JP2013160121A - Air amount measuring device of internal combustion engine and method of measuring air amount - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve the problems that are caused by an error occurring because an average air amount detected is influenced by the size and frequency of intake pulsation and the error may cause a response delay as a frequency may go higher and the output of a higher frequency sensor can't be correctly traced in the incorporation of a predetermined sampling timing if a heat ray type air quantity sensor output is incorporated at a predetermined sampling timing and is converted into an air quantity.SOLUTION: The size and pulsation frequency of a pulsation amplitude ratio are found from the intake pulsation of an internal combustion engine, a correction amount that corrects a frequency response of the heat ray type air amount sensor is found from the pulsation amplitude ratio and pulsation frequency, a correction amount that corrects a sampling error is found, an air amount detected from a corrected amount in response to the intake pulsation is calculated by operating correction thereby finding a final air amount.

Description

  The present invention relates to an air amount measuring device and an air amount measuring method for an internal combustion engine that calculates an air amount based on a detection signal from an air amount detecting device, and in particular, can accurately measure an air amount over a wide rotation region. The present invention relates to an air amount measuring device and an air amount measuring method for an internal combustion engine.

  2. Description of the Related Art Conventionally, in a control device for an internal combustion engine, in order to detect the intake air amount of the internal combustion engine, an air amount detection device is disposed in the intake pipe of the internal combustion engine, and a detection signal detected by the air amount detection device is calculated to calculate the intake air. The amount is obtained, and the amount of fuel injected from the fuel injection valve is controlled using this amount of intake air.

  Reducing harmful substances (for example, carbon monoxide, hydrocarbons, etc.) in the exhaust gas of internal combustion engines has become an important issue, and the intake air pressure can be reduced by using an intake pipe pressure sensor that has been used in the past. The method of measuring air volume indirectly as a volumetric flow rate is disadvantageous for reducing harmful substances in exhaust gas, and recently, the output signal of an air volume detector that can measure mass flow rate has been used. It is coming.

  The detection signal of the air amount detection device capable of measuring the mass flow rate is sampled at a predetermined time timing, and the intake air amount is obtained by digital calculation. This is an important technique for reducing harmful substances in gas.

  Currently, a hot-wire air quantity sensor is widely used as an air quantity detection device capable of measuring a mass flow rate. A detection signal of the air quantity sensor is a voltage corresponding to the air flow rate based on a signal from a heating resistor. A voltage signal that changes the value and a frequency signal that changes the cycle of the output pulse according to the air flow rate are often used.

  On the other hand, further improvement in fuel efficiency is required, and a variable valve timing mechanism that adjusts the opening / closing phase of the intake valve and exhaust valve, a variable lift mechanism that adjusts the lift of the intake valve, and a method of introducing a large amount of EGR gas It has been proposed to achieve both the output of the internal combustion engine and the fuel consumption by adopting the engine.

  By the way, in a multi-cylinder internal combustion engine, an intake pulsation is generated in the intake pipe due to the reciprocating motion of the piston. Therefore, even in a steady state, the detection signal of the hot-wire air quantity sensor fluctuates in synchronization with the rotational speed and the intake pulsation.

  In particular, a hot-wire air quantity sensor arranged upstream of the throttle valve is easily affected by intake pulsation when the throttle valve is operated to the fully open side. It is known to cause errors.

  For this reason, a method for correcting the air amount with respect to the intake pulsation has been proposed as disclosed in Japanese Patent No. 4466309 (Japanese Patent No. 4466309) and Japanese Patent Application Laid-Open No. 2001-5090 (Patent Document 2).

  As described above, in recent internal combustion engines, the influence of the intake valve pulsation is further expanded or the generation region is changed due to the adoption of a variable valve timing mechanism, a variable lift mechanism, or the introduction of a large amount of EGR gas. The degree tends to be complicated.

Registration gazette of Japanese Patent No. 4466309 JP 2001-5090 A

  Also in the method of correcting the intake air amount described in Patent Document 1 and Patent Document 2, when calculating the pulsation rate of intake air, the correction amount is based on the intake pipe pressure, the throttle valve operation amount, or the internal combustion engine speed. Is a method of calculating.

  Here, in the hot-wire air amount sensor, the current value flowing through the heating resistor arranged in the intake air flow to be measured increases when the intake air amount is large, and conversely decreases when the intake air amount is small. Thus, a bridge circuit is configured, and a voltage signal or a frequency signal obtained by voltage-frequency conversion is taken out as an air amount signal by a current flowing through the heating resistor.

  Since the hot wire air quantity sensor uses the heat balance of the heating resistor, the detection accuracy is affected by the magnitude and frequency of the pulsation when it is affected by the intake pulsation, especially as the output of the hot wire air quantity sensor. Has a frequency response characteristic in which a response delay occurs in a higher frequency region. Therefore, when the output characteristic has a non-linear characteristic, the detected output average value and the true air amount average value do not coincide with each other, and an error occurs.

  On the other hand, when the detection signal of the hot-wire air amount sensor is taken at a predetermined sampling timing (for example, every 2 ms) and converted into the air amount, there is little error in the region where the intake pulsation does not occur. The average detected air amount is affected by the size and frequency, and an error occurs.

  For example, since the intake pulsation is caused by the reciprocating motion of the piston, the frequency changes depending on the number of revolutions of the internal combustion engine. As a primary component, in the case of a 4-cylinder internal combustion engine, two pulsations occur in one rotation of the crankshaft. To do.

  That is, in the low rotation speed region, the time required for 1/2 rotation is long, so the number of samplings (parameter) at each predetermined sampling timing increases, and there is a probability that the peak of the output signal of the hot-wire air quantity sensor can be sampled. The error does not increase so much as it becomes higher, but conversely, at high rotation speeds, the time taken per 1/2 rotation is short, so the number of samplings (parameter) at each predetermined sampling timing is reduced, and the hot-wire air quantity sensor The error increases as the probability of sampling the peak of the output signal decreases.

  In addition, depending on the combination of the variable valve timing mechanism and a large amount of EGR, depending on the operating range, not only the primary component but also the frequency of the 0.5th, 1.5th, and secondary components are added to the pulsation. However, the pulsation is not always generated simply by the rotational speed of the internal combustion engine.

