine operating state, and the deviation of the actual time interval tN from the basic time interval tNB during steady engine operation is accumulated SO_XBy quantity.
[0233]
Therefore, in the fourth embodiment according to the present invention, NO is required during steady operation._XA relational expression representing the relationship between the time interval ratio when processing is performed, that is, the steady time interval ratio RST (= tNB / (tNB + ktN)), and the fuel consumption amount QF is obtained._XThe estimated value RSTE of the steady time interval ratio RST when it is determined that the processing should be started is calculated using this relational expression.
[0234]
Specifically, NO during steady operation_XEach time processing is performed, a steady time interval ratio RST is calculated. Then, as shown in FIG. 47, every time the average value calculation time comes, that is, every time the fuel consumption amount QF increases by the second set amount qf2, the steady time interval ratio average value RSTAV is calculated. The primary expression m representing the relationship between the steady time interval ratio average value RSTAV and the fuel consumption QF is determined based on the steady time interval ratio average value RSTAV. Then the previous SO_XAfter the processing is completed (see arrow X1 in FIG. 47), when the fuel consumption amount QF increases by the first set amount qf1, the steady time interval ratio estimated value RSTE at this time is calculated using the primary expression m. The
[0235]
Furthermore, the steady time interval ratio RST and accumulated SO_XThe relationship with the quantity estimated value QSE is obtained in advance by experiments, and is stored in advance in the ROM 42 in the form of a map shown in FIG. Therefore, SO_XIf the steady time interval ratio estimated value RSTE when it is determined that the processing should be performed is calculated, the accumulated SO at this time is calculated._XThe quantity estimated value QSE is obtained using the map of FIG.
[0236]
Therefore, in general terms, NO_XOutflow NO when the air-fuel ratio of the exhaust gas flowing into the absorbent 23 is kept lean_XNO so that the amount does not exceed the predetermined maximum allowable amount_XSet the time interval at which processing takes place, and NO based on the controlled time interval_XThis means that the amount of accumulated sulfur in the absorbent is being sought.
[0237]
FIG. 49 shows the routine for calculating the coefficient kmi of the fourth embodiment according to the present invention. In the fourth embodiment according to the present invention, the SO shown in FIG._XA control routine is executed, and the calculation routine of this coefficient kmi is SO_XThis is executed in step 203 of the control routine.
[0238]
Referring to FIG. 49, in step 440, it is determined whether or not the difference between the current fuel consumption amount QF and the fuel consumption amount QFP2 at the previous average value calculation timing is equal to or larger than a second set amount qf2. When QF-QFP2 <qf2, that is, when the fuel has not been consumed by the second set amount qf2 since the previous steady time interval ratio average value RSSTAV is calculated, the routine then proceeds to step 441, during steady operation and NO_XIt is determined whether or not processing is being performed. Not during steady operation or NO_XWhen the process is not being performed, the process cycle is terminated, and during steady operation and NO_XWhen the process is being performed, the routine proceeds to step 442, where the region to which the current engine operating state belongs is determined. In the subsequent step 443, a steady time interval ratio RST is calculated.
[0239]
RST = tNB / (tNB + ktN)
In the subsequent step 444, the integrated value SRST of the steady time interval ratio RST is calculated (SRST = SRST + RST). In the following step 445, a detection number counter nRST indicating the number of detections of the stationary time interval ratio RST is incremented by 1 (nRST = nRST + 1).
[0240]
When QF−QFP2 ≧ qf2, the routine proceeds from step 440 to step 446, where the steady time interval ratio average value RSTAV is calculated (RSTAV = SRST / nRST). In the subsequent step 447, an average value calculation number counter nAV representing the number of times of calculation of the steady time interval ratio RSTAV is incremented by 1 (nAV = nAV + 1). In the following step 448, the coefficients kmi (i = 1, 2, 3, 4) are calculated based on the following equations, respectively.
