• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The invention relates to a method for resetting the behavioral models of motor vehicle internal combustion engine actuators, the motor being equipped with controlled actuators comprising actuators for phase shift of a camshaft and variable valve lift. , throttle body actuators and injector. The behavior models are readjusted simultaneously for the actuators in at least one operating point of the motor with: - for camshaft phase shift and valve lift actuators, according to a difference between a flow measurement and a model for estimating a flow rate in each cylinder of the engine, - for the throttle body actuator, according to a difference between a flow measurement and a throttle flow estimation model, - for said at least one thrust actuator. injector, according to a gap between a measure of wealth and a wealth directive.
    公开号:FR3052189A1
    申请号:FR1654948
    申请日:2016-06-01
    公开日:2017-12-08
    发明作者:Clement Pouly
    申请人:Peugeot Citroen Automobiles SA；
    IPC主号:
    专利说明:

    METHOD FOR REPRESENTING MODELS FOR BEHAVIOR OF ACTUATORS OF INTAKE LINES AND INJECTION OF INTERNAL COMBUSTION ENGINE
    The invention relates to a method of resetting the behavior models of actuators of intake and injection lines of internal combustion engine of a motor vehicle, the engine being advantageously but not limited to a gasoline engine.
    Due to manufacturing dispersions, wear and clogging of the actuators of the engine, their physical behavior may differ from the behavior models integrated in the engine control. This shift of the actuator models can lead to drifts of wealth, thus to overconsumption or an increase in polluting emissions.
    This shift can also have impacts on the driving pleasure felt by the driver. A motor control, in charge of controlling the actuators by integrating models of the behavior of these actuators, must therefore, throughout the life of the vehicle, exploit the information of various sensors present on the engine and in the intake lines. air, exhaust and motor even to reset the behavior models and adapt them to the actual operation of the actuators.
    It is possible to group the actuators into actuators relating to the intake line and the injection line. Inlet line actuators are actuators for a throttle body, variable intake and exhaust valve lift, intake and exhaust camshaft phase shifters. The injection line actuators are the injector actuators.
    The engine assembly taken in its broad sense including the intake lines, exhaust and injection is equipped with a flow meter measuring an air flow at the throttle body, a sensor of measuring the pressure in an intake plenum and an oxygen sensor measuring the richness in the exhaust line for the control of these actuators.
    There are multiple devices depending actuators to adapt and available measurements. In a first example, a model for an intake manifold estimates intake pressure from the difference of the inflow, namely to the intake manifold at the engine throttle, and the outflows, ie level of the cylinder or cylinders. The difference between the estimated intake manifold pressure and the pressure measured by a pressure sensor at the cylinder (s) gives a coupled image of the throttle model offsets and the cylinder flow estimation model. In addition, the analysis of a measurement of richness from an oxygen probe gives a coupled image of the offset of the estimation model of the cylinder flow and the injector model.
    [0007] These two respective differences in pressure and richness exploited on each stabilized point make it possible, step by step, to find the offsets of the models with respect to the physics of the actuators. In this case, we seek to identify the source of the error and to correct it by including it in the model of behavior of the actuator.
    For this first example, the two calculated differences, namely pressure and richness differences, report offset on several possible sources. The pressure difference may be due to an offset on the butterfly model and / or on the cylinder model. The difference in richness may be due to a shift on the cylinder model and / or the injector model. It is therefore difficult to precisely determine the source of the offsets and therefore to precisely align the actuator models.
    [0009] It is generally by iteration and on a large number of stabilized operating points that the identification takes place. The function can therefore be relatively slow and imprecise. This inaccuracy can result in the fact that the differences in richness and pressure on the points traveled are minimized, but the correction of the actuator models does not necessarily conform to the physical models of these actuators. This has the consequence that on the other operating points, other than stabilized or outside learning areas, the models are not properly recalibrated.
    For the second example, the correction principle turns out to be very heavy in terms of storing the corrections because the corrections on each point or operating area are stored. This can be problematic in determining the correct correction between two points or areas. In addition, the choice in the development phase of zones, their number and their size can be difficult. Finally, apart from the points or areas covered, the system is not recalibrated.