  When intake pulsation is taken into account due to these factors, it becomes more difficult to correctly trace the output value of the hot-wire air quantity sensor as the pulsation increases in the high-frequency region, and this is a true phenomenon called aliasing. Since it becomes equal to a state where a deviation from the air amount average value occurs, there is a problem that the error is enlarged.

  The feature of the present invention is that the magnitude of the pulsation amplitude ratio and the pulsation frequency are obtained from the intake air pulsation of the internal combustion engine, the correction amount for correcting the frequency response of the hot-wire air quantity sensor is obtained from the pulsation amplitude ratio and the pulsation frequency, and sampling is performed. The correction amount for correcting the error is obtained, and the final air amount is obtained by correcting and calculating the air amount detected from the correction amount corresponding to the intake pulsation.

  According to the present invention, the amount of air can be obtained in consideration of the influence of the intake pulsation over a wide rotation region, so that the intake air amount can be accurately calculated regardless of the magnitude or frequency change of the intake pulsation. It will be.

1 is a system configuration diagram showing an overall configuration of an internal combustion engine. It is a block diagram which shows the structure of a control unit. It is a schematic diagram which shows the outline | summary of the internal process in the case of performing the conventional air amount calculating method with a control unit. It is a characteristic view which shows the relationship between the air quantity of a hot wire type air quantity sensor, and an output signal. It is a characteristic view which shows the relationship between air quantity and intake air pulsation. It is a characteristic view which shows the relationship between the magnitude | size of a frequency and pulsation, and air quantity ratio. It is a characteristic view which shows the relationship between the pulsation by sampling timing, a frequency, and an error. It is a schematic diagram which shows the outline | summary of the internal process in the case of performing with the control unit the air quantity calculating method which becomes one Example of this invention. It is explanatory drawing which shows the relationship between the pulsation state of a low frequency area | region, and an air amount calculation process. It is explanatory drawing which shows the relationship between the pulsation state of a high frequency area | region, and an air amount calculation process. It is explanatory drawing explaining the correction coefficient used for the air quantity calculating method which becomes one Example of this invention. It is a control flowchart figure which shows the flowchart which implements the air quantity calculating method which becomes one Example of this invention.

  First, a control apparatus for an internal combustion engine, which is a premise in an embodiment of the present invention, and an outline of its operation will be described.

  FIG. 1 shows a configuration of a so-called MPI (multi-point fuel injection) type internal combustion engine. Hereinafter, an MPI type internal combustion engine will be described as an embodiment of the present invention, but the present invention can also be applied to various types of internal combustion engines such as a type of internal combustion engine in which fuel is directly injected into a cylinder.

  The intake air amount of the internal combustion engine is measured by a hot-wire air amount sensor 10 provided at the outlet of the air cleaner 11. The intake air passes through the intake pipe 12 connected to the air cleaner 11 and the electric throttle body 15 having the throttle valve 15a for adjusting the intake air amount and enters the collector 106. Thereafter, the intake air is distributed to the intake branch pipe 101 of the internal combustion engine 100 and is taken into the cylinder 103c of each cylinder.

  Reference numeral 103b is a cylinder block forming the cylinder 103c. The intake valve and the exhaust valve of the cylinder 103c are provided with a variable valve timing mechanism 101 and a variable valve timing mechanism 102, and the opening / closing phases of the intake valve and the exhaust valve can be adjusted.

  Further, the fuel supplied to the internal combustion engine 100 is sucked and pressurized from the fuel tank 107 by the fuel pump 108, adjusted to a constant pressure by the pressure regulator 106, and then injected from the injector 110 provided in the intake branch pipe 101. The

  In the cylinder 103 c of the internal combustion engine 100, the high voltage generated by the ignition coil 111 is applied to the spark plug 112 to ignite an air-fuel mixture in the intake branch pipe 101 to burn the fuel.

  The exhaust gas after burning in the cylinder 103c of each cylinder passes through the exhaust pipe 104, is purified by the three-way catalyst 105, and is then discharged out of the internal combustion engine.

  Various signals representing the state of the internal combustion engine are input to a control unit 200 that executes control of the internal combustion engine. FIG. 1 shows an intake air temperature sensor 14 and an intake air amount sensor 10 (here, a hot wire type). An air amount sensor), a throttle sensor 13 for detecting the opening degree of the throttle valve 15a, an accelerator sensor 116 for detecting an accelerator pedal depression amount, a water temperature sensor 17 for detecting a coolant temperature, and a crank angle sensor for detecting a crank angle. 18, a knock sensor 113, oxygen sensors 114 and 115 that measure upstream and downstream oxygen concentrations of the three-way catalyst 105, and the like.

  FIG. 2 shows the overall configuration of the control unit 200. The control unit 200 includes sensors necessary for controlling the internal combustion engine, such as a crank angle sensor 18 for detecting the rotational speed and phase of the internal combustion engine, a cam angle. Sensor 16, intake air temperature sensor 14 for detecting the temperature of air taken into internal combustion engine 100, water temperature sensor 17 for detecting the coolant temperature of the internal combustion engine, and atmospheric pressure sensor 121 for detecting atmospheric pressure Signals such as a throttle sensor 13 for detecting the throttle valve opening, an air-fuel ratio sensor 114 for detecting the oxygen concentration in the exhaust gas, a starter switch 117, and a hot-wire air amount sensor 10 are input.

  Detection signals from these sensors are input to an input processing circuit 201 in the control unit 200, and are processed separately as those detected by an A / D converter as analog inputs and those detected at a high / low level. Is done.

  The CPU 203 executes predetermined digital arithmetic processing by a program stored in the ROM 202 in the CPU, outputs control signals for various actuators necessary for controlling the internal combustion engine from the arithmetic result, and outputs the input processing circuit 201. It has a function of controlling various actuators via a circuit.

  For example, each injector 110 of the internal combustion engine has a mass flow rate measured from a detection signal of the hot-wire air amount sensor 10, a rotational speed of the internal combustion engine measured from the crank angle sensor 18, and a water temperature detected from the water temperature sensor 17. The correction amount corresponding to the air-fuel ratio state detected by the air-fuel ratio sensor 114 is added, the fuel injection amount is calculated, and finally output to the injector 110 as the drive pulse width.