[0241]
km1 = km1 + QF
km2 = km2 + QF²
km3 = km3 + QF ・ RST
km4 = km4 + RST
In the following step 449, the current fuel consumption amount QF is stored as QFP2, and the integrated value SRST and the counter nRST are cleared.
[0242]
FIG. 50 shows a routine for setting the processing start time amount QSS of the fourth embodiment according to the present invention. This routine is SO_XIt is executed in step 204 of the control routine.
[0243]
Referring to FIG. 50, in step 450, a steady time interval ratio estimated value RSTE is calculated based on the following equation.
[0244]
[Equation 8]

[0245]
In the subsequent step 451, the accumulated SO is obtained from the map of FIG._XA quantity estimate QSE is calculated. In the following step 452, the processing start time amount QSS is stored in the accumulated SO._XThe quantity estimated value QSE is set. In the subsequent step 453, the coefficient kmi (i = 1, 2, 3, 4) and the average value calculation number counter nAV are cleared.
[0246]
Since the other structure and operation of the fourth embodiment according to the present invention are the same as those of the above-described embodiment, description thereof will be omitted.
[0247]
Next, a fifth embodiment according to the present invention will be described.
[0248]
As described with reference to FIG. 46, the accumulated SO_XNO according to quantity_XThe processing time interval tN varies. Accordingly, the total amount of reducing agent supplied from the reducing agent supply valve 31 per unit period is also stored in the accumulated SO._XIt will vary depending on the amount. That is, accumulated SO_XAs the amount increases, the reducing agent supply amount QR per unit period increases, and as the fuel consumption amount QF increases, the reducing agent supply amount QR per unit period increases.
[0249]
Therefore, in the fifth embodiment according to the present invention, a relational expression representing the relationship between the reducing agent supply amount QR and the fuel consumption amount QF per unit period is obtained, and the SO_XAn estimated value QRE of the reducing agent supply amount QR per unit period when it is determined that the processing should be started is calculated using this relational expression.
[0250]
Specifically, as shown in FIG. 51, every time the supply amount calculation time comes, that is, for example, every time the fuel consumption amount QF increases by the second set amount qf2, the integrated value of the reducing agent supply amount, that is, The reducing agent supply amount QR per unit period is calculated, and a primary expression m representing the relationship between the reducing agent supply amount QR per unit period and the fuel consumption amount QF is determined based on the reducing agent supply amount QR per unit period. The Then the previous SO_XAfter the processing is completed (see arrow X1 in FIG. 51), when the fuel consumption amount QF increases by the first set amount qf1, the estimated value QRE of the reducing agent supply amount per unit period at this time becomes the linear expression m. Is used to calculate.
[0251]
Furthermore, the reducing agent supply amount QR per unit period and the accumulated SO_XThe relationship with the quantity estimated value QSE is obtained in advance by experiments, and is stored in advance in the ROM 42 in the form of a map shown in FIG. Therefore, SO_XIf the estimated value QRE of the reducing agent supply amount per unit period when it is determined that the processing should be performed is calculated, the accumulated SO at this time is calculated._XThe quantity estimated value QSE is obtained using the map of FIG.
[0252]
Therefore, in general terms, NO per unit period_XDetermine the amount of fuel or reducing agent consumed by the treatment, NO per unit period_XNO based on the amount of fuel or reducing agent consumed by the treatment_XAccumulated SO of absorbent_XIt means that we are looking for quantity.
[0253]
FIG. 53 shows a calculation routine of the coefficient kmi of the fifth embodiment according to the present invention. In the fifth embodiment according to the present invention, the SO shown in FIG._XA control routine is executed, and the calculation routine of this coefficient kmi is SO_XThis is executed in step 203 of the control routine.
