JP5446610B2 - Intake air amount measuring method and intake air amount measuring device - Google Patents

Description

  The present invention relates to an intake air amount measurement method and an intake air amount measurement device that are capable of correcting a change with time without being affected by initial variation of a MAF sensor and a duct shape.

  In the control of the engine, the intake air amount (MAF) is an important quantity, so it is necessary to accurately measure the intake air amount. For example, the intake air amount is required to obtain the EGR rate in the EGR control. Further, when diagnosing the performance of actuators such as an EGR valve, an intake throttle, and an exhaust throttle, the intake air amount is used as a quantity that is a basis for the diagnosis.

  Conventionally, the amount of intake air is measured by installing a MAF sensor that measures the flow rate of air in a duct extending from the outside air to the compressor inlet of the turbocharger.

JP 2008-223516 A Japanese Patent No. 4218702 JP-A-7-293347 JP-A-8-338287

  The duct in which the MAF sensor is installed is not straight but winding. The air flow through the duct is not uniform and stable. For this reason, even if the flow rate given to the MAF sensor in the experimental wind tunnel or the like matches the reading value of the MAF sensor, the reading value of the MAF sensor arranged in the duct of the vehicle is actually relative to the flow rate of the air flowing through the duct. It is not always straight. In the case of the intake air amount of the engine, the theoretical value (theoretical air amount) is uniquely determined depending on the engine state, but the air flow rate measured in the duct is influenced by the bending of the duct as described above. It will not be the same as the amount.

  Further, the MAF sensor has individual variations. For this reason, in a vehicle, if the reading value of the MAF sensor is set to the intake air amount as it is, the intake air amount becomes inaccurate due to individual variations of the MAF sensor.

  In addition to these, the readings for the same air flow rate change as the MAF sensor changes over time. For this reason, even if the intake air amount as the reading of the MAF sensor is accurate at the initial stage of attaching the MAF sensor to the vehicle, the intake air amount becomes inaccurate due to long-term use.

  Accordingly, an object of the present invention is to provide an intake air amount measuring method and an intake air amount measuring apparatus that can solve the above-described problems and can correct a change with time without being affected by the initial variation of the MAF sensor and the duct shape. There is.

The present invention relates to an intake air amount measuring method for measuring an intake air amount into an engine with a MAF sensor installed in a duct extending from outside air to a compressor inlet of a turbocharger. The relationship between the first step for determining that the learning execution condition is satisfied when the engine is in the state, and the theoretical air amount estimated from the engine state and the measured air amount based on the reading value of the MAF sensor when the learning execution condition is satisfied And a third step of correcting the measured air amount based on the reading value of the MAF sensor based on the relationship to obtain the intake air amount. In the second step, the learning execution condition is each time but which satisfies the measurement SL between the measured air amount to the theoretical air amount estimated from the engine state based on the readings of the MAF sensor Measurement memory relationship learned when the current learning execution condition is satisfied, measurement memory relationship learned when the previous learning execution condition was satisfied, and learned when the current learning execution condition was satisfied The calculation relationship calculated based on the measurement storage relationship is stored independently, and the measurement storage relationship learned when the current learning execution condition is satisfied is learned when the previous learning execution condition is satisfied. The measured measurement relationship is stored until the next learning execution condition is satisfied and the calculation relationship is calculated . In the third step, the measured air amount based on the reading value of the MAF sensor is calculated based on the calculation relationship. This is a method for measuring the amount of intake air corrected to the amount of intake air.