  Similarly, an ignition signal for controlling the energization timing to the ignition coil 111 necessary for combustion of the internal combustion engine, a signal to the electric throttle 15 for controlling the throttle valve opening degree, and the like. These calculation results are stored in the RAM 204 or the EEPROM 205 which is a nonvolatile RAM.

  Next, the outline of the hot-wire air quantity sensor 10 will be described. The hot-wire air quantity sensor mainly includes a heating resistor arranged in an air flow to be measured, and a current value flowing through the heating resistor. Has a bridge circuit configured to increase when the amount of intake air is large, and conversely decrease when the amount of intake air is small, and takes out the amount of air as a voltage signal by the current flowing through the heating resistor.

  A voltage signal corresponding to the amount of air is output as a voltage value, or converted into a frequency signal by a voltage-frequency conversion circuit and output.

  FIG. 3 shows a calculation method for obtaining the air amount based on the output signal of the hot-wire air amount sensor 10 according to a conventional method, and this calculation is executed by internal processing of the control unit 200.

  When the detection signal from the hot-wire air amount sensor 10 is input to the CPU 203 via the LSI 201, the A / D converter 300 built in the CPU 203 sets the voltage of the signal from the hot-wire air amount sensor 10 to 0V to 5V. The voltage information Vu is converted into a digital value determined by the resolution of the A / D converter within the range.

  For example, assuming that the resolution of the A / D converter 300 is 10 bits and the voltage per bit recognized by the CPU 203 is equivalent to 5 mV, the digital value is 200 when A / D conversion is performed on 1.0 V.

  The voltage information Vu is given to the air amount conversion table 302. The air amount conversion table 302 has an air amount stored in advance corresponding to the voltage information Vu, and is subjected to a search interpolation operation to be converted into a detected air amount. The Hereinafter, description will be made assuming that the detected air amount indicates the air amount converted by the air amount conversion table.

  Although not shown in the figure, when the signal is input as a frequency signal subjected to voltage-frequency conversion, the period of the signal is measured at the port input of the CPU 203, and the value converted from the period or the period to the frequency becomes the input, and the air volume The conversion table is subjected to a search interpolation calculation from a value stored in advance according to the period or frequency, and converted into a detected air amount.

  The air amount signal Q obtained from the air amount conversion table 302 is then subjected to digital filter processing 303 for the purpose of removing high frequency component noise to calculate the detected air amount Qc.

  In the fuel injection pulse width calculation 400, the detected air amount Qc is divided by the rotational speed 305 of the internal combustion engine separately calculated from the signal of the crank angle sensor 18 so as to correspond to the amount of air sucked into the cylinder, and each correction calculation is performed. After that, a time TOUT for injecting the fuel is calculated, and the injector 110 operates to supply the fuel to the internal combustion engine.

  FIG. 4 shows the relationship between the intake air amount and the output signal of a general hot-wire air amount sensor 10. The output signal frequency is low when the intake air amount is small, and the output is high when the intake air amount is large. This is a characteristic curve having a non-linear relationship in which the frequency of the signal to be increased becomes higher.

The reason why the non-linearity characteristic is used is that the following equation called the King equation is mainly used as the air amount Q when the detection signal from the heating resistor is converted into the air amount.
Ih2 ・ Rh = (α + β ・ √Q) ・ (Th−Ta)
Where Ih is the current value of the heating resistor, Rh is the resistance value of the heating resistor, Th is the surface temperature of the heating resistor, Ta is the air temperature, Q is the air flow rate, and α and β are constants determined by the specifications of the heating resistor. It is.

  Generally, since the current value Ih of the heating resistor is controlled so that (Th−Ta) becomes constant, the amount of air is detected by converting it to a voltage value V by a voltage drop of the resistor. The value V is a quartic function expression. For this reason, when converting into an air flow rate, the curvature of the quartic curve, that is, the relationship between the output and the air amount becomes nonlinear.

  As for the characteristic curve, since the output of the hot-wire air amount sensor 10 is set in accordance with the required air amount of the internal combustion engine, the relationship between the frequency and the air amount or the relationship between the voltage and the air amount is reversed or linear. Although there is a relationship case, only the conversion table 302 changes for the arithmetic processing.

  Next, FIG. 5 shows the relationship between the detected air amount based on the output signal of the hot-wire air amount sensor 10 and the intake pulsation. When the horizontal axis is time T (s), (a) has a low frequency. That is, the true air amount when the cycle is long and the pulsation is large is indicated by RL1, and the detected air amount calculated from the output of the hot-wire air amount sensor 10 is indicated by SG1. Similarly, (b) indicates the true air amount when the frequency is high, that is, the cycle is short and the pulsation is large, RL2, and the detected air amount calculated from the output of the hot-wire air amount sensor 10 is SG2.

  As can be seen from this figure, in (a), even if the pulsation is large, the cycle is long, so the difference between the detected air amount SG1 obtained from the output of the hot-wire air amount sensor 10 and the true air amount RL1 is small. On the other hand, in (b), when the pulsation is large, the detected air amount SG2 obtained from the output of the hot-wire air amount sensor 10 traces the amplitude of the pulsation (the detected air amount is detected in the true air amount RL1) with respect to the true air amount RL2. Amount SG2 does not match).

  Since this uses the heat balance of the heating resistor of the hot-wire air quantity sensor 10, when affected by the intake pulsation, the detection accuracy is affected by the magnitude and frequency of the pulsation, and the pulsation is large. This is due to having a frequency response characteristic in which a response delay occurs as the frequency increases.

  Here, in the case where the output characteristic has a non-linear curve, the situation in which the average value of the detected air amount detected when pulsation occurs does not coincide with the true air amount average value and an error occurs will be described with reference to FIG. This will be described in more detail.

  In FIG. 4, if the horizontal axis is assumed to be a true air amount and a sine fluctuation that pulsates as indicated by a dotted line Qr occurs in the detection signal, the true average air amount is QaV1 indicated by a two-dot chain line. is there.