[0254]
Referring to FIG. 53, in step 460, it is determined whether or not the difference between the current fuel consumption amount QF and the fuel consumption amount QFP2 at the previous reducing agent amount calculation timing is equal to or larger than a second set amount qf2. When QF-QFP2 <qf2, that is, when the fuel has not been consumed by the second set amount qf2 after the calculation of the reducing agent supply amount QR per unit, the routine proceeds to step 461, where NO_XIt is determined whether or not processing is being performed. NO_XWhen the process is not performed, the process cycle is ended and NO_XWhen the process is being performed, the process proceeds to step 462, where the current NO._XIn the process, the reducing agent amount qr supplied from the reducing agent supply valve 31 is calculated. In the subsequent step 463, an integrated value Sqr of the reducing agent amount qr is calculated (Sqr = Sqr + qr).
[0255]
When QF−QFP2 ≧ qf2, the process proceeds from step 460 to step 464, and the integrated value Sqr is set as the reducing agent supply amount QR per unit period. In the following step 465, a calculation number counter nQR indicating the number of calculation of the reducing agent supply amount QR per unit is incremented by 1 (nQR = nQR + 1). In the subsequent step 466, the coefficients kmi (i = 1, 2, 3, 4) are calculated based on the following equations, respectively.
[0256]
km1 = km1 + QF
km2 = km2 + QF²
km3 = km3 + QF ・ QR
km4 = km4 + QR
In the following step 467, the current fuel consumption amount QF is stored as QFP2, and the integrated value Sqr and the counter nQR are cleared.
[0257]
FIG. 54 shows a routine for setting the processing start time amount QSS of the fifth embodiment according to the present invention. This routine is SO_XIt is executed in step 204 of the control routine.
[0258]
Referring to FIG. 54, in step 470, an estimated value QRE of the reducing agent supply amount per unit period is calculated based on the following equation.
[0259]
[Equation 9]

[0260]
In the subsequent step 471, the accumulated SO is extracted from the map of FIG._XA quantity estimate QSE is calculated. In the following step 472, the processing start time amount QSS is stored in the accumulated SO._XThe quantity estimated value QSE is set. In the subsequent step 473, the coefficient kmi (i = 1, 2, 3, 4) and the calculation number counter nQR are cleared.
[0261]
The other configurations and operations of the fifth embodiment according to the present invention are the same as those of the above-described embodiment, so that the description thereof is omitted.
[0262]
【The invention's effect】
NO_XThe amount of sulfur stored in the absorbent can be accurately determined.
[Brief description of the drawings]
FIG. 1 is an overall view of an internal combustion engine.
[Fig. 2] NO_XIt is a time chart for demonstrating a process.
FIG. 3 NO_XIt is a diagram showing a basic time interval tNB and a correction amount ktN of processing.
FIG. 4 NO_XIt is a flowchart which shows a control routine.
FIG. 5 is a flowchart showing a routine for updating a correction amount ktN.
FIG. 6 shows an example SO according to the present invention._XIt is a time chart for demonstrating a process.
[FIG. 7] Inflow NO_XIt is a diagram which shows density | concentration CNI.
FIG. 8 is a diagram for explaining a primary expression m.
FIG. 9 Accumulated SO_XQuantity estimate QSE and NO_XIt is a diagram which shows the relationship with the purification rate EFF.
FIG. 10: Accumulated SO_XIt is a diagram which shows possible maximum quantity QSEM of quantity estimated value QSE.
FIG. 11: NO_XIt is a time chart for demonstrating calculation timings, such as purification rate average value EFFAV.
FIG. 12 NO_XIt is a time chart for demonstrating density | concentration detection period DP.
FIG. 13: NO_XIt is a time chart for demonstrating calculation timings, such as purification rate EFF.
FIG. 14: Accumulated SO_XIt is a time chart for demonstrating calculation timings, such as quantity estimated value QSE.
FIG. 15: SO_XIt is a flowchart which shows a control routine.
FIG. 16 SO_XIt is a diagram which shows the decreasing amount qsr.
FIG. 17 is a flowchart showing a calculation routine of a coefficient kmi.
FIG. 18 NO_XIt is a flowchart which shows a purification rate calculation routine.
