JP3612636B2 - Vector control method for synchronous motor - Google Patents

Description

ここに、ここにＲは電機子抵抗を、Ｌは電機子インダクタンスを、ｓは微分演算子ｄ／ｄｔを示している。上式右辺第２項の信号は、電機子電流ｑ軸成分を、直列接続された電機子抵抗と電機子インダクタンスに流した際に発生する降下電圧を示しており、また右辺全体の信号は、電機子電圧ｑ軸成分からこの降下電圧を減じて得られた残電圧値を示している。また、このときの電機子鎖交磁束ｄ軸成分φｄは、直接検出することなく、電機子電流ｄ軸成分ｉｄを用いたつぎの式（２）に従い推定近似しようとするものである。
【数２】

なお、上式におけるΦは、回転子磁束を２軸直交座標系で表現した場合の最大値である。
【００１９】
ところで、同期電動機においては、内部状態において、一般に次式（３）に示す関係が成立している。
【数３】

ここに、φｍｑは回転子磁束φｍのｑ軸成分である。
【００２０】
さてここで、本発明の基本を示す式（１）を式（３）に用いると、ｓφｍｑ＝０なる結果が得られる。ｓφｍｑ＝０なる結果は、φｍｑ＝一定を意味し、回転子磁束が電源角周波数に同期して回転していることを示している。特に、式（２）に示したように、φｄを構成する回転子磁束分としてその最大値Φを利用する場合には、φｍｑ＝０が平衡状態の唯一の解となる。φｍｑ＝０は、とりもなおさず、回転子の磁束と同一方向をｄ軸に選定しこれと直交する軸をｑ軸に選定する回転ｄｑ座標系ための位相が確立されることを意味している。以上の説明より明らかなように、請求項３の本発明によれば、本質的に、回転子磁束と同一方向をｄ軸に選定しこれと直交する軸をｑ軸に選定する回転ｄｑ座標系の位相確立に必要な電源角周波数が、回転子の磁極位置情報を用いることなく、生成されると言う作用が得れる。更に付言するならば、回転子の磁極位置の推定が困難あるいは実質的に不可能な停止あるいは低速回転領域で使用しうる電源角周波数が生成されるという作用が得られる。
【００２１】
請求項３の本発明では、電機子電圧のｑ軸成分、電機子電流のｄ軸成分、ｑ軸成分がノイズ等の影響により正確な値が入手できない場合には、これらの近似推定値により電源角周波数を生成するようにしている。本発明は、回転ｄｑ座標系の位相確立に必要な本質的条件を直接満足させることを電源角周波数の生成原理としているのでいるので、ノイズ等の影響が強い場合にも、上記信号の近似推定値を利用することにより、現実的な電源角周波数を生成できると言う作用が得られる。
【００２２】
請求項４の本発明によれば、請求項３の発明におけるベクトル制御方法であって、ｄ磁束値に対し正の下限値を設定し、ｄ磁束値が該下限値より常時大きな値を取るようにするので、電源角周波数の生成を、残電圧値のｄ磁束値による除算という簡単な演算で実施する場合にも、ゼロあるいは極小値で除算を行うというゼロ割り現象を確実に回避することができるという作用が得られる。
【００２３】
請求項５の本発明によれば、請求項３の発明におけるベクトル制御方法であって、残電圧値の生成工程に、低域通過フィルタによるフィルタリング処理を行うので、残電圧値を与える式（１）の右辺が示すように電機子インダクタンスＬが関与する電圧生成には純粋微分が要求されるが、この純粋微分を回避することができると言う作用が得られる。ひいては、残電圧値が現実的な信号に対して不必要な過大値を示したり、ノイズ等の外乱に対し不都合な過激反応することを回避することができると言う作用が得られる。
【００２４】
請求項６の本発明によれば、回転子磁束と同一方向をｄ軸に選定しこれと直交する軸をｑ軸に選定する回転ｄｑ座標系に必要な位相が、高周波領域に限定されるが、生成されると言う作用が得れることを、先ずノイズ等を含まない正確な信号を用いて説明する。
【００２５】
固定ａｂ座標系上では、電機子電圧、電機子電流、電機子鎖交磁束の回転子磁束分を２次元のベクトル量として夫々ｖ、ｉ、φｍと表現するとき、つぎの関係が成立している。
【数４】

【００２６】
式（４）の右辺は、電機子電圧から、電機子電流を直列接続された電機子抵抗Ｒと電機子インダクタンスＬに流した際に発生する降下電圧値を減じることにより得られる信号を示している。式（４）で表現した信号に対し、例えば直流ゲインが１の低域通過フィルタＦ（ｓ）としてα／（ｓ＋α）を考え（１ーＦ（ｓ））／ｓ＝１／（ｓ＋α）なるフィルタを用意し、フィルタリング処理すると次式（５）に示す信号、すなわち残電圧値が得られる。
【数５】

【００２７】
式（５）の左辺が明確に示すように、上式右辺の残電圧値は、結果として回転子磁束φｍを低域遮断フィルタｓ／（ｓ＋α）でフィルタリング処理した信号そのものとなっている。このときの回転子磁束φｍに関してはつぎの関係式（６）が成立している。
【数６】

【００２８】
式（５）、式（６）より明白なように、式（５）右辺の信号を、回転子磁束の最大値Φで除算すれば、回転ｄｑ座標系に必要な位相（ｃｏｓθ，ｓｉｎθ）が、高周波領域に限定されるが、生成されるという作用が得られる。即ち、高周波領域に限定されるが、磁極位置検出器を用いることなく、電機子の電圧及び電流の情報から回転ｄｑ座標系に必要な位相を生成することができるという作用が得られる。
【００２９】
請求項６の本発明では、回転子磁束φｍに対しては、実質的に、低域遮断フィルタによるフィルタリングと同等な処理を行っているので、本発明の位相生成法によれば、低周波領域のオフセット、ドリフトに頑健であるという作用も得られる。
【００３０】
また、請求項６の本発明では、電機子電圧、電機子電流が高周波領域のノイズ等の影響により正確な値が入手できない場合には、これらの近似推定値により回転ｄｑ座標系に必要な位相を生成するようにしているので、高周波領域のノイズ等の影響が強い場合にも、上記信号の近似推定値を利用することにより、現実的な位相を生成できると言う作用が得られる。
【００３１】
請求項７の本発明によれば、請求項６の発明におけるベクトル制御方法であて、該残電圧値のノルム値を同定するので、回転子磁束の最大値Φが不明の場合にも、同定したノルム値を回転子磁束の最大値Φとして利用することができるので、回転ｄｑ座標系の位相を生成できるという作用が得られる。
【００３２】
【発明の実施の形態】
以下、図面を用いて、本発明の実施形態を詳細に説明する。本発明のベクトル制御方法を適用したベクトル制御装置と同期電動機の１実施形態例の基本的構造を図１に示す。図１における電動機等の１、２、３は、従来のベクトル制御方法を適用した図１２と同一である。同図の１２は、回転ｄｑ座標系上で、電機子電流をｄ軸成分とｑ軸成分に分割し制御する工程を司るベクトルコントローラの基本部であり、これも従来のベクトル制御方法による場合と同様である。図１では、同期電動機に磁極位置検出器は装着されておらず、これに代わって、回転ｄｑ座標系の位相を決定する工程を司る機器として、低周波領域位相生成器９、高周波領域位相生成器１０及び位相合成器１１が新たに用意されている。低周波位相生成器では、回転ｄｑ座標系の位相の低周波領域部分を生成している。一方、高周波領域位相生成器は、回転ｄｑ座標系の位相の高周波領域部分を生成している。位相合成器は、これらの２種の位相を合成して、最終的な位相（ｃｏｓθｆ，ｓｉｎθｆ）を生成し、ベクトルコントローラ基本部１２へ詳細にはベクトルコントローラ基本部内のベクトル回転器６ａ、６ｂへ出力している。このときの２種の位相の合成は、以下詳しく説明するように、周波数的加重平均により実行している。
【００３３】
図２は、図１における低周波領域位相生成器９、高周波領域位相生成器１０、位相合成器１１とベクトルコントローラ基本部１２との関係の詳細を示したものである。本実施形態例では、低周波領域位相生成器９には、電機子電流のｄ軸成分ｉｄ、ｑ軸成分ｉｑ、及び電機子電圧ｑ軸成分の近似推定値としてその指令値ｖｑ＊が、入力されている。また、高周波領域位相生成器１０には、電機子電流のａ，ｂ軸成分及び電機子電圧の近似推定値としてその指令値ｖ＊のａ、ｂ軸成分が、入力されている。
【００３４】
図３は、低周波領域位相生成器９の内部を示したものである。９ａは電源角周波数生成器を、９ｂは積分器を、９ｃは正弦信号発生器を、９ｄは直流ゲインが１の２個の低域通過フィルタＦ（ｓ）を示している。電源角周波数生成器で電源角周波数ω１を生成すると、これは積分器にて積分処理され電源角θ１として出力される。電源角は正弦信号発生器にて複数の正弦信号（ｃｏｓθ１、ｓｉｎθ１）に変換される。この複数の正弦信号は、直流ゲインが１の２個の低域通過フィルタ９ｄに各々入力される。低域通過フィルタ９ｄの役割は、正弦信号（ｃｏｓθ１、ｓｉｎθ１）に、フィルタＦ（ｓ）の周波数特性に応じた、低周波を中心とした周波数的な加重を付加することにある。こうして、低域通過フィルタ９ｄから低周波領域用の位相情報が最終的に出力される。直流ゲインが１の低域通過フィルタＦ（ｓ）として、例えば簡単な１次フィルタ（α／（ｓ＋α））を２個用意するならば（図３の９の出力はこの例を示している）、つぎの信号が、低周波領域用の位相情報として出力される。
【数７】