Further, the present invention provides an intake air amount measuring apparatus that measures an intake air amount to an engine with a MAF sensor installed in a duct extending from outside air to a compressor inlet of a turbocharger. A learning execution condition establishment determination unit that determines that the learning execution condition is established when the engine is in an engine state, and a measurement based on the theoretical air amount estimated from the engine state and the reading value of the MAF sensor when the learning execution condition is established A relationship learning unit that learns a relationship with an air amount; and a correction execution unit that corrects a measured air amount based on a reading value of the MAF sensor based on the relationship to obtain an intake air amount, and the relationship learning unit whenever the learning execution condition is satisfied, the measurement air amount to the theoretical air amount estimated from the engine state based on the readings of the MAF sensor A measurement memory relationship that is learned when the current learning execution condition is satisfied, a measurement memory relationship that is learned when the previous learning execution condition is satisfied, and a learning that is performed when the current learning execution condition is satisfied When the previous learning execution condition is satisfied, the calculation relation calculated based on the measured storage relation is stored independently, and the measurement storage relation learned when the current learning execution condition is satisfied Until the next learning execution condition is established and the calculation relationship is calculated , and the correction execution unit sets the measured air amount based on the reading value of the MAF sensor to the calculation relationship . This is an intake air amount measuring device which corrects based on the intake air amount.

  The present invention exhibits the following excellent effects.

  (1) The influence of the initial variation of the MAF sensor and the duct shape is eliminated.

  (2) A change with time of the MAF sensor can be corrected.

It is a flowchart which shows the process sequence in the intake air amount measuring method of this invention. 1 is a configuration diagram of an intake air amount measuring device including a main part of a vehicle to which the present invention is applied. FIG. 6 is a characteristic diagram of measured air volume versus theoretical air volume for explaining bandwidth setting in the present invention. It is a characteristic diagram of measured air quantity versus theoretical air quantity showing an example of measured air quantity obtained by measurement in the present invention. FIG. 6 is a characteristic diagram of measured air quantity versus theoretical air quantity showing an example of obtaining an intake air quantity in the present invention. It is a measured air quantity versus theoretical air quantity characteristic diagram showing an example of updating a memory curve in the present invention.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

  As shown in FIG. 1, in the processing procedure in the intake air amount measurement method of the present invention, step S1 for determining whether or not the current engine state is an engine state that satisfies the learning execution condition, and the theoretical air amount is calculated. Step S2, Step S3 for calculating the measured air amount, Step S4 for calculating the learning characteristic curve, and Step S5 for correcting the measured air amount based on the reading value of the MAF sensor to obtain the intake air amount.

  In the intake air amount measuring method of the present invention, after learning, the measured air amount based on the reading value of the MAF sensor is corrected based on the above relationship to obtain the intake air amount, and this intake air amount is used for EGR control and actuator performance diagnosis. Use for such as.

  In the intake air amount measurement method of the present invention, learning is executed every time the learning execution condition is satisfied. The measured air amount at the time of learning is stored after learning, and is used for comparison with the measured air amount at the next learning. In other words, in the present invention, the relationship between the theoretical air amount and the measured air amount resulting from the influence of the initial variation of the MAF sensor and the duct shape is learned at the first learning opportunity, and the subsequent change with time each time the learning execution condition is satisfied. As a result of learning, the relationship between the theoretical air volume and the measured air volume is updated.

  As shown in FIG. 2, the vehicle includes an engine 1, a turbocharger 2 that compresses the intake air to the engine 1, an EGR valve 3 that adjusts the EGR amount, and a compressor outlet of the turbocharger 2 to the engine. An intake throttle 5 for opening and closing the intake pipe 4, an exhaust throttle 7 for opening and closing an exhaust pipe 6 from the turbine outlet of the turbocharger 2 to the outside air, and a MAF sensor 9 installed in a duct 8 extending from the outside air to the compressor inlet of the turbocharger 2. And a controller (ECM; Engine Control Module) 10 for controlling the engine 1 and an accelerator pedal sensor 11 for inputting an operation amount of an accelerator pedal to the controller 10. In addition to these, the vehicle is provided with known sensors, actuators, controllers, etc., but description thereof is omitted.

