FR2907169A1 - SYSTEM AND METHOD FOR CONTROLLING THE OPERATION OF AN INTERNAL COMBUSTION ENGINE WITH COMPENSATION OF DERIVATIVES AND DISPERSIONS OF THE ADJUSTED AIR FLOW MEASUREMENT - Google Patents

Abstract

System for monitoring the operation of a motor vehicle internal combustion engine, comprising a member for measuring the flow of air admitted, characterized in that it comprises a means (21) for determining the standard deviation of the spread on the measured airflow as a function of the measured value, a means (24) for comparing the standard deviation thus determined against a calibration value in order to determine an estimated optimum flow rate for gases entering the engine, and a correcting means (25) for deducing a correction function to be applied to the measured airflow on the basis of the estimated optimum flow of gases entering the engine when learning conditions are satisfied.

Description

1 System and method for controlling the operation of a motor

  TECHNICAL FIELD The present invention relates to a system and a method for controlling the operation of an internal combustion engine of a motor vehicle. The control of the operation of the engine is managed by a set of sensors and actuators according to a set of control laws, called software strategies, and characterization parameters or engine calibrations. All of these laws and parameters can be stored in an electronic control unit or ECU. In some engines, there is further provided a turbocharger comprising a turbine driving a compressor so as to increase the pressure of the air admitted into the engine cylinders. The turbine is placed at the outlet of the exhaust manifold of the engine and is driven by the exhaust gas. The power provided by the exhaust gases to the turbine can be modulated by installing a relief valve or by providing vanes with variable geometry on the turbine. The compressor is mounted on the same mechanical axis as the turbine. It compresses the air entering the intake manifold. In addition, the internal combustion engines may be equipped with partial recirculation circuits of the engine exhaust gas (known as EGR for Exhaust Gas Recirculation) to the intake circuit to reduce the amount of pollutant emissions. The control of the operation of the engine is provided electronically to regulate, in particular, the optimal flow rates of air and fuel. The enslavement of the actual flows with respect to the setpoint flows is done thanks to the measurements provided by a set of sensors. These sensors include the flowmeter which measures the flow of fresh air entering the engine.

2907169 2 Problems are generally encountered when using the flow meters because of dispersions of innumerating measures to their design and their use and present from their manufacture. In addition, this dispersion of the results may change over time in the form of a drift between actual and measured air flow. Similarly, the actual flow into the engine, or engine intake capacity, has a dispersion from manufacture. The strategies for controlling the operation of the engine must therefore not only take into account the measured flow rates but also the dispersions of these measured flow rates. To reduce the drift and dispersion of the admitted fresh air flow measurements, the solutions generally relate to the component itself via, for example, technological improvements or changes in manufacturing processes. To obtain substantial gains, these solutions are often expensive. To circumvent these problems, systems and methods for controlling the operation of the engine have been developed so as to use calculation means to correct the dispersions of the measurement results.

French Patent Application No. 2,857,055 (BOSCH) discloses a method for correcting the deviations of the air / fuel mixing ratio from a setpoint value by correcting the air and / or fuel flow rate. French Patent Application 2,860,268 (Renault) also discloses a method of controlling the operation of an internal combustion engine by regulating the flow of air by means of learning a cartography. engine. A patch estimated by mapping as described, is cumbersome and difficult to put into practice. The number of points to map for effective correction makes the application of this method difficult to use in a vehicle. In view of the above, the present invention relates to a system and a method for compensating in a simple manner, without adding additional sensor and for each engine, the dispersion and drift of the fresh air flow measuring means admitted in a 2907169 3 internal combustion engine. The subject of the invention is also the conditions and the logic diagram necessary for triggering a procedure for learning the correction of the measured air flow rate. In one embodiment, the system for controlling the operation of an air-fueled motor vehicle internal combustion engine comprises reducing the drifts and dispersions of a fresh air flow measurement means based on the comparison. between the flow rate measured by the flow meter on the one hand and a gas flow estimator entering the engine on the other hand in the absence of exhaust gas recirculation (EGR). The flow rate of the gases admitted to the estimated engine and the measured air flow rate as well as the standard deviations of the respective dispersions are quantified. In the absence of EGR, the intake air flow is equal to the gas flow admitted into the engine. The measured air flow rate and the flow rate of the gases admitted into the estimated engine are then compared at different times in the life of the vehicle and a corrective table is derived which makes it possible to establish an air flow correction function. It is thus possible to reduce the differences between the measured air flow and the air flow entering the engine, whatever the operating conditions of the engine, including EGR. The invention also provides an improvement of the methods of correction of the airflows by a correction of the measurement of the air flow by learning, such a correction that can be applied in the context of the use of the vehicle by the user. final. In addition, the correction takes place transparently for the user and throughout the life of the vehicle, which guarantees a correction adapted to the evolution over time of the engine. In an advantageous embodiment, the standard deviation of the dispersion of the measurements of the incoming air flow rate is determined. The flow rate of the gases admitted to the optimum estimated engine is then determined after comparing the standard deviations of the dispersions of the measured air flow and the flow rate of the gases admitted into the estimated engine. Finally, a function of correction of the measured air flow is calculated according to the differences observed between the measured air flow rate and the admitted gas flow rate in the estimated optimum engine, during learning phases. In one embodiment, learning of patches is subject to verification of stored learning conditions.