【００３５】
図４は、電源角周波数生成器９ａの内部の構成を示したものである。同図の９ａａは、電機子電圧ｑ軸成分指令値から電機子電流のｑ軸成分を直列接続された電機子抵抗と電機子インダクタンスに流した際に発生する降下電圧を減じた残電圧値を生成する残電圧生成部である。９ａｂは、電機子電流ｄ軸成分ｉｄと回転子磁束の最大値とから電機子鎖交磁束ｄ軸成分の磁束推定近似値を生成するｄ磁束生成部であり、９ａｃは、ｄ磁束生成部の出力信号と電源角周波数との積が上記残電圧生成部の出力信号と直接的に等しくなるように電源角周波数を決定する電源角周波数演算部である。
【００３６】
図５は、残電圧値生成部９ａａの１実施形態例を示したものである。図５の例では、低域通過フィルタｆ／（ｓ＋ｆ）によるフィルタリング処理を実施している点に特徴がある。残電圧生成部の構成は、基本的には式（１）の右辺に従っており、この右辺が示すように電機子インダクタンスＬが関与する電圧生成には理想的状況では純粋微分が要求されるが、図５の例は、低域通過フィルタｆ／（ｓ＋ｆ）によりこの純粋微分を回避していることを、明確に示している。
【００３７】
図６は、ｄ磁束生成部の１実施形態例の詳細を示したものである。ｄ磁束生成部は、基本的は式（２）に示した信号処理を行い、ｄ磁束値をその出力信号としている。この例では、理解を容易にするため、機能別に構成を分けた３段構成で、ｄ磁束生成部を構成している。ｄ磁束生成部第１段は基本部９ａｂａであり、式（２）に示した関係を忠実に具現化したものである。ｄ磁束生成部第２段９ａｂｂはバランス部であり、前述の残電圧生成部の出力信号に対しゲインと位相のバランスを保つことを目的に用意されている。したがって、この１例では、簡単には、残電圧生成部に導入した低域通過フィルタｆ／（ｓ＋ｆ）と同一のものを設置すればよい。ｄ磁束生成部第３段はリミッタ部９ａｂｃであり、これにより、ｄ磁束生成部の最終的な出力信号であるｄ磁束値が予め設定した正の下限値より、常時大きな値を取ることができる。
【００３８】
電源角周波数演算部９ａｃの働きは、ｄ磁束生成部の出力信号と電源角周波数との積が該残電圧生成部の出力信号と直接的に等しくなるように電源角周波数を決定することにあるが、本例では、残電圧生成部の出力信号をｄ磁束生成部の出力信号により除算すると言う最も簡単な決定法を利用している。除算による場合にも、前述のｄ磁束生成部においてｄ磁束値が予め設定した下限値より常時大きな値を取るようにしているので、ゼロあるいは極小値で除算を行うというゼロ割り現象を確実に回避することができる。
【００３９】
つぎに、高周波領域位相生成器１０の１実施形態について説明する。図７は、高周波領域位相生成器１０の本実施形態での詳細を示したものである。この高周波領域位相生成器内では、簡明のため、作用の説明で使用したフィルタを用い、式（４）、（５）、（６）に準じた処理を実施する例を示した。また、本実施形態では、固定ａｂ座標系の電機子電圧指令値ｖ＊と電機子電流情報ｉがベクトル信号として入力され、それぞれａ軸成分、ｂ軸成分の信号としてまとめられ、例えばａ軸成分の信号に関しては、まず次式に示した処理がフィルタ部１０ａで実施される。
【数８】

図８はフィルタ部１０ａにおけるフィルタリング処理のための具体的なフィルタの１実現例をブロックで示したものであり、このフィルタの出力信号が、残電圧値となる。式（８）および図８より明白なように、この処理は、電機子電圧の推定近似値の１つである電機子電圧指令値から、電機子電流を直列接続された電機子抵抗と電機子インダクタンスに流した際に発生する降下電圧値を減じることにより得られる信号を、直流ゲインが１の低域通過フィルタＦ（ｓ）＝α／（ｓ＋α）と（１ーＦ（ｓ））／ｓなる関係をもつフィルタ（１／（ｓ＋α））に通して残電圧値を生成している。
ａ軸成分を例に残電圧値生成の説明したが、フィルタ部１０ｂによるｂ軸成分の残電圧値生成はａ軸成分の場合と同一であるので、説明は省略する。
【００４０】
ａ軸成分、ｂ軸成分の両成分について残電圧値が生成されたならば、図７に示すように、この両信号に対し、演算部１０ｃにおいて回転子磁束の最大値Φで単に除算することにより、該残電圧値の位相を決定する。こうして決定された残電圧値の位相は、図７及び次式（９）に示すように、既に式（７）に双対の周波数的加重を付与したものとなっている。
【数９】

低周波遮断の周波数特性を加重として付与されたこの位相が、回転ｄｑ座標系の位相の高周波領域部分として使用される。
【００４１】
位相合成器１１には、低周波領域位相生成器９によって生成された位相（式（７）参照）が、また高周波領域位相生成器１０によって生成された位相（式（９）参照）が、入力されている。位相合成器の入力信号は既に周波数的な加重が付与されているので、位相合成器では、次式が示すように、これらの入力信号を各成分ごとに単純に加算することにより平均化し、回転ｄｑ座標系の最終的な位相ベクトル［ｃｏｓθｆ，ｓｉｎθｆ］を生成し、これをベクトル回転器６ａ，６ｂに出力している。
【数１０】