  On the other hand, since the output of the hot-wire air amount sensor 10 is detected as non-linear fluctuations, if the waveform indicated by the dotted line Sg1 on the vertical axis is output, the average value when converted to the air amount is indicated by the two-dot chain line described above. A return error does not occur in QaV1.

However, it has the frequency response characteristics as described in FIG. 5, and when the amplitude of the waveform changes to a waveform as indicated by the solid line Sg2, the average value takes the value indicated by the alternate long and short dash line and is the average when converted to the air amount. Since the value takes the value indicated by QaV2, the amount of air indicated by the difference QEr1 between QaV1 and QaV2 becomes a minus error.
FIG. 6 shows the relationship between the air amount ratio and the pulsation frequency, with the ratio obtained by dividing the detected air amount obtained from the sensor output by the true air amount as the air amount ratio.

  In FIG. 6, when the detected air amount obtained from the sensor output and the true air amount coincide with each other, the air amount ratio is 1.0, the solid line Cf1 indicates that the pulsation is small, and the dotted line Cf2 indicates that the pulsation is large. Is shown.

  In particular, when the frequency is 10 Hz or higher, the higher the frequency, the smaller the air amount ratio is less than 1.0, that is, the air amount error obtained from the output of the hot-wire air amount sensor 10 becomes negative, and the larger the pulsation, the larger the negative error. ing.

  For this reason, it can be seen that the error with respect to the frequency response characteristics as described above can be reduced if correction is performed with a correction amount having opposite characteristics according to the magnitude of the pulsation and the pulsation frequency.

  Therefore, one feature is that if the detected air amount is corrected by a correction amount determined by the magnitude of the pulsation of the intake air and the change of the pulsation frequency, the error due to the frequency response characteristic can be reduced and the detection accuracy can be improved.

  Next, FIG. 7 shows an error in the average value of the detected air amount depending on the magnitude and frequency of the intake pulsation when the output of the hot-wire air amount sensor 10 is taken at a predetermined sampling timing (for example, every 2 ms) and converted into the air amount. Is shown to occur.

In FIG. 7, the horizontal axis represents the pulsation frequency, and the vertical axis represents the error. Here, the error is obtained by the following equation.
Error (%) = {(Average value of detected air amount at 2 ms sampling timing) − (True air amount average value)} / (True air amount average value)
FIG. 7 shows the relationship between the pulsation frequency and the magnitude of the pulsation by obtaining this error in accordance with the change in frequency. The solid line Cnf1 shows the case where the pulsation is small, and the dotted line Cnf2 shows the case where the pulsation is large. Yes.

  As shown in FIG. 7, the larger the pulsation and the higher the frequency, the smaller the error in the air amount obtained from the output of the hot-wire air amount sensor 10, and the larger the error.

  For this reason, it can be understood that the error with respect to the sampling timing as described above can be reduced if correction is performed with a correction amount having reverse characteristics according to the magnitude of the pulsation and the pulsation frequency.

  Therefore, another feature is that if the detected air amount is corrected by the correction amount determined by the magnitude of the pulsation of the intake air and the change of the pulsation frequency, the error due to the sampling timing can be reduced and the detection accuracy can be improved.

  Here, in the case of a four-cylinder internal combustion engine, the frequency of pulsation may be substituted for the number of revolutions of the internal combustion engine because pulsation occurs in one cycle of the crankshaft of the internal combustion engine. However, depending on the combination of the variable valve timing mechanism and the introduction of the mass EGR system, depending on the operating range of the internal combustion engine, not only the primary component but also the frequencies such as the 0.5th order component, the 1.5th order component, and the second order component In some cases, pulsation is taken into account, and pulsation is not necessarily generated simply depending on the rotational speed of the internal combustion engine. Therefore, it is desirable to obtain the pulsation frequency from the actual detection value.

  Based on the above findings, an air amount calculation method that takes into account intake pulsation and fluctuations in its frequency according to an embodiment of the present invention will be described in detail.

  A method for suppressing an error caused by the magnitude or frequency of the intake pulsation when correcting the detection error of the hot-wire air amount sensor 10 caused by the intake pulsation by a correction coefficient corresponding to the intake pulsation will be described with reference to FIGS. explain.

  FIG. 8 shows an outline of internal processing when the control unit 200 executes the air amount calculation method using the output signal of the hot-wire air amount sensor 10 according to an embodiment of the present invention.

  This internal processing represents a control function or an arithmetic function as a block in the drawing, and this internal processing block receives an output Vu from the A / D converter 300 as an input and a sampling timing (first predetermined period). For example, sampling means 310 that refers to an A / D conversion value (detection output) at 2 ms), an air amount conversion table 302, a direct output signal from sampling means 301, or a converted air amount from air amount conversion table 302 Is sampled by a sampling means (not shown), and a pulsation amplitude ratio and pulsation frequency calculation means 320 for calculating a pulsation amplitude ratio and a pulsation frequency with reference to these results, and further, a hot-wire air quantity based on these calculation results Frequency response correction means 330 for correcting the frequency response of the sensor 10 and a support for correcting the sampling error. The air quantity Qc is calculated from the sampling error correcting means 340, the non-linearity correcting means 350 for correcting the non-linearity, and the corrected air quantity calculating means 360 for calculating the corrected air quantity from these results. Has been.

  Further, the pulsation amplitude ratio and pulsation frequency calculating means 320 is provided in the sampling means 310 for a second predetermined period (for example, 20 ms) longer than a sampling time (for example, 2 ms) that is the first predetermined period of the sampling means 310. Sampling value storage means 321 for storing and storing the detected air amount as a sampling result, difference amount calculation means 322 for calculating a difference amount between sampling values temporally adjacent with a positive / negative sign, and the number of sign inversions Inversion number calculating means 323 for calculating, and maximum, minimum and overall average value calculating means for calculating the overall average value of the sampling values during the maximum, minimum and second periods from the sampling values stored in the sampling value storage means 321. 324, the amplitude ratio calculating means 235 for calculating the pulsation amplitude ratio, and the inversion number calculating means 323 And a frequency calculating means 326 Metropolitan for calculating the number.

  The operation and function of the air amount calculation function configured as described above will be described below. FIG. 9 shows an example when the rotational speed in the low rotational speed region is 1200 r / min in an in-line four-cylinder internal combustion engine.