  In the intake air amount measurement device according to the present invention, when the engine state is an engine state in which the preset intake air amount is stable, it is determined that the learning execution condition is satisfied and learning learning condition establishment determination is started to start learning. When the learning execution condition is satisfied, the relationship learning unit 13 that learns the relationship between the theoretical air amount estimated from the engine state and the measured air amount based on the reading value of the MAF sensor 9, and the reading of the MAF sensor 9 And a correction execution unit that corrects the measured air amount based on the value based on the relationship to obtain the intake air amount. The learning execution condition establishment determination unit 12, the relationship learning unit 13, and the correction execution unit 14 may be provided in the controller 10 for engine control, or may be provided in a controller different from the controller 10 for engine control.

  Hereinafter, the intake air amount measurement method of the present invention will be described in detail according to the procedure of FIG.

  In step S1, it is determined whether the learning execution condition is satisfied.

  The learning execution condition is that the engine is in an ideal state. The ideal state is a state where the intake air amount is stable. Specifically, the learning execution condition is satisfied when all the conditions listed below are satisfied.

  1) The actuator performance is not being diagnosed.

  When diagnosing whether the actuators such as the EGR valve 3, the intake throttle 5 and the exhaust throttle 7 are broken, the amount of intake air is the basis of the diagnosis, so learning is not performed during the performance diagnosis of the actuator.

  2) The MAF sensor 9 has no electrical abnormality.

  It is abnormal that the voltage or current giving the reading value of the MAF sensor 9 is a value that is not actually possible (a value exceeding the upper limit value or the lower limit value) or constant regardless of the engine state. Since 9 is broken, we don't learn.

  3) The atmospheric pressure is within the set range.

  No learning is performed when the atmospheric pressure measured by an atmospheric pressure sensor (not shown) exceeds a preset upper limit value or less than a lower limit value. This is because if the atmospheric pressure is too high or too low, the intake air amount is significantly affected.

  4) The engine 1 is warm. Don't be too hot.

  If the water temperature of the engine coolant measured by a water temperature sensor (not shown) is less than a preset lower limit value or exceeds an upper limit value, learning is not performed. This is because the intake air amount is not stable when the engine 1 is not warmed or extremely hot.

  5) The intake air temperature is within the set range.

  If the intake air temperature measured by an intake air temperature sensor (not shown) is less than a preset lower limit value or exceeds an upper limit value, learning is not performed. This is because the reading value of the MAF sensor 9 is affected when the density of the air greatly changes depending on the temperature.

  6) The outlet temperature of the EGR cooler (not shown) is within the set range.

  If the temperature of the EGR gas measured by a temperature sensor (not shown) installed at the outlet of the EGR cooler is less than a preset lower limit value or exceeds an upper limit value, learning is not performed. This is because the intake air amount is affected. That is, if there is a large difference between the temperature of the inhaled air and the temperature of the EGR gas returning to the air, it becomes difficult to inhale the air.

  7) The exhaust brake 7 is not operating.

  8) DPD (Diesel Paticulate Defuser) (not shown) is not in post-processing (reproduction processing).

  9) A predetermined time has elapsed since the engine was started.

  Immediately after the engine is started, the engine state is unstable and the intake air amount is not stable, so learning is not performed.

  10) The engine speed fluctuation is within the set range.

  When the engine 1 is hunting and the fluctuation of the engine speed is large, the amount of intake air is not stable and learning is not performed.

  11) The variation of the fuel injection amount is within the set range.

  When the variation in the fuel injection amount is large, the intake air amount is not stable, so learning is not performed.

  12) The boost pressure fluctuation of the turbocharger 2 is within the set range.

  When the fluctuation of the boost pressure is large, the intake air amount is not stable, so learning is not performed.

  13) The opening degree of the EGR valve 3 is not more than a set value.

  It is desirable that the EGR valve 3 is fully closed.

  14) The opening degree of the intake throttle 5 is not less than the set value.

  It is desirable that the intake throttle 5 is fully open.