【００４２】
図９は、本発明を用いた上記実施形態のベクトル制御装置により、同期電動機を駆動制御した場合の実験結果の１例を示したものである。左上が電機子電流ｑ軸成分の指令値を、右上が生成された回転ｄｑ座標系用の最終的位相ベクトルの１つであるｃｏｓθｆを、左下が発生トルクを、右下が回転子の速度を、各々５秒間表示している。実験結果は、α＝２０と設計した１例であるが、力行・回生性能、１３００（ｒ／ｍ）の正転逆転を１．５秒という加減速性能、正確なトルク発生性能と言ったサーボの重要性能を示しており、これから容易に理解されるように、優れたサーボ性能が得られている。
【００４３】
以上、１実施形態を説明したが、上記の例に代わって、以下に示すような実施形態も可能であることを、低周波領域位相生成器９、高周波領域位相生成器１０、位相合成器１１に分けて、指摘しておく。
【００４４】
低周波領域位相生成器９に関しては、第１に、電源角周波数生成器９ａに使用する電流信号として上の例では電機子電流のｄ軸成分ｉｄ及びｑ軸成分ｉｑを用いたが、この近似推定値の１つとしてこれらの電流指令を用いてもよく、また、電機子電圧ｑ軸成分の信号として上の例ではその指令値ｖｑ＊を用いたが、当然のことながら、電圧の実測値あるいはこの他の推定近似値を用いてよいことを指摘しておく。
【００４５】
また、残電圧生成部９ａａに使用したフィルタの次数は１次である必要はなく、２次、３次の高次であってよい。また、このフィルタの第１義的目的は、純粋微分の回避であるので、これが回避できる他の方法・手段によっても差し支えないことを指摘しておく。
【００４６】
ｄ磁束生成部９ａｂに関しては、第１、２、３段に分けて構成する１例を示したが、第１段と第２段は順序を入れ替えることが可能である。また、ｄ磁束生成部を第１、２、３段のように分けることなく、一体的に構成するこも可能であることを指摘しておく。また、ｄ磁束生成部内のバランス部は、残電圧生成部で使用したフィルタに対し、両生成部出力信号間のゲインと位相を保つために導入されたものであり、この帯域が十分広い場合、あるいは電機子電流ｄ軸成分の指令値ｉｄ＊が常時一定の場合には、省略可能であることも指摘しておく。当然のことながら、残電圧生成部でフィルタを使用しない場合には、バランス部は必要ない。同期電動機のベクトル制御では、電機子電流ｄ軸成分の指令値ｉｄ＊を常時ゼロに設定する場合が多いが、この場合には、ｄ磁束生成部の出力信号は、回転子磁束の最大値Φそのものであり、回転子磁束の最大値Φが既知であればバランス部もリミッタ部も不要で、極めて簡単にｄ磁束生成部を構成することができる。
【００４７】
回転子磁束の最大値Φが未知の場合には、これを何等かの方法で同定し、ｄ磁束生成部９ａｂではこの同定値を回転子磁束の最大値Φに代わって使用すればよい。回転子磁束の最大値Φの同定方法の１例に関しては、図１０、１１に関連して、後に詳しく説明する。
【００４８】
残電圧生成部９ａａの出力信号に対し、スケーリング処理を実施する場合には、ｄ磁束生成部９ａｂの出力信号に対しても、同様なスケーリング処理が当然必要であることを指摘しておく。
【００４９】
電源角周波数演算部９ａｃにおける演算処理に関して、除算による方法を上に例示したが、除算を一切使用しない繰り返し的方法で電源角周波数を決定できることを指摘しておく。
【００５０】
低周波領域位相生成器９内の低域通過フィルタ９ｄは、直流ゲインが１で、かつ高周波領域位相生成器１０内のフィルタと周波数加重としての周波数特性に関して合致していれば、上の示した実施形態例のように１次フィルタに限定する必要はない。２次、３次のフィルタを利用してよいことを指摘しておく。
【００５１】
ｄ軸インダクタンスＬｄとｑ軸インダクタンスＬｑが異なる突極性を有する同期電動機に対しては、これまで説明における電機子電流のｄ軸成分に関連したインダクタンスＬをＬｄに、電機子電流のｑ軸成分に関連したインダクタンスＬをＬｑに置換すれば、これまでの説明してきた方法がそのまま成立することを指摘しておく。
【００５２】
つぎに、高周波領域位相生成器１０に関する他の実施形態について指摘しておく。高周波領域位相生成器１０で使用するフィルタとしては、Ｆ（ｓ）を直流ゲインが１の低域通過フィルタとするとき（１ーＦ（ｓ））／ｓと同様の周波数特性をもつフィルタであればよく、正確に（１ーＦ（ｓ））／ｓである必要はない。Ｆ（ｓ）に比較し十分に通過帯域が広くかつ直流ゲインが１の低域通過フィルタをＧ（ｓ）とするとき、高周波領域位相生成器１０で使用するフィルタとしてＧ（ｓ）（１ーＦ（ｓ））／ｓを用いてよいことを指摘しておく。
【００５３】
高周波領域位相生成器１０では、回転子磁束の最大値Φが未知の場合には、これを同定し、所要の高周波領域の位相を出力することが可能であることを指摘しておく。図１０は、この１実施形態をブロック図で示したものである。図１０と図９の違いは、新たにノルム同定器１０ｄが付加されている点にある。図１１は、このノルム同定器１０ｄの同定処理の内容を流れ図で示したものである。起動前にすなわちステップｓ１で先ずノルムの同定初期値を可能な限り適切に与えておく。起動後、ステップｓ２で１０ａ、１０ｂから残電圧値のａ軸成分、ｂ軸成分を取得する。シテップｓ３でこの取得信号の周波数を検査する。つぎのステップｓ４で検査した周波数が高周波領域に属するか否かを判定する。このときの高周波領域とは、高周波位相生成器が位相生成を担う周波数領域である。周波数が高周波領域に問題なく属すると判定された場合に限りつぎのステップｓ５へ進み、取得信号からノルム値を新たに計算し、これを用いて旧ノルム値を更新する。更新された最新のノルム同定値がつぎのステップｓ６で演算部１０ｃへ向け出力される。ステップｓ４で周波数が高周波領域に問題なく属さないと判定された場合には、旧ノルム値の更新は行われず、ステップｓ６では、最新のノルム同定値として、旧ノルム値がそのまま出力される。以下、ステップｓ２に戻り、同様な処理が繰り返される。図１０に示すように、演算部１０ｃでは、最新の同定値をノルム同定器１０ｄより受け取ると、これを用いて残電圧値のａ軸成分、ｂ軸成分を除算して、該残電圧値の位相を決定する。この残電圧値の位相は、これまでの説明で明白なように、回転ｄｑ座標系の位相の高周波領域部分として使用される。
【００５４】
本高周波領域位相生成器に図３ー６に示した低周波領域位相生成器を併用する実施形態では、ステップｓ３における周波素の領域判定に使用する周波数は、図３の電源角周波数生成器９ａの出力信号ω１を利用すればよい。また、本高周波領域位相生成器で同定したノルム値は、電源角周波数生成器９ａ内のｄ磁束生成部９ａｂで必要とされる回転子磁束の最大値Φが未知の場合には、この同定値として利用することができる。
【００５５】
図２を用いて説明した実施形態では、低周波領域用の位相決定方法を装置化した低周波領域位相生成器９として、図３ー６に示したものを使用した。また、高周波領域用の位相決定方法を装置化した高周波領域位相生成器１０として、図７、８、１０に示したものを使用した。本発明の基本構成を示した図１においては、低周波領域位相生成器９、高周波領域位相生成器１０としてはこれら２個の位相生成器いずれか１つまたは全部を、他のものと置換してよいことを指摘しておく。
【００５６】
図２を用いて説明した実施形態では、低周波領域位相生成器９、高周波領域位相生成器１０の内部で位相に周波数的加重を付加した。このため、位相合成器では、単に２種の加重付き位相を加算すれば、加重平均を実施できた。例示した実施形態以外の低周波領域位相生成器あるいは高周波領域位相生成器による場合には、位相合成器内で周波数加重を実施し、加重平均する方が合理的であり得る可能性があることを指摘しておく。
【００５７】
以上説明した低周波領域位相生成器９、高周波領域位相生成器１０、位相合成器１１の実現は、連続時間的な実現と同様に、離散時間的実現も可能であることを指摘しておく。離散時間的実現に際しては、ｚ演算子による離散時間表現が必要であるが、これには、例えば簡単には、これまでの説明で使用したｓ演算子をオイラー近似、台形近似等を用いてｚ演算子の表現に置換すればよい。離散時間的実現に際しては、ハードウェアー的実現のみならず、ソフトウェアー的実現も、当然ながら、可能であることを指摘しておく。特に、本発明の実行に要する演算量は、最近のマイクロプロセッサ等の演算素子によれば、何等の問題なくソフトウェアー的実現を達成することができる。
【００５８】
【発明の効果】
低速域と高速域との間での急激かつ不断の加速減速性能を必要とする同期電動機のベクトル制御方法においては、回転ｄｑ座標系の位相の決定は不可欠であり、従来よりこれに関連して回転子の磁極位置検出器が使用されており、この使用に起因する種々の問題が不可避的に残置されてきた。しかし、以上の説明より明白なように、本発明によれば、磁極位置検出器を一切必要としない、かつ低速域と高速域との間での急激かつ不断の加速減速性能を達成できるベクトル制御方法を実現できるので、従来の方法が不可避的に内臓していた磁極位置検出器に起因する種々の問題を解決でき、本発明の目的を十分に達成し得る優れた効果が得られる。
【００５９】
特に、請求項１の本発明によれば、回転子の磁極位置検出器を用いない場合にも、低速域と高速域との間での急激かつ不断の加速減速性能を必要とするサーボ性能を発揮し得るベクトル制御に必要な回転ｄｑ座標系の位相を生成できるので、この結果、回転子の磁極位置検出器を用いることなく、サーボ応用に必要な低速域と高速域との間での急激かつ不断の加速減速性能を発揮し得るベクトル制御を実現できるという効果が得られる。
【００６０】
特に、請求項２の本発明によれば、位相信号にフィルタリング処理を施すという簡単処理で位相信号に周波数的な加重を付加できるので、周波数的な加重平均を通じて最終位相を生成する請求項１のベクトル制御方法を簡単に実現できるという効果が得られる。
【００６１】
特に、請求項３の本発明によれば、回転ｄｑ座標系の位相確立に必要な電源角周波数が、回転子の磁極位置の推定が困難あるいは実質的に不可能な停止あるいは低速回転領域で使用しうる電源角周波数が生成されるという作用が得られるので、この作用の結果、磁極検出器を一切用いない低周波領域用のベクトル制御方法を実現できるという効果が得られる。
【００６２】
特に、請求項４の本発明によれば、電源角周波数の生成を、残電圧値にｄ磁束値による除算という簡単な演算で実施する場合にも、ゼロあるいは極小値で除算を行うというゼロ割り現象を確実に回避することができるという作用が得られるので、この結果、請求項３のベクトル制御方法を簡単に実現できるようになるという効果が得られる。
【００６３】
特に請求項５の本発明によれば、残電圧値の生成工程において、純粋微分を回避することができると言う作用が、ひいては、残電圧値が現実的な信号に対して不必要な過大値を示したり、ノイズ等の外乱に対し不都合な過激反応することを回避することができると言う作用が得られるので、この結果、請求項３のベクトル制御方法の実現に、ノイズ等の外乱に対しロバスト性を付与できると言う効果が得られる。
【００６４】
特に、請求項６の本発明によれば、回転子の磁極位置検出器を用いない場合にも、回転ｄｑ座標系に必要な位相が、高周波領域に限定されるが、生成されると言う作用が得れるので、この結果、回転子検出器を一切用いないで高周波領域用のベクトル制御方法を実現できるという効果が得られる。
【００６５】
特に、請求項７の本発明によれば、回転子磁束の最大値Φを正の有意の値で近似同定できるので、回転子磁束の最大値Φが不明の場合にも、請求項６のベクトル制御方法を実現できるという効果が得られる。
【図面の簡単な説明】
【図１】１実施形態に係わるベクトル制御装置の基本構成を示すブロック図
【図２】１実施形態に係わるベクトル制御装置の概略構成を示すブロック図
【図３】同ベクトル制御装置の低周波領域位相生成器内部の概略構成を示すブロック図
【図４】同低周波領域位相生成器内部の電源角周波数生成器の概略構成を示すブロック図
【図５】同電源角周波数生成器を構成する残電圧生成部の概略構成を示すブロック図
【図６】同電源角周波数生成器を構成するｄ磁束生成部の概略構成を示すブロック図
【図７】図２における高周波領域位相生成器の概略構成を示すブロック図
【図８】同高周波領域位相生成器内部のフィルタ部の構成を示すブロック図
【図９】実施形態に係わる実験結果の１例
【図１０】図２における高周波領域位相生成器の他の実施形態による概略構成を示すブロック図
【図１１】同高周波領域位相生成器内のノルム同定部における処理内容を示す流れ図
【図１２】従来のベクトル制御装置の概略構成を示すブロック図
【符号の説明】
１ 同期電動機
２ 磁極位置検出器
３ 電力変換器
４ 電流検出器
５ａ ３相２相変換器
５ｂ ２相３相変換器
６ａ ベクトル回転器
６ｂ ベクトル回転器
７ 電流制御器
８ 正弦信号発生器
９ 低周波領域位相生成器
９ａ 電源角周波数生成器
９ａａ 残電圧生成部
９ａｂ ｄ磁束生成部
９ａｂａ 基本部
９ａｂｂ バランス部
９ａｂｃ リミッタ部
９ａｃ 電源角周波数演算部
９ｂ 積分器
９ｃ 正弦信号発生器
９ｄ 低域通過フィルタ
１０ 高周波領域位相生成器
１０ａ フィルタ部
１０ｂ フィルタ部
１０ｃ 演算部
１０ｄ ノルム同定部
１１ 位相合成器
１２ ベクトルコントローラ基本部[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a method for controlling a synchronous motor, and relates to a vector control method that can be applied to a servo that requires rapid and constant acceleration / deceleration performance between a low speed range and a high speed range. In particular, the present invention relates to a vector control method in which the same direction as the magnetic flux of the rotor is selected as the base axis of the rotation orthogonal coordinate system and the rotor magnetic flux position detector is not required for establishing the vector control.
[0002]
[Prior art]
When using a synchronous motor in a servo application that requires rapid and constant acceleration / deceleration performance between a low speed range and a high speed range, as a control method for this purpose, the state of the synchronous motor is changed to zero in the rotor flux. There is known a vector control method for capturing and controlling on a rotating dq coordinate system synchronized with.
[0003]
FIG. 12 is a block diagram schematically showing a representative example when this vector control method is implemented as an apparatus and mounted on a synchronous motor. 1 is a synchronous motor, 2 is a rotor magnetic pole position detector, 3 is a power converter, 4 is a current detector, 5a and 5b are a 3-phase 2-phase converter and a 2-phase 3-phase converter, respectively. 6a and 6b are vector rotators, 7 is a current controller, and 8 is a sine signal generator. In FIG. 12, various devices from 5a, 5b to 8 constitute a vector control device.
[0004]
In particular, the three types of devices 5a, 5b, 6a, 6b, and 7 mainly constitute a means for controlling the armature current by dividing it into a d-axis component and a q-axis component on the rotating dq coordinate system. It is. The three-phase current detected by the current detector 4 is converted into a two-phase current on the fixed ab coordinate system by the three-phase two-phase converter 5a, and then the two-phase current id, converted to iq. It is sent to the current controller 7. The current controller generates command voltages vd * and vq * on the dq coordinate system so that the conversion currents id and iq follow the command values id * and iq * commanded from the outside, and sends them to the vector rotator 6b. send. In 6b, this two-phase signal is converted into a two-phase command voltage in the fixed ab coordinate system and sent to the two-phase / three-phase converter 5b. In 5 b, the two-phase signal is converted into a three-phase command voltage and output as a command value to the power converter 3. The power converter 3 generates electric power according to the command, applies it to the synchronous motor 1, and drives it.
[0005]
The sine signal generator 8 constitutes means for determining the phase of the rotating dq coordinate system. That is, when the magnetic pole position information is sent from the magnetic pole position detector 2, the sine signal generator 8 generates a plurality of sine signals (cos, sin signals) using this signal, and outputs it as phase information of the rotating dq coordinate system. To the vector rotators 6a and 6b. In the figure, for the sake of simplicity, a plurality of sine signals are captured as one phase vector and represented by a single thick line. In the following, also in other drawings, the bold lines indicate vector signal lines. As will be more readily appreciated that phase information in the form of a sine signal is provided to the vector rotator 6b leading to the power converter 3, this phase information will determine the power supply angular frequency.
[0006]
[Problems to be solved by the invention]
In the above conventional vector control method, as clearly shown in the figure, the magnetic pole position information of the rotor is indispensable for determining the phase of the rotating dq coordinate system. For this reason, a magnetic pole position detection system centered on a magnetic pole position detector represented by an absolute encoder must be constructed. However, the magnetic pole position detection system including the magnetic pole position detector is extremely fragile as compared with the synchronous motor main body, and the reliability of the entire drive system including the synchronous motor is remarkably lowered. For example, the magnetic pole position detector is extremely weak against the impact received by the motor, and the signal of the magnetic pole position detector is generally a few volts, and special consideration must be given to noise during transmission. Consideration was also required for safe wiring of power lines and signal lines. In addition, countermeasures are required when a weak magnetic pole position detection system is abnormal. The implementation of such a magnetic pole position detection system is disadvantageous in terms of cost. In addition, when the detector is mounted, the axial dimension of the motor is increased, which is disadvantageous in the installation of the motor.