  The following description describes the operation of the amplitude ratio and pulsation frequency calculation means 320, which includes sampling means 310, sampling value storage means 321, difference amount calculation means 322, inversion number calculation means 323, maximum and minimum. These are executed by the overall average value calculating means 324, the amplitude ratio calculating means 235, the inversion number calculating means 323, and the frequency calculating means 326.

  This example is a case of 40 Hz when simply converted into a frequency, and shows an example of a pulsation state on the low frequency region side. The vertical axis represents the air amount Q (kg / h) and the horizontal axis represents the time (ms). The solid line that periodically changes is the true air amount Qr1 that changes with time, and between this 20 ms The average amount of air is indicated by a two-dot chain line QaVr.

  On the other hand, each point indicated by the black circle point s1 to the black circle point s11 is a value when the air amount is converted by the air amount conversion table 302 at a sampling timing of every 2 ms, and the adjacent portions of the black circle point s1 to the black circle point s11 are indicated. A dotted line that is quantized and connected is shown as an air amount Qs1 when the air amount is converted at a sampling timing of every 2 ms, and similarly an average value for 20 ms is shown as a broken line QaVs.

  As can be seen from this, since the sample air amount of Qs1 indicated by the black circle point s1 to the black circle point s11 can almost trace the true air amount Qr1 in the low frequency side region, the average air amount QaVr and the average air amount QaVs. Since the difference between the two is small and the error between the two is small, the correction amount may be small.

  Next, a method for obtaining the pulsation frequency Pf from the waveform shown in FIG. 9 will be described. As shown in the lower side of FIG. 9, at each black dot point indicated by the black dot point s1 to the black dot point s11, for example, temporally. When the difference between points adjacent to each other is calculated with a plus / minus sign, (s2-s1) has a sign (+), (s3-s2) has a sign (-), and the sign at the point indicated by point a is (+ ) To (−) and the number of inversions is incremented by one.

Similarly, (s10-s9) indicated by point b is inverted from (−) to (+), so the number of inversions is incremented by +1. (The number of inversions is also the number of inflection points.)
Therefore, if the second predetermined period is 20 ms, the number of inversions for 20 ms that is the second predetermined period is stored by accumulating the number of inversions during this period.

  Next, in order to convert this to the pulsation frequency Pf, in the case of a four-cylinder internal combustion engine, ½ of the number of inversions in 1 s corresponds to the frequency, so a value 50 times that in 20 ms is halved. Double it.

  That is, in FIG. 9, since the number of inflection points = 2, 50 × 2/2 = 50 (Hz) is obtained as the pulsation frequency Pf. In the example shown in FIG. 9, as described above, when the rotational speed is 1200 r / min, the actual frequency is 40 Hz, and an error occurs, but it can be obtained as a substantially close frequency.

  In addition, what is necessary is just to take the 2nd predetermined period long in order to improve the calculation precision of the pulsation frequency Pf. However, the stored number of each black dot is associated with the second predetermined period, and the RAM capacity increases when the stored number is increased. Therefore, the setting may be made in consideration of the RAM capacity and accuracy.

  Next, a method for obtaining the maximum value average Qmx1, the minimum value average Qmn1 and the overall average value QaVs in the second predetermined period will be described.

  By storing and saving the sampling value every 2 ms in the second predetermined period, if the sign of point a is (−) at the timing of updating the number of inversions described above, the black circle of the previous sampling value The point s2 is accumulated as the maximum value ΣQmx, and at the same time, the number of ΣQmx is accumulated as +1.

  At point b, the sign is (+). In this case, the previous sampling value black circle point S9 is accumulated as the minimum value ΣQmn, and at the same time, the number of ΣQmn times is stored as +1.

  The maximum average value Qmx1 can be obtained by dividing the integrated maximum value ΣQmx by the number of ΣQmx. However, in FIG. 9, there is only one inversion in the second predetermined period of 20 ms. This means the black circle point s2, and similarly, the minimum average Qmn1 means the black circle point s9, and the maximum and minimum values can be obtained respectively.

  On the other hand, when the second predetermined period is set to 20 ms, the number of samples during this period can be fixed to 10 at the sample timing every 2 ms. Therefore, the total value of each black circle point s1 to the black circle point s10 is obtained to obtain 10 samples. By dividing, an average value QaVs during this period can be obtained.

  Also, an amplitude ratio that is a ratio of the pulsation amplitude is calculated by calculating a pulsation amplitude amount dQ from the difference between the maximum value average Qmx1 and the minimum value average Qmn1, and dividing the pulsation amplitude amount dQ by the average value QaVs (dQ / QaVs). Pw is required.

  The above is the description of the low frequency region. Next, the case of the high frequency region will be described with reference to FIG. FIG. 10 shows the same thing as the example of FIG. 9, but is the one when the rotational speed of the internal combustion engine is operated at 3000 r / min, and the pulsation state on the high frequency region side of 100 Hz is simply converted into the frequency. It is a representation.

  As in FIG. 9, the solid line is the true air amount Qr1, the dotted line is the air amount Qs1 when the air amount is converted at 2 ms sampling timing, and the average value of the true air amount Qr1 is the average value QaVr indicated by the two-dot chain line. The average value of the air amount Qs1 is the average value QaVs indicated by a broken line.

  As can be seen from comparison with FIG. 9, the sampled detected air amount indicated by the air amount Qs1 in FIG. 10 cannot trace the true air amount Qr1, and the difference between the average value QaVr and the average value QaVs is Ers. It can be seen that the correction amount needs to be increased because the error is large as shown in FIG.

  The method for obtaining the pulsation frequency Pf from the waveform is also the same as the method shown in FIG. 9, and as shown in the lower side of FIG. When the calculation is performed and it is determined that the sign is inverted, the number of inversions for 20 ms is stored by adding the number of inversions to +1.

When this is converted into a pulsation frequency, the number of inflection points is 4 in FIG. 10, and 50 × 4/2 = 100 (Hz) is obtained as the pulsation frequency Pf.
Next, similarly for the method of obtaining the maximum, minimum and overall average values of the second predetermined period, the sampling values every 2 ms in the second predetermined period are stored and stored, and the above-described inversion count is updated. In the case of (−) according to the sign calculated at the timing, the previous sampling value is accumulated as the maximum value ΣQmx, and at the same time, the number of ΣQmx times is accumulated as +1.