Once verified, a trigger signal is sent to the comparison means to calculate the corrections of the measured airflow. Verification of learning conditions is provided by a set of logic gates comparing the information collected by a set of sensors to the reference data. The first logic gate ensures that the motor operates in sufficiently stable conditions to ensure the reliability of the correction process. These stability conditions can be restricted to stored temperature and pressure ranges. For example and without limitation, there may be mentioned a motor water temperature range, an external temperature range or an atmospheric pressure range. In the case of an engine equipped with a turbocharger and a controlled partial exhaust gas redirection valve, the second door makes it possible to verify that the exhaust gas redirection valve is closed and that the turbocharger compensated for this closure by maintaining the flow of gases admitted into the engine at a constant level. The last door receives the signal from the first door, the means for checking the opening state of the partial exhaust gas recirculation valve and the means for comparing the measurement of the intake air flow with the reference value. This latter gate triggers the transmission of the signal triggering learning if each of the three signals provides a positive response. In an advantageous embodiment, the patches resulting from the comparison between the flow rate of the gases entering the optimum estimated engine and the measured air flow rate are stored in a corrective table, as well as the flow rates. corresponding measured air, for each learning point. To ensure the durability of the data, the reliability of the correction throughout the life of the engine, and to avoid restarting the learning process after each engine shutdown, the patch table is stored in a non-volatile memory. A corrective function of the air flow measured from the previously defined patch table is then determined.

Fixes and correction function can be defined as additive or multiplicative error corrections. In a preferred embodiment, the measured airflow is corrected by the corrective function. This corrected airflow is then provided to the control strategies using the fresh air flow measurement to decrease the impact of the fresh air measurement dispersions on their operation. Other objects, features and advantages of the invention will become apparent on reading the following description, given solely by way of nonlimiting example and with reference to the appended drawings in which: FIG. 1 schematically illustrates the main elements a combustion ignition engine, and supercharged with air. FIG. 2 illustrates, in one embodiment, the logic diagram governing the correction strategy. FIG. 3 illustrates, in one embodiment, the logic diagram used for the verification of the learning conditions. In Figure 1 is shown schematically an internal combustion engine 1 comprising four cylinders la. The fresh air entering the intake manifold of the engine 2 passes through a compressor 3a of a turbocharger 3 beforehand. The turbocharger 3 is composed of a compressor 3a and a turbine 3b arranged on the same axis. The fresh air taken from the outside, symbolized by the arrow F, first passes through an air filter 4, then a flow meter 5, before entering the compressor 3a. Compressed air by the compressor 3a passes through, an intake pipe 6 containing a pressure sensor 7 measuring the boost pressure of the engine 1 followed by a three-way valve 8 capable of regulating a partial recirculation flow of the engines. exhaust gas (EGR valve) before entering the intake manifold 2. The exhaust gases from the exhaust manifold 9, after combustion in the engine 1, through the exhaust pipe and are directed in part in the bypass line 10 of the EGR circuit and the valve 8, in order to be partly recirculated after mixing in the intake air directed to the intake manifold 2. The other part of the exhaust gas exhaust is fed through the exhaust pipe 12 to the turbine 3b to drive the compressor 3a. At the outlet of the turbine 3b, the exhaust gases pass through the exhaust elbow 13, then an optional nitrogen oxide catalytic filter 14 and / or an optional particulate filter 15, before being rejected in 16. The flowmeter 5, the turbocharger 3, the pressure sensor 7 and the EGR valve 8 are connected to the control means 17 via electrical connections 18. The control means 17 is connected to the atmosphere. electronically to a nonvolatile memory 19. The assembly 20 formed by the control means 17 and the non-volatile memory 19 is integrated in an electronic control unit (ECU) referenced 20 which furthermore ensures the management of the operation of the engine 1.