[0007]
For this reason, methods for estimating the magnetic pole position without using a magnetic pole position detector such as an encoder have been studied. However, these methods basically use a single method for the rotor magnetic pole position or the rotating dq axis. This is to determine the phase of the coordinates, and this is one of the main causes, inducing several of the problems such as detection magnetic pole position detection accuracy, detection range, detection stability, low speed range and high speed Applications to servos that require rapid and constant acceleration / deceleration performance with the region are not progressing. Vector control is originally intended for servo applications with rapid acceleration / deceleration, and various problems resulting from the use of the magnetic pole position detector have been left as problems to be solved by vector control.
[0008]
The present invention has been made based on the background described above, and its object is to solve the problems caused by the use of the magnetic pole position detector, without requiring any magnetic pole position detector, and to achieve high acceleration / deceleration. An object of the present invention is to provide a vector control method capable of exhibiting performance.
[0009]
[Means for Solving the Problems]
In order to achieve the above object, the invention according to claim 1 is directed to the armature current on the rotating dq coordinate system in which the same direction as the magnetic flux of the rotor is selected as the d axis and the axis orthogonal thereto is selected as the q axis. A synchronous motor vector control method comprising a step of dividing and controlling a d-axis component and a q-axis component, and a step of determining a phase of a rotating dq coordinate system, wherein the phase of the rotating dq coordinate system is determined. The phase determination method for the low frequency region and the phase determination method for the high frequency region are used to generate phases, and the phase generated by the phase determination method for the low frequency region and the high frequency The phase generated by the region phase determination method is weighted and averaged in frequency to obtain the phase of the rotated dq coordinate system.
[0010]
The invention according to claim 2 is the vector control method for the synchronous motor according to claim 1, wherein the frequency characteristic of the low-pass filter F (s) having a DC gain of 1 is generated by the phase determination method for the low-frequency region. The frequency weighted average is performed using the frequency characteristic of the low-frequency cutoff filter (1-F (s)) as the main weight of the phase generated by the phase determination method for the high frequency region. It is a feature.
In the invention of claim 3, the armature current is converted into the d-axis component and the q-axis component on the rotating dq coordinate system in which the same direction as the magnetic flux of the rotor is selected as the d-axis and the axis orthogonal thereto is selected as the q-axis. A vector control method for a synchronous motor having a step of dividing and controlling and a step of determining a phase of a rotating dq coordinate system, wherein the step of determining the phase of the rotating dq coordinate system includes: From the product of the d-axis component of the alternating magnetic flux or the d-magnetic flux value expressed by the estimated magnetic flux value and the q-axis component of the armature voltage or the estimated voltage value of the armature, the q-axis component of the armature current or the current estimate thereof. In order to directly equalize the residual voltage value obtained by reducing the drop voltage generated when the approximate value is connected to the armature resistor connected in series and the armature inductance, the residual voltage value and the d magnetic flux The power supply angular frequency is determined by calculation processing with the value. And the determined integration and trigonometric function processed value of the power supply angular frequency and is characterized in that for use in the low frequency region portion of the rotating dq coordinate system phase.
[0011]
The invention according to claim 4 is the vector control method for the synchronous motor according to claim 3, wherein a positive lower limit value is set for the d magnetic flux value, and the d magnetic flux value is always larger than the lower limit value. It is characterized by that.
[0012]
A fifth aspect of the present invention is the vector control method for a synchronous motor according to the third aspect, wherein a filtering process using a low-pass filter is performed in the step of generating the residual voltage value.
[0013]
According to the sixth aspect of the present invention, the armature current is converted into the d-axis component and the q-axis component on the rotating dq coordinate system in which the same direction as the magnetic flux of the rotor is selected as the d-axis and the orthogonal axis is selected as the q-axis. A vector control method for a synchronous motor having a step of dividing and controlling, and a step of determining a phase of a rotating dq coordinate system, wherein the phase of the rotating dq coordinate system is determined in the step of determining the phase of the rotating dq coordinate system. The armature voltage and current are expressed on a fixed ab coordinate system as a reference of the armature, and the armature current and the estimated approximate value thereof are connected in series from the armature voltage or the estimated approximate value thereof. When the signal obtained by reducing the drop voltage value generated when the signal is passed through is a low-pass filter with a DC gain of 1 (1−F (s)) / s, the same frequency as F (s) Pass through a filter with characteristics A pressure value is generated, a phase of the residual voltage value is determined from an a-axis component and a b-axis component of the residual voltage value, and the determined phase is used for a high frequency region portion of the phase of the rotating dq coordinate system. It is what.
[0014]
The invention according to claim 7 is the vector control method for a synchronous motor according to claim 6, wherein the norm of the residual voltage value is newly identified only when the residual voltage value belongs to a high frequency region, and the residual voltage value The phase of the residual voltage value is determined by dividing the a-axis component and the b-axis component by the latest identification value of the norm.
[0015]
Next, the operation of the present invention will be described. According to the present invention of claim 1, the phase determination method for the low frequency region suitable for the phase determination in the low frequency region can be used, and the phase determination method for the high frequency region suitable for the phase determination in the high frequency region can be used. These phase determination methods can be used. Moreover, according to the present invention, two types of phases each determined by a suitable method can be combined in frequency continuously by a frequency weighted average, so that this final phase is used as the phase of the rotating dq coordinate system. By doing so, it is possible to generate a phase that can be used for vector control for servo applications that require rapid and constant acceleration / deceleration performance between a low speed region and a high speed region.
[0016]
According to the second aspect of the present invention, in the vector control method according to the first aspect, the frequency characteristic of the low-pass filter F (s) having a DC gain of 1 and low frequency can be used as a main weight for frequency weighted averaging. Since the frequency characteristics of the band cut-off filter (1-F (s)) are used, the frequency weighting can be simplified by simply filtering each phase generated by the two phase determination methods with these filters. The effect that it can be added is obtained. In addition, the frequency weighting can be easily changed by simply changing the frequency characteristics of the filter. For example, when the cutoff characteristic of the filter is set to an extremely low frequency, two types of weights for the extremely low frequency region and the low, medium, and high frequency regions can be easily set.
[0017]
According to the third aspect of the present invention, the power supply angular frequency for establishing the phase necessary for the rotating dq coordinate system in which the same direction as the rotor magnetic flux is selected as the d axis and the axis orthogonal thereto is selected as the q axis is: The fact that the action of being generated without using the magnetic pole position information of the rotor can be obtained will be described first using an accurate signal that does not include noise or the like.
[0018]
If the power supply angular frequency is expressed as ω1, the d-axis component of the armature flux linkage as φd, the q-axis component of the armature voltage as vq, the d-axis component of the armature current as id, and the q-axis component as iq, In the present invention, the signal represented by the right side of the equation (1) is directly calculated by an operation such as division by the signal φd so that the following relational expression (1) including no rotor magnetic pole position signal is satisfied. The power source angular frequency ω1 is to be generated.
[Expression 1]