  On the other hand, when the sign calculated at the timing of updating the number of inversions is (+), the previous sampling value is accumulated as the minimum value ΣQmn, and at the same time, the number of ΣQmn is accumulated as +1.

  In FIG. 10, the number of inversions in the second predetermined period of 20 ms is 4 times, the maximum average is the maximum value ΣQmx obtained by dividing the maximum value ΣQmx by the number of ΣQmx, and the minimum value average is the minimum integrated. The value ΣQmn is obtained as the minimum value average Qmn1 by two points obtained by dividing the value ΣQmn by the number of ΣQmn times.

  On the other hand, when the second predetermined period is set to 20 ms, the number of samples during this period can be fixed to 10 at the sample timing of every 2 ms as in FIG. 9. By dividing by Equation 10, the mean value QaVs during this period can be obtained.

  Also, an amplitude ratio that is a ratio of the pulsation amplitude is calculated by calculating a pulsation amplitude amount dQ from the difference between the maximum value average Qmx1 and the minimum value average Qmn1, and dividing the pulsation amplitude amount dQ by the average value QaVs (dQ / QaVs). Pw is required.

  Further, in this embodiment, based on the calculation result of the pulsation amplitude ratio Pw and the pulsation frequency Pf described above, the frequency response correction unit 330 in the subsequent stage of the pulsation amplitude ratio and pulsation frequency calculation unit 320 shown in FIG. An air amount Qc is calculated by an after-correction air amount calculation unit 360 that calculates the finally corrected air amount from the error correction unit 340 and the nonlinearity correction unit 350.

  FIG. 10 shows the corrected air amount at that time as Qc1, and the average value is shown as QaVc. The difference between the corrected average value QaVc and the true air amount average value QaVr can be reduced as indicated by Erc.

  FIG. 11 shows specific correction values of the above-described frequency response correction means 330, sampling error correction means 340, and nonlinearity correction means 350. FIG. 11A shows the frequency response of the hot-wire air quantity sensor 10. It is a map of the correction coefficient KQf to be corrected, (b) is a map of the correction coefficient KQs for correcting the sampling error, and (c) shows a table of the correction coefficient KQb for correcting the non-linearity.

  First, the correction coefficient KQf for correcting the frequency response of the hot-wire air quantity sensor 10 is calculated with reference to a three-dimensional map with the pulsation frequency Pf and the pulsation amplitude ratio Pw as axes. The pulsation frequency Pf and the pulsation amplitude ratio Pw are obtained by the method as described above.

  In this map, a correction amount inverse to the characteristics shown in FIG. 6 may be set. For example, the higher the frequency is, the larger the coefficient is, and the larger the pulsation amplitude ratio is, the larger the coefficient is 1.0. Set to a larger value. If each value is previously measured and mapped as a characteristic of the hot-wire air quantity sensor 10, it is not necessary to adapt the coefficient depending on the internal combustion engine.

  Next, the correction coefficient KQs for correcting the sampling error is calculated with reference to a three-dimensional map having the pulsation frequency Pf and the pulsation amplitude ratio Pw as axes. In this map as well, it is only necessary to set an inverse function correction amount of the characteristic shown in FIG. 7, and for example, the higher the pulsation frequency Pf, the larger the coefficient, and the larger the pulsation amplitude ratio Pw, the larger the coefficient. It is set with a value larger than 1.0. The correction coefficient need not be adapted depending on the fuel engine if a correction amount corresponding to an error for reading the pulsation amplitude ratio Pw and the pulsation frequency Pf is calculated and mapped in advance.

  Further, the correction coefficient KQb for correcting the non-linearity is configured such that the correction amount is increased for a portion having a large degree of non-linearity in the characteristics shown in FIG.

  Depending on the degree of linearity, for example, when the value of the air amount Q per unit of output voltage of the sensor is large, the two-dimensional value is larger than 1.0 with the air amount Q set to be large as the axis. Determined by table. The correction amount may be set in advance according to the sensor characteristic curve.

  In the present embodiment, each of the above-described correction coefficients is set to a value larger than 1.0 because it is desired to set the uncorrected state to 1.0 at the time of multiplication. Each correction coefficient may be set with reference to 0, but in this case, the same effect can be obtained by adding 1.0 to each coefficient when multiplying.

  Next, a method for calculating the finally corrected air amount using the three correction coefficients (KQf, KQs, KQb) described above will be described.

The corrected air amount Qc is given by the following formula (a) or the following formula (b). In this embodiment, the adjustment coefficients C1 to C3 for adjusting the correction coefficient are 1.0 or more. The corrected air amount Qc is obtained by multiplying the detected air amount Q at each sample timing.
Qc = Q × {(C1 × KQf) × (C2 × KQs) × (C3 × KQb)} (a) Qc = Q × {1+ (C1 × KQf) + (C2 × KQs) + (C3 × KQ) } ... (b)
Here, in the formula (b), each correction coefficient is set on the basis of 0.

  Whether to use equation (a) or equation (b) may be selected according to the magnitude and effect of the correction amount.

  Here, the adjustment coefficients C1 to C3 are set because they are both corrected based on the pulsation amplitude ratio Pw and the pulsation frequency Pf, and are set as adjustments in the case of overcorrection. The adjustment coefficients C1 to C3 may be set as a map or table corresponding to the rotational speed or load of the internal combustion engine and the throttle valve opening according to the operation region.

Furthermore, as another calculation method, there is a calculation method according to the following expression (c), which is calculated at each sampling timing. However, in this case, it is necessary to set each coefficient to 0 reference. In addition, since the difference is used in the calculation of the following formula (c), the effect of the correction is small and the coefficient needs to be large. However, since the divergence is 1.0, it is applied when a large correction is necessary Can not.
Qc = G × (Q−Qc_old) + Q (c)
G = coefficient multiplied by the detected air amount Q in the equation (b), (C1 × KQf) + (C2 × KQs) + (C3 × KQ). However, the condition is <1.0.
Qc_old: Qc 2 ms before
Since each correction coefficient (KQf, KQs, KQb) is obtained from the second predetermined period, the result of the next second predetermined period is obtained using the result of the immediately preceding second predetermined period. Until then, the same correction coefficients (KQf, KQs, KQb) are used.