  15) There is no abnormality in the actuator.

  If actuators such as the EGR valve 3, the intake throttle 5, and the exhaust throttle 7 are broken, the intake air amount is affected, so learning is not performed.

  When all the conditions are satisfied, the engine state is an ideal state in which the intake air amount is stable, which is suitable for learning the intake air amount measurement. Therefore, when the learning execution condition is satisfied, the process proceeds to steps S2 to S5 to execute learning.

  In step S2, a theoretical air amount is calculated.

  The theoretical air amount is the amount of air that should enter the cylinder in the current engine state. The theoretical air amount is generally determined by the engine speed, but is calculated in detail by the following theoretical formula.

Volumetric efficiency is the density of air entering the cylinder and depends on temperature. The total cylinder volume is set for each engine model. The engine speed denominator = 2 is for the case where the engine 1 has four cylinders, and means two cylinders per revolution. The air density of the atmosphere is a fixed value of 1.293 kg / m ³ . The standard intake pressure is a fixed value of 101.3 kPa.

  In step S3, a measurement air amount is calculated.

  As a preparation, a plurality of measurement bands are set with respect to the theoretical air amount on the vertical axis of FIG. At this time, the number of measurement bands is set as a constant in advance. Next, the width of the measurement band is set. For example, 15 measurement bands are set, and the start points of the measurement bands are MSBAND1 to MSBAND15. Specific values of MSBAND1 to MSBAND15 are set in a one-dimensional map referred to by atmospheric pressure. On the other hand, the widths of the measurement bands are all the same, and the value is defined as a bandwidth MAFBAND. A specific value of the bandwidth MAFBAND is set in a one-dimensional map that is referred to by atmospheric pressure. The reason why the starting point and width of the measurement band are changed according to the atmospheric pressure is to cope with the difference in the sea level.

  As shown in FIG. 3, the range from the theoretical air amount MSBAND1 to MSBAND1 + MAFBAND is one measurement band, and the range from the theoretical air amount MSBAND2 to MSBAND2 + MAFBAND is another measurement band. This figure is an example. Adjacent measurement bands are spaced from each other, but the distance between the measurement band start points is made narrower than the bandwidth so that the adjacent measurement bands partially overlap. May be.

  In the graph of FIG. 3, the measured air amount based on the reading value of the MAF sensor is plotted in correspondence with the theoretical air amount estimated from the engine state using the above theoretical formula. In each measurement band, the relationship between the theoretical air amount and the measured air amount is examined. In the illustrated example, among the two plotted measurement points, the measurement point in the measurement band of MSBAND1 indicates that the theoretical air amount and the measurement air amount substantially coincide, whereas the measurement in the measurement band of MSBAND2 The point indicates that the theoretical air amount and the measured air amount are deviated. That is, ideally, the measured air amount and the theoretical air amount are 1: 1, but actually the measured air amount varies.

  It is desirable to collect 10 or more such measurement points for each measurement band.

  In step S4, a learning characteristic curve is calculated.

For each measurement band, the average value of the measurement air volume is calculated for all measurement points in the measurement band, and the measurement air volume (representative point) in the measurement band is calculated. These representative points are plotted in the graph of FIG. 4 (FIG. 4 shows an example with eight representative points). From these representative points, a quadratic approximate curve is obtained by the method of least squares. Here, the approximate curve to be obtained is
Y = AX ² + BX + C
Y is a vertical axis variable, X is a horizontal axis variable, and A, B, and C are variable constants.
And

Since the representative point (Xi, Yi) (i = 1 to 15) is a point on the approximate curve, variable constants A, B, and C that minimize Yi−Y are obtained. That is,
S = Σ (Yi−AXi ² −BXi−C) ²
The variable constants A, B, and C that minimize the S of are obtained.