Here, R represents an armature resistance, L represents an armature inductance, and s represents a differential operator d / dt. The signal in the second term on the right side of the above equation indicates the voltage drop that occurs when the armature current q-axis component is passed through the armature resistor and the armature inductance connected in series, and the signal on the entire right side is: The residual voltage value obtained by subtracting this drop voltage from the armature voltage q-axis component is shown. Further, the armature flux linkage d-axis component φd at this time is to be estimated and approximated according to the following equation (2) using the armature current d-axis component id without being directly detected.
[Expression 2]

In the above equation, Φ is the maximum value when the rotor magnetic flux is expressed in a biaxial orthogonal coordinate system.
[0019]
By the way, in the synchronous motor, the relationship represented by the following equation (3) is generally established in the internal state.
[Equation 3]

Here, φmq is a q-axis component of the rotor magnetic flux φm.
[0020]
Here, when Expression (1) indicating the basis of the present invention is used for Expression (3), a result of sφmq = 0 is obtained. The result of sφmq = 0 means that φmq = constant, and indicates that the rotor magnetic flux rotates in synchronization with the power supply angular frequency. In particular, as shown in Equation (2), when the maximum value Φ is used as the rotor magnetic flux component constituting φd, φmq = 0 is the only solution in the equilibrium state. φmq = 0 means that, for the time being, the phase for the rotating dq coordinate system is established in which the same direction as the magnetic flux of the rotor is selected as the d axis and the axis orthogonal to this is selected as the q axis. Yes. As is apparent from the above description, according to the present invention of claim 3, a rotating dq coordinate system is essentially selected in which the same direction as the rotor magnetic flux is selected as the d-axis and the axis perpendicular thereto is selected as the q-axis. The power source angular frequency necessary for establishing the phase can be generated without using the magnetic pole position information of the rotor. Further, it is possible to obtain an effect that a power source angular frequency that can be used in a stop or low-speed rotation region in which estimation of the magnetic pole position of the rotor is difficult or substantially impossible is generated.
[0021]
According to the third aspect of the present invention, when accurate values of the q-axis component of the armature voltage, the d-axis component of the armature current, and the q-axis component cannot be obtained due to the influence of noise or the like, An angular frequency is generated. Since the present invention uses the generation principle of the power supply angular frequency to directly satisfy the essential condition necessary for establishing the phase of the rotating dq coordinate system, even if the influence of noise or the like is strong, the approximate estimation of the signal is performed. By using the value, an effect that a realistic power angular frequency can be generated is obtained.
[0022]
According to the fourth aspect of the present invention, in the vector control method according to the third aspect of the invention, a positive lower limit value is set for the d magnetic flux value so that the d magnetic flux value is always larger than the lower limit value. Therefore, even when the generation of the power supply angular frequency is performed by a simple calculation such as division of the residual voltage value by the d magnetic flux value, it is possible to reliably avoid the zero division phenomenon of dividing by zero or the minimum value. The effect that it can be obtained.
[0023]
According to the fifth aspect of the present invention, in the vector control method according to the third aspect of the present invention, since the filtering process using the low-pass filter is performed in the residual voltage value generation step, the equation (1) As shown on the right side of (), a pure differentiation is required for the voltage generation involving the armature inductance L, and an effect that this pure differentiation can be avoided is obtained. As a result, it is possible to avoid an unnecessary excessive reaction with respect to a disturbance such as noise or a residual voltage value indicating an unnecessary excessive value with respect to a realistic signal.
[0024]
According to the present invention of claim 6, the phase required for the rotating dq coordinate system in which the same direction as the rotor magnetic flux is selected as the d-axis and the axis orthogonal thereto is selected as the q-axis is limited to the high frequency region. The fact that the action of being generated can be obtained will be described first using an accurate signal that does not include noise or the like.
[0025]
On the fixed ab coordinate system, when the armature voltage, the armature current, and the rotor flux component of the armature linkage flux are expressed as two-dimensional vector quantities as v, i, and φm, respectively, the following relations are established. Yes.
[Expression 4]

[0026]
The right side of equation (4) shows a signal obtained by subtracting the voltage drop generated when the armature current is passed through the armature resistor R and the armature inductance L connected in series from the armature voltage. Yes. For the signal expressed by Equation (4), for example, α / (s + α) is considered as a low-pass filter F (s) having a DC gain of 1 (1−F (s)) / s = 1 / (s + α). When a filter is prepared and subjected to filtering processing, a signal represented by the following equation (5), that is, a residual voltage value is obtained.
[Equation 5]