  Next, calculation of each correction coefficient and the corrected air amount Qc will be described based on the flowchart of FIG.

  FIG. 12 shows a process of starting the detection signal output from the hot-wire air quantity sensor 10 at a sampling timing of, for example, every 2 ms, which is a first predetermined period. It is something to start.

  First, in step 1000, a signal detected from the airflow sensor 10 is converted into an air amount Q every 2 ms.

  Next, in step 1010, a flag is checked to determine whether measurement is being performed for the second predetermined period. If it is determined that flag = 0, that is, the measurement is completed, the process proceeds to step 2000; Proceed to 1020.

  In step 1020, a second predetermined period (for example, 20 ms) is being measured, and the air amount Q calculated in step 1000 is added to the previous value, integrated as ΣQ, and the process proceeds to step 1030.

  In step 1030, the difference amount is calculated with a sign by calculation of subtracting the previous air amount Q stored from the current air amount Q.

  In step 1040, the sign obtained by the subtraction calculation result of the previous stored air quantity Q and the previous air quantity Q is compared with the sign calculated in step 1030. For example, when the air amount Q is 2-byte data, it is determined that the data has been inverted when the MSB bits of the data do not match.

  If it is determined that the rotation has been reversed, the number of reversals is integrated in step 1050. In step 1060, it is compared whether the sign calculated in step 1030 is comparison (+). If (+), the process proceeds to step 1070, and the value immediately before the inversion is determined to be the lower limit inflection point. The inverted air amount Q one time before is added to ΣQmn as a lower limit and integrated, and at the same time, the number of ΣQmn is integrated.

  When the sign is (−) in step 1060, it is determined that the value immediately before the inversion is the upper limit inflection point, and the inverted air amount Q one time before the inversion is added to ΣQmx as the upper limit, and at the same time, The number of times of ΣQmx is integrated and the process proceeds to Step 1080.

  If it is determined in step 1040 that the sign has not been inverted, the processing from step 1050 to 1070 is not performed and the processing proceeds to step 1080.

  On the other hand, if it is determined in step 1010 that the measurement is being performed during the second predetermined period = 0, that is, if it is determined that the measurement has been completed, the process proceeds to step 2000 to calculate each correction coefficient.

  First, in step 2000, the minimum value average Qmn1 is obtained by dividing the minimum value ΣQmn by the number of ΣQmn from the minimum values ΣQmn and ΣQmn times calculated from steps 1020 to 1070 during the second predetermined period. Similarly, the maximum average value Qmx1 is obtained by dividing the maximum value ΣQmx by the number of ΣQmx times. Further, the overall average QaVs is calculated from the integrated air amount ΣQ / 10. Here, 10 is a value determined from the number of times of 2 ms sampling for 20 ms.

  Next, in step 2010, a pulsation amplitude amount dQ is calculated from the difference between the maximum value average Qmx1 and the minimum value average Qmn1, and the pulsation amplitude amount dQ is divided by the average value QaVs (dQ / QaVs) to thereby calculate the pulsation amplitude ratio. Is obtained.

  In step 2020, calculation to convert to the pulsation frequency Pf is performed, and in the case of a four-cylinder internal combustion engine, ½ of the number of inversions for 1 s corresponds to the frequency based on the inversion number integrated value calculated in step 1050. The pulsation frequency Pf can be obtained by halving the 50 times value for 20 ms.

  Three correction coefficients (KQf, KQs, KQb) are obtained in steps 2040 to 2050 using the pulsation amplitude ratio Pw obtained in step 2010 and the pulsation frequency Pf obtained in step 2020, respectively. Each correction coefficient is read from a map, a table or the like according to the magnitude of the pulsation amplitude ratio Pw and the pulsation frequency Pf as shown in FIG.

  After obtaining each correction coefficient, in step 2060, each integrated value and stored information in the second predetermined period are cleared for the next measurement, and in step 2070, the second measurement is performed for the next remeasurement. The period flag is set (1) and the process proceeds to step 1020.

  In step 1080, the corrected air amount Qc is calculated using the correction coefficients calculated in steps 2000 to 2070. Here, each correction coefficient is calculated with the same value until the calculation for the second predetermined period is updated.

  In step 1090, measurement is performed for a second predetermined period. For example, in the case of 20 ms, the count is incremented until 2 ms reaches 10 times.

  In step 1100, it is determined whether the second predetermined period has ended. For example, it is determined that the measurement is finished when the number of times of measurement in step 1090 reaches 10. If it is not determined to end in step 1100, the processing is terminated as it is. If it is determined to end in step 1100, the flag for the second predetermined period is reset (0) in step 1110, and the first step is performed in the next 2ms. Each correction coefficient is calculated in a routine starting from 2000. Although not described, at the time of the first calculation, steps 1020 and after are calculated with values set with the respective correction coefficients = 1.0 and the second predetermined period flag = 1 as initial values.

  As described above, in the hot-wire air amount sensor, there is a phenomenon that a response delay occurs in a higher frequency region, and the detected air amount average value and the true air amount average value do not match and an error occurs. The effect of reducing this error is expected.

  In addition, when taking in at a predetermined sampling timing and converting it into an air amount, there is a phenomenon that an error occurs in the detected average air amount depending on the magnitude and frequency of the intake pulsation. According to the present invention, this error can be reduced. Be expected.

DESCRIPTION OF SYMBOLS 10 ... Hot wire type air quantity sensor, 11 ... Air cleaner, 15 ... Throttle body, 12 ... Intake pipe, 17 ... Water temperature sensor, 101 ... Intake branch pipe, 103c ... Cylinder, 106 ... Collector, 107 ... Fuel tank, 108 ... Fuel pump 106 ... Pressure regulator, 111 ... Ignition coil, 112 ... Spark plug, 104 ... Exhaust pipe, 105 ... Catalyst,
110 ... Injector, 200 ... Control unit, 114 ... Air-fuel ratio sensor,
203... CPU, 201... I / O LSI.