The learning strategy of the patch table and the correction of the incoming fresh air flow measurement is illustrated in FIG. 2. The strategy starts with the measurement of the flow of fresh air admitted by the flowmeter 4 , then the estimate of the standard deviation of the air flow measurement by a calculation means 21. Simultaneously, an estimation means 22 estimates the flow rate of the gases admitted to the engine followed by a calculation means 23 which calculates the standard deviation on the estimation of the flow rate of the gases admitted into the engine. The values of the standard deviation of the intake air flow rate and the flow rate of the gases admitted to the engine are provided by the calculation means 24 which compares these values with a calibration value (6) and calculates the gas flow rate. entering the optimal estimated engine. The correction means 25 calculates the corrections and the measured air flow rate correction function, based on the flow of air supplied by the flow meter 4, and the flow rate of the gases entering the optimum estimated engine supplied by the means. 24, if it has received a learning trigger signal 26 from the verification means of the learning conditions 27. The verification means 27 receives the flow rate of the gases admitted to the engine estimated by the estimation means 22 and the air flow measured by the flowmeter 4.

The logic diagram of the verification means 27 is detailed in FIG. 3. Finally, the correction function calculated by the correction means 25 is supplied to the correction means 28 in parallel with the measured fresh air measurement measured by the flowmeter. 4 which then calculates the admitted fresh air flow corrected.

The corrected value of the intake air flow rate is provided to CHIP 20 for use in engine operation control strategies. The flow rate of the gases admitted into the optimum estimated engine, coming from the correction means 25, is calculated by comparing the standard deviations of the dispersions of the measured air flow rate (6Qa mes) and the flow rate of the gases admitted to the estimated engine ( 6Qmot is). For this, we use the function: zz ~ Qair_mes + ~ Qmot_est 2 Qmotest ùg 30 In the case where this inequality is satisfied, the flow rate of the gases admitted into the optimal estimated engine is given by: / ~ / 1 _ Qmot_est + Qair If not, the flow rate of the gases admitted into the optimal estimated engine is equal to 2907169. Qmot estopti = Qmot est In a particular embodiment, it is possible to use whatever the standard deviations 6Qmot is and 6Qair my Qmot estopti = Qmot is in the previous equations and later, we have: 10 Q.ot est: Gas flow admitted in the engine estimated Qmot_est_opti: Gas flow admitted to the engine estimated 8 optimal Qair myAmounted airflow 6Qmot is: Standard deviation of the dispersion of the flow rate of gases admitted to the engine estimated 6Qair mes: Standard deviation of the dispersion of the measured air flow rate 8: calibration variable The verification means of the learning conditions is 20 detailed in Figure 3. I1 is composed three logic gates connected on the one hand to the comparison means themselves connected to the different sensors and calculating means and on the other hand to actuators or to the training means of the patch table. A first AND-type gate 29 sees on its inputs signals from the comparison means comparing the reference values to the corresponding measurements, namely the stability conditions 30, the environmental conditions 31, as well as the conditions on the time elapsed since the last iteration 32. Its outputs are connected to the second 34 and third gate 37. The environmental conditions include the measurement results of a set of sensors (not shown) of temperature and pressure of the conditions outside the vehicle (outside temperature). , atmospheric pressure, ....) and internal (cooling water temperature, ..) 2907169 9 A second AND type gate 34 sees on its inputs signals from the first gate 29 and the comparison means 33 the flow of gases admitted into the engine 1 with a reference value. Its outlets are connected to the control device 35 of the turbocharger 3 and to the control device 36 of the valve 7. This door 33 is only necessary in the case where the engine 1 is equipped with an EGR valve 7 and a turbocharger 3. A third AND gate 37 sees on its inputs signals from the control device 36 of the EGR valve 7, 10 of the first gate 29 and the comparison means 38 of the flow rate measured by the flowmeter. 4 with a stored value range. Its output sends a signal 26 to the correction means 25 triggering the learning of the patch table. The reference values used by the different comparison means form the learning conditions. It should be noted that in the case of the presence of an EGR valve 7 and a turbocharger 3, the EGR valve 7 is closed so that the only flow entering the engine 1 is the flow of fresh air. Under these conditions, the fresh air flow is equal to the flow rate of the gases admitted into the engine. It is then possible to determine the measurement error of the flowmeter 5. The closure of the EGR valve 7 may lead to a pressure drop at the input of the engine 1 that can move the flow of the gases entering the engine estimated from the operating point that has triggered the 'learning. In order to mitigate this pressure drop as well as the variations in the boost pressure that may result therefrom, specific parameters may be imposed on the supercharging regulation so that the flow rate of gases admitted to the engine without exhaust gas control remains at most identical to the flow rate of the gases admitted to the recirculating engine.