FIG. 19 NO_XIt is a flowchart which shows a purification rate calculation routine.
FIG. 20 NO_XIt is a flowchart which shows a purification rate calculation routine.
FIG. 21: Accumulated NO_XIt is a flowchart which shows the calculation routine of quantity estimated value QNE.
FIG. 22 NO_XIt is a diagram which shows increase amount qna.
FIG. 23: Inflow NO_XIt is a flowchart which shows the calculation routine of density correction value CNIC.
FIG. 24 is a diagram showing a delay time DLY.
FIG. 25: Inflow NO_XIt is a time chart for demonstrating the calculation method of density correction value CNIC.
FIG. 26: Outflow NO_XIt is a flowchart which shows the calculation routine of density correction value CNOC.
FIG. 27: NO_XIt is a flowchart which shows the calculation routine of the purification rate correction value EFFC.
FIG. 28 is a diagram showing correction coefficients kE1, kE2, and kE3.
FIG. 29 is a flowchart showing a routine for calculating a thermal deterioration counter nTCH.
FIG. 30 is a flowchart showing an oil supply determination routine.
FIG. 31 is a diagram for explaining a fuel supply determination method;
FIG. 32 is a flowchart showing a routine for setting a processing start time amount QSS.
FIG. 33 is a flowchart showing a routine for setting a processing start time amount QSS.
FIG. 34 is a flowchart showing a routine for setting a processing start time amount QSS.
FIG. 35 is a diagram for explaining a first embodiment according to the present invention.
FIG. 36 is a diagram for explaining a second embodiment according to the present invention.
FIG. 37 is a view for explaining a second embodiment according to the present invention.
FIG. 38: Accumulated SO_XIt is a diagram which shows the relationship between quantity estimated value QSE and maximum value CPK.
FIG. 39 is a flowchart showing a calculation routine of a coefficient kmi according to the second embodiment of the present invention.
FIG. 40 is a flowchart showing a routine for setting a processing start time amount QSS according to the second embodiment of the present invention;
FIG. 41 is a diagram for explaining a third embodiment according to the present invention.
FIG. 42 is a view for explaining a third embodiment according to the present invention.
FIG. 43: Accumulated SO_XQuantity estimate QSE and spill NO_XIt is a diagram which shows the relationship with integration amount SQL.
FIG. 44 is a flowchart showing a calculation routine of a coefficient kmi of the third embodiment according to the present invention.
FIG. 45 is a flowchart showing a routine for setting a processing start time amount QSS according to the third embodiment of the present invention;
FIG. 46 is a view for explaining a fourth embodiment according to the present invention.
FIG. 47 is a view for explaining a fourth embodiment according to the present invention.
FIG. 48: Accumulated SO_XIt is a diagram which shows the relationship between quantity estimated value QSE and stationary time interval ratio RST.
FIG. 49 is a flowchart showing a calculation routine of a coefficient kmi according to the fourth embodiment of the present invention.
FIG. 50 is a flowchart showing a routine for setting a processing start time amount QSS according to the fourth embodiment of the present invention.
FIG. 51 is a diagram for explaining a fifth embodiment according to the present invention;
FIG. 52: Accumulated SO_XIt is a diagram which shows the relationship between quantity estimated value QSE and reducing agent supply amount QR.
FIG. 53 is a flowchart showing a calculation routine of a coefficient kmi according to the fifth embodiment of the present invention.