Claims (13)

In the internal combustion engine air amount measuring device provided with a hot-wire air amount sensor that measures the amount of air flowing through the intake passage of the internal combustion engine,
A sampling means for detecting an output signal of the hot-wire air amount sensor or a detected air amount converted from the output signal for each predetermined sampling period;
An intake pulsation and pulsation frequency calculation stage for obtaining the magnitude and pulsation frequency of intake pulsation from a plurality of output signals or detected air amounts sampled by the output sampling means;
A frequency response error correction means for calculating a correction amount for correcting the frequency response of the hot air sensor from the magnitude of the intake pulsation and the pulsation frequency obtained in the intake pulsation and pulsation frequency calculation stage;
Sampling error correction means for calculating a correction amount for correcting a sampling error of the hot-wire air amount sensor from the intake pulsation and the pulsation frequency obtained in the intake pulsation and pulsation frequency calculation stage;
An air amount measurement for an internal combustion engine comprising correction air amount calculation means for correcting the detected air amount by a correction amount for correcting the frequency response and a correction amount for correcting the sampling error to obtain a correction air amount. apparatus. In the internal combustion engine air quantity measuring device according to claim 1,
The sampling means samples the output signal or the detected air amount of the hot-wire air amount sensor every first predetermined time, and the intake pulsation and pulsation frequency calculation stage is a second longer than the first predetermined time. An air quantity measuring device for an internal combustion engine, characterized in that an intake pulsation magnitude and a pulsation frequency are obtained from a plurality of the output signals or detected air quantities sampled during a predetermined time. The air quantity measuring device for an internal combustion engine according to claim 2,
The intake pulsation and pulsation frequency calculation stage includes storage means for storing the output signal or the detected air amount sampled at the first predetermined time for the second predetermined time; Calculation means for calculating the maximum value average and minimum value average and the overall average value within a predetermined time, calculation means for calculating the amplitude amount of pulsation from the difference between the maximum value average and the minimum value average, and the amplitude amount An air quantity measuring device for an internal combustion engine, comprising a calculating means for obtaining a ratio of pulsation amplitude from the overall average value. The air quantity measuring device for an internal combustion engine according to claim 2,
The intake pulsation and pulsation frequency calculation stage calculates the difference between the output signals sampled over the second predetermined time or the temporally adjacent ones of the detected air amount with a positive / negative sign, and reverses the sign An air amount measuring apparatus for an internal combustion engine, comprising: calculating means for calculating the number of times; and calculating means for calculating the period or frequency of the intake pulsation from the second predetermined time and the number of inversions. In the air quantity measuring device for an internal combustion engine according to any one of claims 3 to 4,
The frequency response error correction means obtains a correction amount according to the magnitude of the pulsation amplitude ratio and the pulsation frequency, and sets the correction amount to be larger as the pulsation amplitude ratio is larger and the pulsation frequency is higher. An air quantity measuring device for an internal combustion engine characterized by the above. In the air quantity measuring device for an internal combustion engine according to any one of claims 3 to 4,
The sampling error correction means obtains a correction amount according to the magnitude of the pulsation amplitude ratio and the pulsation frequency, and sets the correction amount to be larger as the pulsation amplitude ratio is larger and the pulsation frequency is higher. An air quantity measuring device for an internal combustion engine. In the internal combustion engine air quantity measuring device according to claim 1,
The correction air amount calculating means corrects the detected air amount by a correction amount from a nonlinear correction means for calculating a correction amount for correcting a nonlinear error determined by the magnitude of the detected air amount with respect to the overall average value. An air amount measuring device for an internal combustion engine. In the internal combustion engine air amount measuring device provided with a hot-wire air amount sensor that measures the amount of air flowing through the intake passage of the internal combustion engine,
The air amount measuring device is
Sampling the detected air amount converted from the output signal of the hot-wire air amount sensor every predetermined sampling period,
The intake pulsation magnitude and pulsation frequency are calculated from the plurality of sampled detected air amounts,
From the magnitude of the intake pulsation and the pulsation frequency, a correction amount for correcting the frequency response of the hot-wire air quantity sensor is calculated,
From the magnitude of the intake pulsation and the pulsation frequency, a correction amount for correcting the sampling error of the hot-wire air amount sensor is calculated,
A method for measuring an air amount of an internal combustion engine, comprising: calculating a correction air amount by correcting the detected air amount by a correction amount for correcting the frequency response and a correction amount for correcting the sampling error. In the internal combustion engine air amount measuring method according to claim 8,
The output signal of the hot-wire air amount sensor or the detected air amount is sampled every first predetermined time, and the plurality of the detections sampled during a second predetermined time longer than the first predetermined time. An air quantity measuring method for an internal combustion engine, characterized in that a magnitude and a pulsation frequency of intake pulsation are obtained from an air quantity. The method for measuring an air amount of an internal combustion engine according to claim 9,
The detected air amount sampled every first predetermined time is stored for the second predetermined time, and the maximum value average, minimum value average, and overall average value within the second predetermined time are stored. And calculating a pulsation amplitude amount from the difference between the maximum average value and the minimum average value, and calculating a ratio of pulsation amplitude from the amplitude amount and the overall average value. Air volume measurement method. The method for measuring an air amount of an internal combustion engine according to claim 9,
A difference amount between the detected air amounts sampled over the second predetermined time in terms of time is calculated with a plus / minus sign to calculate the number of sign inversions, and the second predetermined time And calculating the period or frequency of the intake pulsation from the number of inversions. The method for measuring an air amount of an internal combustion engine according to any one of claims 11 to 12,
A correction amount for correcting a frequency response error according to the magnitude of the pulsation amplitude ratio and the pulsation frequency is obtained, and the correction amount is set to be larger as the pulsation amplitude ratio is larger and the pulsation frequency is higher. An air amount measurement method for an internal combustion engine. The method for measuring an air amount of an internal combustion engine according to any one of claims 11 to 12,
A correction amount for correcting a sampling error according to the magnitude of the pulsation amplitude ratio and the pulsation frequency is obtained, and the correction amount is set to be larger as the pulsation amplitude ratio is larger and the pulsation frequency is higher. A method for measuring the amount of air in an internal combustion engine.

Images