[0027]
As clearly shown on the left side of Equation (5), the residual voltage value on the right side of the above equation is the signal itself obtained by filtering the rotor magnetic flux φm with the low-frequency cutoff filter s / (s + α). With respect to the rotor magnetic flux φm at this time, the following relational expression (6) is established.
[Formula 6]

[0028]
As is clear from the equations (5) and (6), if the signal on the right side of the equation (5) is divided by the maximum value Φ of the rotor magnetic flux, the phase (cos θ, sin θ) required for the rotating dq coordinate system is obtained. Although it is limited to the high frequency region, the effect of being generated can be obtained. That is, although it is limited to a high frequency region, an effect is obtained that a phase necessary for the rotating dq coordinate system can be generated from information on the voltage and current of the armature without using a magnetic pole position detector.
[0029]
In the present invention of claim 6, since the rotor magnetic flux φm is processed substantially equivalent to the filtering by the low-frequency cutoff filter, according to the phase generation method of the present invention, the low-frequency region It is also possible to obtain an effect that is robust against offset and drift.
[0030]
Further, in the present invention of claim 6, when the armature voltage and the armature current cannot obtain accurate values due to the influence of noise in the high frequency region, the phase necessary for the rotating dq coordinate system is calculated based on these approximate estimates. Therefore, even when the influence of high-frequency noise or the like is strong, an effect that a realistic phase can be generated by using the approximate estimated value of the signal can be obtained.
[0031]
According to the present invention of claim 7, in the vector control method of the invention of claim 6, the norm value of the residual voltage value is identified, so that the identification is performed even when the maximum value Φ of the rotor magnetic flux is unknown. Since the norm value can be used as the maximum value Φ of the rotor magnetic flux, there is an effect that the phase of the rotating dq coordinate system can be generated.
[0032]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 shows the basic structure of an embodiment of a vector control apparatus and a synchronous motor to which the vector control method of the present invention is applied. 1, 2 and 3 of the electric motor etc. in FIG. 1 are the same as FIG. 12 which applied the conventional vector control method. 12 in the figure is a basic part of a vector controller that controls the process of dividing the armature current into the d-axis component and the q-axis component on the rotating dq coordinate system. This is also the case with the conventional vector control method. It is the same. In FIG. 1, the magnetic pole position detector is not mounted on the synchronous motor. Instead, the low frequency region phase generator 9 and the high frequency region phase generator are used as devices for controlling the phase of the rotating dq coordinate system. A device 10 and a phase synthesizer 11 are newly prepared. The low frequency phase generator generates a low frequency region portion of the phase of the rotating dq coordinate system. On the other hand, the high frequency region phase generator generates a high frequency region portion of the phase of the rotating dq coordinate system. The phase synthesizer synthesizes these two types of phases to generate a final phase (cos θf, sin θf), and to the vector controller basic unit 12 in detail, to the vector rotators 6a and 6b in the vector controller basic unit. Output. The synthesis of the two types of phases at this time is performed by frequency weighted averaging, as will be described in detail below.
[0033]
FIG. 2 shows details of the relationship between the low frequency region phase generator 9, the high frequency region phase generator 10, the phase synthesizer 11 and the vector controller basic unit 12 in FIG. In this embodiment, the low frequency region phase generator 9 receives the command value vq * as an approximate estimated value of the d-axis component id, the q-axis component iq, and the armature voltage q-axis component of the armature current. Has been. Further, the a and b axis components of the command value v * are input to the high frequency region phase generator 10 as approximate estimated values of the a and b axis components of the armature current and the armature voltage.
[0034]
FIG. 3 shows the inside of the low frequency region phase generator 9. Reference numeral 9a denotes a power source angular frequency generator, 9b denotes an integrator, 9c denotes a sine signal generator, and 9d denotes two low-pass filters F (s) having a DC gain of 1. When the power supply angular frequency generator generates the power supply angular frequency ω1, it is integrated by the integrator and output as the power supply angle θ1. The power supply angle is converted into a plurality of sine signals (cos θ1, sin θ1) by a sine signal generator. The plurality of sine signals are respectively input to two low-pass filters 9d having a DC gain of 1. The role of the low-pass filter 9d is to add frequency weighting centered on the low frequency to the sine signal (cos θ1, sin θ1) according to the frequency characteristics of the filter F (s). In this way, phase information for the low frequency region is finally output from the low pass filter 9d. For example, if two simple primary filters (α / (s + α)) are prepared as the low-pass filter F (s) having a DC gain of 1 (the output of 9 in FIG. 3 shows this example). The next signal is output as phase information for the low frequency region.
[Expression 7]

[0035]
FIG. 4 shows the internal configuration of the power supply angular frequency generator 9a. 9aa in the figure shows a residual voltage value obtained by subtracting the voltage drop generated when the q-axis component of the armature current is passed through the armature resistor and the armature inductance connected in series from the armature voltage q-axis component command value. It is a residual voltage generation part to generate. 9ab is a d magnetic flux generation unit that generates a magnetic flux estimation approximate value of the armature flux linkage d-axis component from the armature current d-axis component id and the maximum value of the rotor magnetic flux, and 9ac is the d magnetic flux generation unit. A power supply angular frequency calculation unit that determines a power supply angular frequency so that a product of the output signal and the power supply angular frequency is directly equal to the output signal of the residual voltage generation unit.
[0036]
FIG. 5 shows an embodiment of the remaining voltage value generation unit 9aa. The example of FIG. 5 is characterized in that a filtering process using a low-pass filter f / (s + f) is performed. The configuration of the residual voltage generation unit basically follows the right side of Equation (1), and as shown on the right side, the voltage generation involving the armature inductance L requires a pure differentiation in an ideal situation. The example of FIG. 5 clearly shows that this pure differentiation is avoided by the low-pass filter f / (s + f).
[0037]
FIG. 6 shows details of one embodiment of the d magnetic flux generator. The d magnetic flux generator basically performs the signal processing shown in Equation (2), and uses the d magnetic flux value as its output signal. In this example, in order to facilitate understanding, the d magnetic flux generation unit is configured in a three-stage configuration in which the configuration is divided according to function. The first stage of the d magnetic flux generation unit is a basic unit 9aba, which faithfully embodies the relationship shown in Expression (2). The d magnetic flux generation unit second stage 9abb is a balance unit, and is prepared for the purpose of maintaining a balance between gain and phase with respect to the output signal of the residual voltage generation unit. Therefore, in this example, the same filter as the low-pass filter f / (s + f) introduced into the residual voltage generation unit may be simply installed. The third stage of the d magnetic flux generation unit is a limiter unit 9abc, which allows the d magnetic flux value, which is the final output signal of the d magnetic flux generation unit, to always take a value larger than a preset positive lower limit value. .
[0038]
The function of the power supply angular frequency calculation unit 9ac is to determine the power supply angular frequency so that the product of the output signal of the d magnetic flux generation unit and the power supply angular frequency is directly equal to the output signal of the residual voltage generation unit. However, in this example, the simplest determination method of dividing the output signal of the residual voltage generation unit by the output signal of the d magnetic flux generation unit is used. Even in the case of division, since the d magnetic flux value is always larger than the preset lower limit value in the above-described d magnetic flux generation section, the zero division phenomenon of dividing by zero or a minimum value is surely avoided. can do.
[0039]
Next, an embodiment of the high frequency domain phase generator 10 will be described. FIG. 7 shows details of the high-frequency region phase generator 10 in this embodiment. In this high-frequency region phase generator, for the sake of simplicity, an example has been shown in which processing according to the equations (4), (5), and (6) is performed using the filter used in the description of the operation. In the present embodiment, the armature voltage command value v * and the armature current information i in the fixed ab coordinate system are input as vector signals and are combined as signals of an a-axis component and a b-axis component, respectively. With respect to the above signal, first, the processing shown in the following equation is performed by the filter unit 10a.
[Equation 8]

FIG. 8 is a block diagram showing a specific implementation example of a filter for filtering processing in the filter unit 10a, and an output signal of this filter becomes a residual voltage value. As is clear from the equation (8) and FIG. 8, this processing is performed by using the armature voltage command value, which is one of the estimated approximate values of the armature voltage, and the armature resistance and the armature in which the armature current is connected in series. A signal obtained by subtracting the voltage drop generated when the current flows through the inductance is represented by a low-pass filter F (s) = α / (s + α) and (1−F (s)) / s with a DC gain of 1. The residual voltage value is generated through a filter (1 / (s + α)) having the following relationship.
Although the residual voltage value generation has been described by taking the a-axis component as an example, the generation of the residual voltage value of the b-axis component by the filter unit 10b is the same as that in the case of the a-axis component, and thus the description thereof is omitted.
[0040]
If the residual voltage values are generated for both the a-axis component and the b-axis component, as shown in FIG. 7, the two signals are simply divided by the maximum value Φ of the rotor magnetic flux in the arithmetic unit 10c. To determine the phase of the residual voltage value. The phase of the residual voltage value thus determined has already been obtained by adding dual frequency weighting to equation (7) as shown in FIG. 7 and the following equation (9).
[Equation 9]