FIG. 54 is a flowchart showing a routine for setting a processing start time amount QSS according to the fifth embodiment of the present invention;
[Explanation of symbols]
1 ... Engine body
20a, 20b ... exhaust pipe
23 ... NO_XAbsorbent
31 ... Reducing agent supply valve
51 ... NO_XSensor

Claims (22)

In an exhaust passage of an internal combustion engine burning fuel under a lean air-fuel ratio is carried out, stored the NO _X in the exhaust gas air-fuel ratio of the inflowing exhaust gas flows when the lean air-fuel ratio of the exhaust gas flowing In an internal combustion engine in which an NO _X absorbent that reduces the amount of NO _X stored by reducing the stored NO _X when the reducing agent is contained in the exhaust gas when the exhaust gas is reduced is provided, the NO _X absorbent and NO _X sensor arranged in _the NO _X absorbent in the exhaust passage downstream of to detect the outflow amount of NO _X is the amount of the NO _X in the exhaust gas flowing out from, are stored in _the NO _X absorbent An exhaust emission control device comprising means for obtaining an accumulated sulfur amount that is the amount of sulfur based on the outflow NO _x amount, a means for performing sulfur treatment for reducing the accumulated sulfur amount of the NO _x absorbent, and sulfur treatment Whether or not Further comprising means for, with when it is determined to be subjected to sulfur treatment seeking accumulated sulfur amount of the NO _X absorbent in this case, the processing start time amount is set to the accumulation of sulfur amount determined the NO Sulfur treatment necessary to reduce the amount of accumulated sulfur of the _X absorbent from the starting amount of the treatment to a predetermined lower limit or less is performed, and the obtained accumulated amount of NO _X absorbent is determined in advance. It was when greater than the boundary amount necessary accumulated sulfur amount of the NO _X absorbent to start processing when the amount is set to the accumulation of sulfur amount obtained said to decrease from the start of processing at the amount to below the lower limit amount such performed sulfur treatment, the when accumulated sulfur amount of the NO _X absorbent obtained is less than the boundary amount, the processing start time amount is set independent amounts and accumulated sulfur amount obtained the NO _X Suck Exhaust purification device the accumulation amount of sulfur agent to perform the sulfur treatment required to reduce from the start of processing at the amount to below the lower limit amount. Unrelated amount and the determined accumulation amount of sulfur, to claim 1 in which the fuel is set to the maximum sulfur content that may be stored in _the NO _X absorbent assuming that the sulfur concentration is a fuel reference density An exhaust gas purification apparatus for an internal combustion engine as described . _The NO _X absorbent calculated accumulated amount of sulfur, the allowable minimum outflow amount of NO _X said detected is predetermined based on these outflow amount of NO _X with spaced from each other time intervals outflow amount of NO _X to detect a plurality of times when the number of time less than the amount exceeds the upper limit number, the processing at the start amount accumulated sulfur amount of the NO _X absorbent to start processing at the amount set independent amounts and accumulated sulfur amount obtained the 2. The exhaust emission control device for an internal combustion engine according to claim 1, wherein sulfur treatment necessary to reduce the amount to less than the lower limit amount is performed . NO _X in the exhaust gas flowing in when the air-fuel ratio of the inflowing exhaust gas is lean is stored in the exhaust passage of the internal combustion engine in which combustion is performed under a lean air-fuel ratio, and the air-fuel ratio of the inflowing exhaust gas is In an internal combustion engine in which an NO _X absorbent that reduces the amount of NO _X stored by reducing the stored NO _X when the reducing agent is contained in the exhaust gas when the exhaust gas is reduced is provided , the NO _X absorbent and NO _X sensor arranged in _the NO _X absorbent in the exhaust passage downstream of to detect the outflow amount of NO _X is the amount of the NO _X in the exhaust gas flowing out from, are stored in _the NO _X absorbent An exhaust emission control device comprising means for obtaining an accumulated sulfur amount that is the amount of sulfur based on the outflow NO _x amount, a means for performing sulfur treatment for reducing the accumulated sulfur amount of the NO _x absorbent, and sulfur treatment Whether or not Further comprising means for, with when it is determined to be subjected to sulfur treatment seeking