Differentiating S with each of the variable constants A, B, and C, and the simultaneous equations obtained with = 0,
A (ΣXi ⁴ ) + B (ΣXi ³ ) + C (ΣXi ² )
= ΣXi ² Yi
A (ΣXi ³ ) + B (ΣXi ² ) + C (ΣXi)
= ΣXiYi
A (ΣXi ² ) + B (ΣXi) + C (Σ1)
= ΣYi
It becomes. Variable constants A, B, and C are obtained from these simultaneous equations.

  The approximate curve of FIG. 4 obtained in this way represents the relationship between the theoretical air amount and the measured air amount resulting from the initial variation of the MAF sensor 9 and the influence of the duct shape. This approximate curve is stored as a learning memory curve.

  In step S5, that is, after learning, the measured air amount based on the reading value of the MAF sensor 9 is applied to the learning memory curve to read the theoretical air amount, and the measured air amount is corrected using the theoretical air amount as the intake air amount. It will be. As a result, the initial variation of the MAF sensor 9 and the influence of the duct shape are removed, and an accurate intake air amount can be measured.

  Next, the update of the memory curve will be described.

  As shown in FIG. 5, when the MAF sensor 9 changes with time, the learning memory curve obtained at the first learning opportunity is obtained when learning is performed by the above-described procedure to obtain an approximate curve (measurement memory curve). Deviation occurs. This measurement memory curve is saved and used in the procedure for correcting the measured air amount from the next time. For example, the theoretical air amount is read by applying the measured air amount based on the reading value of the MAF sensor 9 to the measurement memory curve, and this theoretical air amount is set as the intake air amount.

  In this way, by updating the measurement memory curve every time learning is performed, the change over time of the MAF sensor 9 is always used using the measurement memory curve that matches the progress of the MAF sensor 9 over time. It can be corrected.

  In the present embodiment, the measurement memory curve is updated as follows.

  As shown in FIG. 6, the measurement memory curve (current measurement memory curve; indicated by a broken line) created in the current learning with respect to the measurement memory curve (previous memory curve; indicated by a solid line) created and saved in the previous learning. ) Is shifted due to the time-dependent change of the MAF sensor 9. At this time, the current measurement memory curve is not used as it is for the subsequent correction procedure, but an intermediate curve (current calculation curve; indicated by a dotted line) between the previous memory curve and the current measurement memory curve is created. The current calculation curve is obtained from, for example, the average value of the previous memory curve and the current measurement memory curve at the same theoretical air amount. This calculated curve is stored and used for the correction procedure of the measured air amount after the current learning. This time, the measurement memory curve is stored and used to obtain an average value with the measurement memory curve created in the next learning.

  In this way, since the measurement memory curve created in the current learning and the measurement memory curve created / saved in the previous learning are taken in between, the current measurement memory curve created in the current learning due to some disturbance is assumed. Even if is inaccurate, the effect of the previous memory curve remains on the current calculated curve, so that the intake air amount is prevented from becoming extremely inaccurate.

  As described above, according to the intake air amount measuring method of the present invention, a learning memory curve representing the relationship between the theoretical air amount and the measured air amount is created by learning, and after learning, the reading value of the MAF sensor 9 is used. Since the measured air amount is applied to the learning memory curve to obtain the intake air amount, the initial variation of the MAF sensor 9 and the influence of the duct shape are removed, and the accurate intake air amount can be measured.

  Further, according to the intake air amount measurement method of the present invention, the engine state where the intake air amount is stabilized is set as the learning execution condition in advance, and the learning is executed when the engine execution state satisfies the learning execution condition. The relationship between the air amount and the measured air amount can be accurately obtained, and the correction after learning becomes accurate.

  Also, according to the intake air amount measurement method of the present invention, the measurement memory curve is updated each time learning is performed, so that the change over time of the MAF sensor 9 can be corrected.