This phase provided with the frequency characteristic of the low frequency cutoff as a weight is used as a high frequency region portion of the phase of the rotating dq coordinate system.
[0041]
The phase synthesizer 11 receives the phase generated by the low frequency domain phase generator 9 (see equation (7)) and the phase generated by the high frequency domain phase generator 10 (see equation (9)). Has been. Since the input signal of the phase synthesizer has already been weighted in frequency, the phase synthesizer averages and rotates these input signals by simply adding them for each component as shown in the following equation. The final phase vector [cos θf, sin θf] of the dq coordinate system is generated and output to the vector rotators 6a and 6b.
[Expression 10]

[0042]
FIG. 9 shows an example of an experimental result when the synchronous motor is driven and controlled by the vector control apparatus of the above embodiment using the present invention. The upper left is the command value of the armature current q-axis component, the upper right is cos θf, which is one of the final phase vectors for the generated rotation dq coordinate system, the lower left is the generated torque, and the lower right is the rotor speed. Each is displayed for 5 seconds. The result of the experiment is an example designed with α = 20, but the power running / regenerative performance, 1300 (r / m) forward / reverse rotation 1.5 seconds acceleration / deceleration performance, accurate torque generation servo As can be easily understood, excellent servo performance is obtained.
[0043]
In the above, one embodiment has been described. However, in place of the above example, the following embodiments are also possible. The low frequency domain phase generator 9, the high frequency domain phase generator 10, and the phase synthesizer 11 are described below. I will point it out separately.
[0044]
Regarding the low frequency region phase generator 9, first, the d-axis component id and the q-axis component iq of the armature current are used as the current signal used for the power supply angular frequency generator 9a in the above example. These current commands may be used as one of the estimated values, and the command value vq * is used in the above example as the signal of the armature voltage q-axis component. Alternatively, it should be pointed out that other estimated approximation values may be used.
[0045]
Further, the order of the filter used in the residual voltage generation unit 9aa need not be the first order, and may be the second order or the third order higher order. Also, it should be pointed out that the primary purpose of this filter is to avoid pure differentiation, so that other methods and means that can avoid this may be used.
[0046]
As for the d magnetic flux generation unit 9ab, an example in which the first magnetic flux generation unit 9ab is divided into the first, second, and third stages is shown, but the order of the first stage and the second stage can be changed. Further, it should be pointed out that the d magnetic flux generation unit can be integrally configured without being divided into the first, second, and third stages. Moreover, the balance part in d magnetic flux generation part was introduced in order to maintain the gain and phase between both generation part output signals to the filter used in the residual voltage generation part, and when this band is sufficiently wide, Alternatively, it should be pointed out that it can be omitted if the command value id * of the armature current d-axis component is always constant. As a matter of course, when no filter is used in the residual voltage generation unit, the balance unit is not necessary. In the vector control of the synchronous motor, the command value id * of the armature current d-axis component is often set to zero at all times. In this case, the output signal of the d magnetic flux generator is the maximum value Φ of the rotor magnetic flux. In other words, if the maximum value Φ of the rotor magnetic flux is known, neither the balance unit nor the limiter unit is necessary, and the d magnetic flux generation unit can be configured very easily.
[0047]
When the maximum value Φ of the rotor magnetic flux is unknown, this is identified by some method, and the d magnetic flux generation unit 9ab may use this identification value instead of the maximum value Φ of the rotor magnetic flux. One example of a method for identifying the maximum value Φ of the rotor magnetic flux will be described in detail later with reference to FIGS.
[0048]
It should be pointed out that when the scaling process is performed on the output signal of the residual voltage generator 9aa, the same scaling process is naturally necessary for the output signal of the d magnetic flux generator 9ab.
[0049]
Regarding the calculation processing in the power supply angular frequency calculation unit 9ac, the method by division is exemplified above, but it should be pointed out that the power supply angular frequency can be determined by an iterative method without using any division.
[0050]
The low-pass filter 9d in the low-frequency region phase generator 9 is shown above if the DC gain is 1 and the filter in the high-frequency region phase generator 10 matches the frequency characteristics as frequency weighting. It is not necessary to limit to the first order filter as in the embodiment. Note that second and third order filters may be used.
[0051]
For a synchronous motor having a saliency in which the d-axis inductance Ld and the q-axis inductance Lq are different, the inductance L related to the d-axis component of the armature current in the description so far is set to Ld, and the q-axis component of the armature current is set to It should be pointed out that if the related inductance L is replaced with Lq, the method described so far can be applied as it is.
[0052]
Next, other embodiments relating to the high frequency domain phase generator 10 will be pointed out. The filter used in the high-frequency region phase generator 10 is a filter having a frequency characteristic similar to (1−F (s)) / s when F (s) is a low-pass filter having a DC gain of 1. It does not have to be exactly (1−F (s)) / s. When a low-pass filter having a sufficiently wide pass band and a DC gain of 1 as G (s) compared with F (s) is G (s), G (s) (1− Note that F (s)) / s may be used.
[0053]
It is pointed out that the high-frequency region phase generator 10 can identify the maximum value Φ of the rotor magnetic flux, and can output the phase of the required high-frequency region. FIG. 10 is a block diagram showing this one embodiment. The difference between FIG. 10 and FIG. 9 is that a norm identifier 10d is newly added. FIG. 11 is a flowchart showing the contents of the identification process of the norm identifier 10d. Before starting, that is, in step s1, first, the norm identification initial value is given as appropriately as possible. After startup, the a-axis component and the b-axis component of the remaining voltage value are acquired from 10a and 10b in step s2. In step s3, the frequency of the acquired signal is checked. It is determined whether or not the frequency inspected in the next step s4 belongs to the high frequency region. The high frequency region at this time is a frequency region in which the high frequency phase generator is responsible for phase generation. Only when it is determined that the frequency belongs to the high frequency region without any problem, the process proceeds to the next step s5, a new norm value is calculated from the acquired signal, and the old norm value is updated using this. The updated new norm identification value is output to the calculation unit 10c in the next step s6. If it is determined in step s4 that the frequency does not belong to the high frequency region without any problem, the old norm value is not updated, and in step s6, the old norm value is output as it is as the latest norm identification value. Thereafter, the process returns to step s2, and similar processing is repeated. As shown in FIG. 10, when the latest identification value is received from the norm identifier 10d, the arithmetic unit 10c divides the a-axis component and the b-axis component of the remaining voltage value using this, and calculates the remaining voltage value. Determine the phase. The phase of this residual voltage value is used as the high frequency region portion of the phase of the rotating dq coordinate system, as is apparent from the above description.
[0054]
In the embodiment in which the low frequency domain phase generator shown in FIGS. 3-6 is used in combination with the high frequency domain phase generator, the frequency used for the frequency domain domain determination in step s3 is the power supply angular frequency generator 9a in FIG. The output signal ω1 may be used. The norm value identified by the high frequency region phase generator is the identification value when the maximum value Φ of the rotor magnetic flux required by the d magnetic flux generator 9ab in the power supply angular frequency generator 9a is unknown. Can be used as
[0055]
In the embodiment described with reference to FIG. 2, the low frequency region phase generator 9 in which the low frequency region phase determination method is implemented is the one shown in FIGS. 3-6. Further, as the high-frequency region phase generator 10 in which the phase determination method for the high-frequency region is implemented, the one shown in FIGS. In FIG. 1 showing the basic configuration of the present invention, one or all of these two phase generators are replaced with other ones as the low frequency domain phase generator 9 and the high frequency domain phase generator 10. Point out that it is okay.
[0056]
In the embodiment described with reference to FIG. 2, frequency weighting is added to the phase inside the low frequency region phase generator 9 and the high frequency region phase generator 10. For this reason, the phase synthesizer can perform weighted averaging simply by adding two kinds of weighted phases. In the case of using a low-frequency domain phase generator or a high-frequency domain phase generator other than the illustrated embodiment, it may be more reasonable to perform frequency weighting in the phase synthesizer and perform weighted averaging. Please point out.
[0057]
It should be pointed out that the low-frequency domain phase generator 9, the high-frequency domain phase generator 10, and the phase synthesizer 11 described above can be realized in discrete time as well as in continuous time. In the discrete time realization, a discrete time expression by the z operator is necessary. For this purpose, for example, the s operator used in the above description is expressed by using Euler approximation, trapezoidal approximation, or the like. Replace with an operator expression. It should be pointed out that, in discrete time implementation, not only hardware implementation but also software implementation is possible. In particular, the amount of computation required to execute the present invention can be achieved by software according to a recent computing element such as a microprocessor without any problem.
[0058]
【The invention's effect】
In a vector control method for a synchronous motor that requires rapid and constant acceleration / deceleration performance between a low speed region and a high speed region, it is essential to determine the phase of the rotating dq coordinate system. Rotor magnetic pole position detectors have been used, and various problems resulting from this use have inevitably remained. However, as is clear from the above description, according to the present invention, the vector control that does not require any magnetic pole position detector and can achieve rapid and constant acceleration / deceleration performance between the low speed range and the high speed range. Since the method can be realized, it is possible to solve various problems caused by the magnetic pole position detector that is unavoidably incorporated in the conventional method, and an excellent effect that can sufficiently achieve the object of the present invention is obtained.
[0059]
In particular, according to the first aspect of the present invention, even when the magnetic pole position detector of the rotor is not used, the servo performance that requires rapid and constant acceleration / deceleration performance between the low speed range and the high speed range is achieved. As a result, the phase of the rotating dq coordinate system necessary for vector control that can be exhibited can be generated. As a result, a rapid change between the low speed range and the high speed range required for servo application can be achieved without using the magnetic pole position detector of the rotor. And the effect that the vector control which can exhibit a constant acceleration deceleration performance is realizable is acquired.
[0060]
In particular, according to the present invention of claim 2, since the frequency weight can be added to the phase signal by a simple process of filtering the phase signal, the final phase is generated through the frequency weighted average. There is an effect that the vector control method can be easily realized.
[0061]
In particular, according to the present invention of claim 3, the power source angular frequency necessary for establishing the phase of the rotating dq coordinate system is used in a stop or low-speed rotation region where it is difficult or practically impossible to estimate the magnetic pole position of the rotor. Since an effect that a power source angular frequency that can be generated is obtained is obtained, as a result of this action, an effect that a vector control method for a low frequency region without using a magnetic pole detector can be realized.
[0062]
In particular, according to the present invention of claim 4, even when the generation of the power source angular frequency is performed by a simple operation of dividing the residual voltage value by the d magnetic flux value, the zero division of dividing by zero or the minimum value is performed. Since the effect that the phenomenon can be surely avoided is obtained, the result is that the vector control method of claim 3 can be easily realized.
[0063]
In particular, according to the present invention of claim 5, in the residual voltage value generating step, the effect that the pure differentiation can be avoided, and consequently the residual voltage value is an excessive value that is unnecessary for a real signal. Or an adverse reaction to disturbances such as noise can be avoided. As a result, the vector control method according to claim 3 can be realized with respect to disturbances such as noise. An effect that robustness can be imparted is obtained.
[0064]
In particular, according to the present invention of claim 6, even when the magnetic pole position detector of the rotor is not used, the phase required for the rotating dq coordinate system is limited to the high frequency region but is generated. As a result, it is possible to obtain a vector control method for the high frequency region without using any rotor detector.
[0065]
In particular, according to the present invention of claim 7, since the maximum value Φ of the rotor magnetic flux can be approximately identified with a positive significant value, the vector of claim 6 can be used even when the maximum value Φ of the rotor magnetic flux is unknown. The effect that a control method is realizable is acquired.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a basic configuration of a vector control apparatus according to an embodiment.
FIG. 2 is a block diagram showing a schematic configuration of a vector control apparatus according to one embodiment.
FIG. 3 is a block diagram showing a schematic configuration inside a low frequency region phase generator of the vector control device;
FIG. 4 is a block diagram showing a schematic configuration of a power supply angular frequency generator inside the low frequency region phase generator.
FIG. 5 is a block diagram showing a schematic configuration of a residual voltage generation unit constituting the power supply angular frequency generator.
FIG. 6 is a block diagram showing a schematic configuration of a d magnetic flux generation unit constituting the power source angular frequency generator.
7 is a block diagram showing a schematic configuration of a high-frequency region phase generator in FIG.
FIG. 8 is a block diagram showing a configuration of a filter unit in the high frequency region phase generator
FIG. 9 shows an example of experimental results according to the embodiment.
10 is a block diagram showing a schematic configuration according to another embodiment of the high-frequency region phase generator in FIG.
FIG. 11 is a flowchart showing processing contents in a norm identification unit in the high frequency region phase generator;
FIG. 12 is a block diagram showing a schematic configuration of a conventional vector control apparatus.
[Explanation of symbols]
1 Synchronous motor
2 Magnetic pole position detector
3 Power converter
4 Current detector
5a 3 phase 2 phase converter
5b 2 phase 3 phase converter
6a vector rotator
6b vector rotator
7 Current controller
8 Sine signal generator
9 Low frequency phase generator
9a Power supply angular frequency generator
9aa residual voltage generator
9ab d magnetic flux generator
9aba basic part
9abb balance
9abc limiter
9ac Power supply angular frequency calculator
9b integrator
9c sine signal generator
9d low-pass filter
10 High-frequency domain phase generator
10a Filter section
10b Filter section
10c arithmetic unit
10d norm identification part
11 Phase synthesizer
12 Vector controller basic part