accumulated sulfur amount of the NO _X absorbent in this case, the processing start time amount is set to the accumulation of sulfur amount determined the NO _The sulfur treatment necessary to reduce the amount of accumulated sulfur in the _X absorbent from the amount at the start of the treatment to a predetermined lower limit or less is performed, and the set amount at the start of treatment is obtained in the market for fuel. If it is assumed that it is the fuel with the highest sulfur concentration among the possible fuels, and if it is greater than the maximum amount of sulfur that can be stored in the NO _X absorbent, the amount at the start of processing can be stored. An exhaust emission control device configured to perform sulfur treatment necessary to reduce the accumulated sulfur amount of the NO _X absorbent from the starting amount to the lower limit amount or less by setting the maximum amount . The exhaust emission control device for an internal combustion engine according to claim 1 or 4, wherein when the fuel is consumed by a predetermined first set amount after completion of the previous sulfur treatment, it is determined that the sulfur treatment should be performed . Inflow into the exhaust passage of an internal combustion engine where combustion occurs under a lean air-fuel ratio NO in the exhaust gas flowing in when the air-fuel ratio of the exhaust gas to be lean is lean _X NO when storing the reducing agent in the exhaust gas when the air-fuel ratio of the exhaust gas flowing in decreases _X NO is being reduced and stored _X NO decreases in quantity _X In an internal combustion engine with an absorbent, NO _X NO in exhaust gas flowing out from absorbent _X The amount of spillage NO _X NO to detect quantity _X NO placed in the exhaust passage downstream of the absorbent _X Sensor and NO _X The amount of accumulated sulfur, which is the amount of sulfur stored in the absorbent, is converted to the outflow NO. _X NOx purification apparatus comprising means for obtaining based on quantity, NO _X NO in the exhaust gas flowing into the absorbent _X Inflow NO _X Means for determining the amount and inflow NO _X Volume and spill NO _X NO based on quantity _X Absorbent NO _X Means for determining the purification rate and the determined NO _X NO based on purification rate _X An exhaust emission control device further comprising means for determining the amount of accumulated sulfur in the absorbent. NO _X Absorbent NO _X The purification rate is obtained several times at intervals of each other, and these obtained NOs are obtained. _X A relational expression representing the relation between the purification rate and the fuel consumption is obtained, and the NO at the judgment time is determined from the fuel consumption at the judgment time and the relational expression. _X The purification rate is obtained, and NO at the determined judgment time. _X The exhaust emission control device for an internal combustion engine according to claim 6, wherein the amount of accumulated sulfur at the determination timing is obtained based on a purification rate. The determined NO _X The exhaust gas purification apparatus for an internal combustion engine according to claim 7, wherein the relationship between the purification rate and the fuel consumption is expressed by a linear expression, and the linear expression is obtained by a least square method. NO _X Absorbent NO _X The purification rate is obtained several times at intervals of each other, and the NO obtained during the period from the previous average value calculation time to the current average value calculation time every time the average value calculation time arrives. _X The average value of the purification rate is calculated, and the calculated NO _X The exhaust emission control device for an internal combustion engine according to claim 7, wherein a relational expression representing a relationship between an average value of the purification rate and fuel consumption is obtained. NO _X NO stored in the absorbent _X NO _X NO stored in the absorbent _X NO to reduce the amount of _X NO temporarily switching the air-fuel ratio of the exhaust gas flowing into the absorbent to rich _X Means for processing and NO _X Outflow NO when the air-fuel ratio of the exhaust gas flowing into the absorbent is kept lean _X NO so that the amount does not exceed the predetermined maximum allowable amount _X Means for controlling the time interval during which processing is performed, and NO _X Inflow NO when the air-fuel ratio of the exhaust gas flowing into the absorbent is lean _X Volume and spill NO _X While calculating the amount, these inflow NO _X Volume and spill NO _X NO based on quantity _X Absorbent NO _X The purification rate is calculated, and NO is determined during