The errors in fresh air flow measurement can be interpreted mathematically, either as multiplicative factors or as additive factors. The fix table stores the measured and corrected airflow torque for each learning point. The average of several fresh air flow measurements and several gas flow calculations allowed in the estimated engine can be averaged prior to storing the values. The table thus obtained is stored in a non-volatile memory 19 making it possible to ensure the durability of the correctives learned during the stopping of the engine 1.

The correction factor is then interpolated from the patch table and the corrected airflow. In the case of a multiplicative error, the fix is calculated as 10 Cor (i) = mot_esti_opti Qair mes And the air flow corrected as: corrected = kx Qairmes With k = f (Cor) multiplicative correction factor from the patch table In the case of an additive error, the fix is calculated as follows: Cor '(i) = Qmot_esti_opti ù Qair_mes And the flow rate: Qaircorrigé ù + Qair_mes With k' = f (Cor ') additive correction factor from the correction table This correction is valid regardless of the operating conditions of the engine 1, including with the recirculation of the exhaust gas. The corrected air flow thus obtained is provided to the different strategies of controlling the operation of the engine via the unit 20. The present invention makes it possible to define a system and a method for correcting dispersions and flow drifts. measured fresh air, based on a method of learning a patch table in particular operating conditions of the engine, so-called learning conditions. A correction function is deduced from this table and is used to correct the measurement of the fresh air flow measured.

Claims (10)

  1. A system for controlling the operation of an internal combustion engine of a motor vehicle comprising a member for measuring the flow rate of the intake air (4) in the engine (1), a device for estimating the flow rate of the admitted gases (22). ) in the engine (1), and an electronic control unit (ECU) (20), characterized in that it comprises means (21) for determining the standard deviation of the dispersion of the measured air flow rate as a function of the measured value, means for comparing (24) the thus determined standard deviation with a calibration value to determine a flow rate of the gases entering the optimum estimated engine, and a correction means (25) for deducing an airflow correction function measured from the flow rate of the gases entering the optimum estimated engine and the measured airflow, when learning conditions are verified.   The control system of claim 1 wherein the comparing means (27) is capable of comparing the instantaneous conditions to the stored learning conditions and capable of providing a trigger signal (26) to the correction means (25).   3. Control system according to claims 1 or 2 wherein the comparison means (27) comprises a set of AND logic gates controlling the emission of the trigger signal (26) to the correction means (25).   4. Control system according to claim 3 wherein a first gate sees on its inputs signals from the comparison means to the learning values of the measurements of the stability conditions (30), the environmental conditions (31), as well as the time elapsed since the last iteration (32). 2907169 13   5. Control system according to claim 4 for an engine equipped with a turbocharger (3) and a controlled partial exhaust gas recirculation (EGR) valve (8), wherein a second door (34) sees on its inputs signals from the first gate (29) and the means (31) for comparing the flow rate of the gases admitted into the engine with the learning values, the outputs of the second gate being connected to control means (35) the turbocharger (3) and the partial exhaust gas recirculation valve (8). 10   6. Control system according to claim 5 wherein a third door (37) sees on its inputs signals from the control means (36) of the recirculation valve, the first door (29) and the comparison means. (38) of the air flow rate measured by the flow meter 4 to the training conditions, the output of the third gate sending a trigger signal (26) to the correction means (25) to trigger the storage of the engine operating parameters . 20   7. Control system according to one of the preceding claims wherein the correction means comprises a corrective table capable of storing patches resulting from a comparison between the flow rate of the gas entering the engine estimated optimum and the air flow rate. measured as well as the corresponding measured airflows.   8. Control system according to claim 7 comprising a nonvolatile memory (19) in which is stored the patch table.   9. A method of controlling the operation of an internal combustion engine in which the flow rate of fresh air admitted into the engine (1) is measured and the flow rate of the gases admitted into the engine is estimated, characterized in that the The standard deviation of the measured air flow rate is determined as a function of the measured value, which compares the thus determined standard deviation with a calibration value to determine a flow rate of the incoming gases. in the optimum estimated engine, and a flow table and airflow correction function measured when learning conditions are verified are deduced from the flow rate of gases entering the optimum estimated engine.   10. The control method according to claim 9, characterized in that the patches and the airflow correction function are calculated as corrections to multiplicative and / or additive errors.