Claims (7)

On the rotating dq coordinate system in which the same direction as the rotor magnetic flux is selected as the d-axis and the axis orthogonal to this is selected as the q-axis, the armature current is divided and controlled into a d-axis component and a q-axis component; A vector control method for a synchronous motor having a step of determining a phase of a rotating dq coordinate system,
In the step of determining the phase of the rotating dq coordinate system, two types of phase determination methods, a phase determination method for a low frequency region and a phase determination method for a high frequency region, are used to generate phases, respectively, A synchronous motor characterized in that the phase generated by the phase determination method of (1) and the phase generated by the phase determination method for the high frequency region are weighted and averaged in frequency to obtain the phase of the rotating dq coordinate system. Vector control method. The frequency characteristic of the low-pass filter F (s) having a DC gain of 1 is the main weight of the phase generated by the phase determination method for the low-frequency region, and the frequency of the low-frequency cutoff filter (1−F (s)) 2. The vector control method for a synchronous motor according to claim 1, wherein the frequency weighted average is performed by using the characteristic as a main weight of the phase generated by the phase determination method for the high frequency region. On the rotating dq coordinate system in which the same direction as the rotor magnetic flux is selected as the d-axis and the axis orthogonal to this is selected as the q-axis, the armature current is divided and controlled into a d-axis component and a q-axis component; A vector control method for a synchronous motor having a step of determining a phase of a rotating dq coordinate system,
In the step of determining the phase of the rotation dq coordinate system, the product of the power source angular frequency and the d-axis component of the armature linkage flux or the d-magnetic flux value expressed by the estimated magnetic flux approximation value and the q-axis of the armature voltage The residual voltage obtained by subtracting the voltage drop generated when the q-axis component of the armature current or the estimated current value of the armature current is passed through the armature resistor and the armature inductance connected in series from the component or the estimated voltage value thereof. In order to directly equalize the voltage value, the power supply angular frequency is determined by calculating the residual voltage value and the d magnetic flux value, and the determined integral of the power supply angular frequency and the trigonometric function processing value are rotated. A vector control method for a synchronous motor, wherein the vector control method is used for a low frequency region portion of a phase of a dq coordinate system. 4. The vector control method for a synchronous motor according to claim 3, wherein a positive lower limit value is set for the d magnetic flux value, and the d magnetic flux value is always larger than the lower limit value. 4. The vector control method for a synchronous motor according to claim 3, wherein a filtering process using a low-pass filter is performed in the step of generating the residual voltage value. On the rotating dq coordinate system in which the same direction as the rotor magnetic flux is selected as the d-axis and the axis orthogonal to this is selected as the q-axis, the armature current is divided and controlled into a d-axis component and a q-axis component; A vector control method for a synchronous motor having a step of determining a phase of a rotating dq coordinate system,
In the step of determining the phase of the rotating dq coordinate system, the voltage and current of the armature are expressed on a fixed ab coordinate system that is a reference for determining the phase of the rotating dq coordinate system, and from the armature voltage or an estimated approximate value thereof, A signal obtained by subtracting a voltage drop generated when an armature current or an estimated approximate value thereof is passed through an armature resistor and an armature inductance connected in series, and F (s) is a low DC gain. When a band-pass filter is used, a residual voltage value is generated through a filter having a frequency characteristic similar to (1−F (s)) / s, and the residual voltage is calculated from the a-axis component and the b-axis component of the residual voltage value. A vector control method for a synchronous motor, wherein a phase of a value is determined and the determined phase is used for a high frequency region portion of a phase of the rotating dq coordinate system. Only when the residual voltage value belongs to the high frequency region, the norm of the residual voltage value is newly identified, and the a-axis component and the b-axis component of the residual voltage value are divided by the latest identification value of the norm. 7. The vector control method for a synchronous motor according to claim 6, wherein the phase of the voltage value is determined.

Images