the period from the previous average value calculation time to the current average value calculation time. _X When the number of times the purification rate has been obtained is less than the lower limit number, NO during that period _X The exhaust emission control device for an internal combustion engine according to claim 9, wherein an average value of the purification rate is not calculated. NO calculated every time the average value calculation time arrives _X The average value of the purification rate is NO _X NO weighted according to whether steady engine operation or transient engine operation is performed when the purification rate is determined _X The exhaust emission control device for an internal combustion engine according to claim 9, which is a weighted average value of the purification rate. The exhaust emission control device for an internal combustion engine according to claim 9, wherein it is determined that the next average value calculation time has come when the fuel is consumed by a predetermined second set amount from the previous average value calculation time. NO _X Means for performing sulfur treatment to reduce the amount of accumulated sulfur in the absorbent, and the determined NO _X When the accumulated sulfur amount of the absorbent is larger than a predetermined boundary amount, it is determined that the fuel is a high-sulfur concentration fuel, and the obtained NO _X And a means for determining that the fuel is a low-sulfur concentration fuel when the amount of accumulated sulfur in the absorbent is less than the boundary amount, and when the fuel is determined to be a high-sulfur concentration fuel, the processing start amount is determined. NO is set to the obtained accumulated sulfur amount. _X When the sulfur treatment necessary to reduce the amount of accumulated sulfur in the absorbent from the amount at the start of the treatment to a predetermined lower limit or less is performed, and it is determined that the fuel is a low-sulfur concentration fuel, the amount at the start of treatment NO is set to an amount irrelevant to the obtained accumulated sulfur amount. _X The exhaust emission control device for an internal combustion engine according to claim 7, wherein sulfur treatment necessary to reduce the amount of accumulated sulfur in the absorbent from the amount at the start of treatment to the lower limit amount or less is performed. The determined NO _X Means for obtaining a deviation of the purification rate from the relational expression is further provided, and when the obtained deviation changes beyond a predetermined threshold when it is determined that the fuel is a high-sulfur concentration fuel, the fuel Is changed to a low-sulfur concentration fuel, and when it is determined that the fuel is a low-sulfur concentration fuel, the fuel is changed to a high-sulfur concentration fuel when the obtained deviation changes beyond the threshold value. The exhaust emission control device for an internal combustion engine according to claim 13, wherein the exhaust gas purification apparatus is determined to have been performed. NO _X Outflow NO set during a period when the air-fuel ratio of the exhaust gas flowing into the absorbent is lean _X Outflow NO during volume detection period _X While calculating the average value of the amount, the outflow NO _X Inflow NO during volume detection period _X Calculate the average amount of these inflow NO _X Average volume and spill NO _X The spill NO based on the average amount _X NO in quantity detection period _X The exhaust gas purification apparatus for an internal combustion engine according to claim 6, wherein the purification rate is obtained. Inflow NO _X Volume and spill NO _X The amount of exhaust gas is required at almost the same time, and the exhaust gas part is inflow NO. _X NO after quantity is requested _X Means for determining the delay time required to reach the sensor, and inflow NO with the delay time _X Volume or spill NO _X And a means for correcting the amount, the corrected inflow NO. _X Volume or spill NO _X NO based on quantity _X The exhaust gas purification apparatus for an internal combustion engine according to claim 6, wherein the purification rate is obtained. The determined NO _X NO purification rate _X NO when the absorbent is in a predetermined reference state _X A means for converting to a purification rate is further provided, and the converted NO _X NO based on purification rate _X The exhaust emission control device for an internal combustion engine according to claim 6, wherein the accumulated sulfur amount of the absorbent is obtained. NO _X Means for determining the degree of thermal degradation of the absorbent and the determined NO _X The obtained NO based on the degree of thermal deterioration of the