  Further, according to the intake air amount measurement method of the present invention, in calculating the measurement air amount based on the reading value of the MAF sensor 9, a plurality of measurement bands are set, and a plurality of measurement points are obtained for each measurement band. Since a representative point consisting of the average value is obtained for each measurement band, noise factors such as electrical external noise entering the MAF sensor 9 are eliminated, and the measurement air amount (representative point) becomes accurate. And since an approximate curve is calculated | required by the least squares method from the representative point of a some measurement zone | band, the mislearning resulting from a noise factor is avoided.

  Further, in the method for measuring the intake air amount of the present invention, when calculating the measurement air amount based on the reading value of the MAF sensor 9, the start point and the width of the measurement band are changed with reference to the map according to the atmospheric pressure. ing. In the United States and other parts of the world, there are high-altitude areas where people live at an altitude of 3000 m or more. In vehicles used in such high-altitude areas, it is desirable that a learning memory curve suitable for the high-altitude areas can be learned. According to this, since the start point and width of the measurement band are appropriately set in conjunction with the atmospheric pressure, a learning memory curve suitable for the region can be learned regardless of the difference in sea level.

1 Engine 2 Turbocharger 3 EGR valve 5 Intake throttle 7 Exhaust throttle 9 MAF sensor 10 ECM
12 learning execution condition establishment determination unit 13 relation learning unit 14 correction execution unit

Claims (2)

In an intake air amount measurement method for measuring an intake air amount into an engine with a MAF sensor installed in a duct extending from outside air to a turbocharger compressor inlet,
A first step of determining that the learning execution condition is satisfied when the engine state is an engine state in which a preset intake air amount is stable;
A second step of learning a relationship between a theoretical air amount estimated from an engine state and a measured air amount based on a reading value of the MAF sensor when a learning execution condition is satisfied;
A third step of correcting the measured air amount based on the reading value of the MAF sensor based on the relationship to obtain an intake air amount;
With
In the second step, every time the learning execution condition is satisfied, the current learning execution condition is satisfied by the measurement storage relationship between the theoretical air amount estimated from the engine state and the measured air amount based on the reading value of the MAF sensor. Calculated based on the measurement memory relationship learned when the learning execution condition was satisfied, the measurement memory relationship learned when the previous learning execution condition was satisfied, and the measurement memory relationship learned when the current learning execution condition was satisfied And the next learning execution condition as the measurement memory relation learned when the previous learning execution condition is satisfied, and the measurement storage relation learned when the current learning execution condition is satisfied. Is stored until the calculation relationship is calculated ,
In the third step, the intake air amount measuring method is characterized in that the measured air amount based on the reading value of the MAF sensor is corrected based on the calculation relationship to be an intake air amount. In an intake air amount measuring device that measures an intake air amount to an engine with a MAF sensor installed in a duct extending from outside air to a turbocharger compressor inlet,
A learning execution condition establishment determination unit that determines that the learning execution condition is established when the engine state is an engine state in which a preset intake air amount is stable;
A relationship learning unit for learning a relationship between a theoretical air amount estimated from an engine state and a measured air amount based on a reading value of the MAF sensor when a learning execution condition is satisfied;
A correction execution unit that corrects a measured air amount based on the reading value of the MAF sensor based on the relationship to obtain an intake air amount;
With
The relationship learning unit is a measurement storage relationship between the theoretical air amount estimated from the engine state and the measured air amount based on the reading value of the MAF sensor each time the learning execution condition is satisfied, and the current learning execution condition is satisfied. Calculated based on the measurement memory relationship learned when the learning execution condition was satisfied, the measurement memory relationship learned when the previous learning execution condition was satisfied, and the measurement memory relationship learned when the current learning execution condition was satisfied And the next learning execution condition as the measurement memory relation learned when the previous learning execution condition is satisfied, and the measurement storage relation learned when the current learning execution condition is satisfied. Is stored until the calculation relationship is calculated ,
The correction execution unit corrects the measured air amount based on the reading value of the MAF sensor based on the calculation relationship to obtain an intake air amount.

Images