absorbent. _X Means for correcting the purification rate, and the corrected NO. _X NO based on purification rate _X The exhaust emission control device for an internal combustion engine according to claim 6, wherein the accumulated sulfur amount of the absorbent is obtained. The engine further includes means for detecting an engine operating state, and the inflow NO is based on the detected engine operating state. _X The exhaust emission control device for an internal combustion engine according to claim 6, wherein the amount is calculated. NO in the exhaust gas flowing into the exhaust passage of the internal combustion engine where the combustion is performed under a lean air-fuel ratio when the air-fuel ratio of the exhaust gas flowing in is lean _X NO when storing the reducing agent in the exhaust gas when the air-fuel ratio of the exhaust gas flowing in decreases _X NO is being reduced and stored _X NO decreases in quantity _X In an internal combustion engine with an absorbent, NO _X NO in exhaust gas flowing out from absorbent _X The amount of spillage NO _X NO to detect quantity _X NO placed in the exhaust passage downstream of the absorbent _X Sensor and NO _X The amount of accumulated sulfur, which is the amount of sulfur stored in the absorbent, is converted to the outflow NO. _X NOx purification apparatus comprising means for obtaining based on quantity, NO _X NO stored in the absorbent _X NO _X NO stored in the absorbent _X NO to reduce the amount of _X NO temporarily switching the air-fuel ratio of the exhaust gas flowing into the absorbent to rich _X Means for processing and NO _X Outflow NO generated when processing is performed _X Means for detecting the maximum value of the quantity, and the detected outflow NO. _X NO based on local maximum _X An exhaust emission control system that determines the amount of accumulated sulfur in the absorbent. NO in the exhaust gas flowing into the exhaust passage of the internal combustion engine where the combustion is performed under a lean air-fuel ratio when the air-fuel ratio of the exhaust gas flowing in is lean _X NO when storing the reducing agent in the exhaust gas when the air-fuel ratio of the exhaust gas flowing in decreases _X NO is being reduced and stored _X NO decreases in quantity _X In an internal combustion engine with an absorbent, NO _X NO in exhaust gas flowing out from absorbent _X The amount of spillage NO _X NO to detect quantity _X NO placed in the exhaust passage downstream of the absorbent _X Sensor and NO _X The amount of accumulated sulfur, which is the amount of sulfur stored in the absorbent, is converted to the outflow NO. _X NOx purification apparatus comprising means for obtaining based on quantity, NO _X NO stored in the absorbent _X NO _X NO stored in the absorbent _X NO to reduce the amount of _X NO temporarily switching the air-fuel ratio of the exhaust gas flowing into the absorbent to rich _X Means for processing and NO _X Processing was done Sometimes NO _X NO flowing out of the absorbent _X Outflow NO that is the accumulated amount of _X Means for obtaining the integrated amount of the quantity, and the obtained outflow NO. _X NO based on the integrated amount _X An exhaust emission control system that determines the amount of accumulated sulfur in the absorbent. NO in the exhaust gas flowing into the exhaust passage of the internal combustion engine where the combustion is performed under a lean air-fuel ratio when the air-fuel ratio of the exhaust gas flowing in is lean _X NO when storing the reducing agent in the exhaust gas when the air-fuel ratio of the exhaust gas flowing in decreases _X NO is being reduced and stored _X NO decreases in quantity _X In an internal combustion engine with an absorbent, NO _X NO in exhaust gas flowing out from absorbent _X The amount of spillage NO _X NO to detect quantity _X NO placed in the exhaust passage downstream of the absorbent _X Sensor and NO _X The amount of accumulated sulfur, which is the amount of sulfur stored in the absorbent, is converted to the outflow NO. _X NOx purification apparatus comprising means for obtaining based on quantity, NO _X NO stored in the absorbent _X NO _X NO stored in the absorbent _X NO to reduce the amount of _X NO temporarily switching the air-fuel ratio of the exhaust gas flowing into the absorbent to rich _X Means for processing and NO _X Outflow NO when the air-fuel ratio of the exhaust gas flowing into the absorbent is kept lean _X NO so that the amount does not exceed the predetermined maximum allowable amount _X Means for setting a time interval at which processing is performed, and NO based on the controlled time interval _X An exhaust emission control device further comprising means for determining the amount of accumulated sulfur in the absorbent.

Images