    The document FR-A-2 970 348 describes a method of controlling an actuator for performing a function for an internal combustion engine. The method comprises activating the actuator with a setpoint chosen according to the operating point of the motor, the registration of the setpoint by comparison between a measured signal resulting from the activation of the actuator and a reference signal which is function of the operating point of the motor.
    Prior to resetting the set point, the method further comprises receiving a request for resetting the setpoint, the resetting request being triggered by the user of the motor vehicle. However, this document does not describe a simultaneous registration of actuator behavior models as close as possible to their actual behavior.
    Therefore, the problem underlying the invention is to compensate for the shifts between the actuator models and their physical behavior, the offsets may be due to manufacturing dispersions, wear, fouling for models of behavior of the actuators present in the intake and injection lines of an internal combustion engine of a motor vehicle.
    To achieve this objective, it is provided according to the invention a method of resetting the behavior models of air intake lines and fuel injection of internal combustion engine motor vehicle engines, the engine being equipped with pilot actuators comprising an intake and exhaust camshaft phase shift actuators and variable intake and exhaust valve lift actuators, a throttle valve actuator and a throttle actuator injector for each cylinder of the engine, characterized in that the registration of the behavior models is done simultaneously for the actuators of the intake and injection lines in at least one operating point of the engine with: - for phase shift actuators of a camshaft and variable intake or exhaust valve lift, according to a cylinder flow differential between an actual flow measurement and a model for estimating a flow rate in each cylinder, - for the throttle body actuator, according to a throttle flow gap between an actual flow measurement and a throttle flow estimation model, - for said at least one actuator injector, according to a wealth gap between a measure of real wealth and a wealth directive.
    By recalibrating the models of behavior of the actuators of the system as close as possible to their actual behavior, the manufacturing dispersions, the wear and the fouling of these actuators are compensated. This avoids the shift of the actuator models, thus limiting the excesses of the wealth, thus an overconsumption or an increase of the polluting emissions and the impacts on the pleasure of driving felt by the driver.
    The proposed invention differs from existing inventions by the fact that it decouples errors, calculated deviations reporting system drifts can be attributed to only one source so a single actuator behavior model. The invention therefore proves to be faster and more precise in terms of identifying the corrections to be made to the behavior models of the actuators.
    In addition, the method according to the invention makes it possible to identify the source of the differences between the model and the physical behavior of the actuator. The behavior model of the actuator is corrected with the minimum of parameters and the correction is valid everywhere, even outside the areas where the error has been learned.
    The invention therefore relates to an engine control method for compensating for the offsets between the models of the actuators and their physical behavior. These shifts may be due to manufacturing dispersions, wear, fouling, etc ... and may have consequences on consumption, pollutant emissions and drivability. These offsets can not all be measured directly at the actuator. It is then used indirect information from the engine sensors. The advantage of the invention is to determine an engine control method that makes it possible to translate the information from the sensors of the engine into correction of the models of behavior of the actuators.
    Unlike the current system of the state of the art, the method proposed by the present invention allows an adaptation based on the identification of the source of the difference between the models and the physics of the actuators. This type of adaptation makes it possible not to have to go through all the operating zones to correctly reset the system. The method according to the invention also proposes a precise identification of the source of the errors based on the knowledge of the sensitivities of the engine to the various parameters, in particular the camshaft dephaser position, the exhaust camshaft dephaser position. , lift position of the intake and exhaust valves and opening of the butterfly flap. This ensures a certain robustness in identifying sources of errors, including in areas where you can not learn.
    The method according to the present invention therefore leads to a faster and more accurate system registration including in areas outside the adaptation domain. This results in a better refocusing of the wealth error, which is a limitation of consumption and pollutant emissions and a better behavior of the engine in terms of the driver's perceptible approval.
    Advantageously, it is defined at least one adaptive for each actuator, said at least one adaptive being: - for the throttle body actuator, an adaptive on a passage section by the throttle, - for phase shift actuators d camshaft, adaptive to a phase shift value for each type of intake or exhaust phase-shifter, - for variable valve lift actuators, an adaptive lift height for each type of intake valve or exhaust, - for said at least one injector actuator, its model of behavior of the injector similar to a line with modulation of an estimate of the gain and an offset, adaptive respectively on the gain and the shift.
    Advantageously, said at least one operating point of the engine is selected according to operating conditions and stability of the point, - the operating conditions relating to unitarily or in combination a motor temperature being in a temperature range of the engine predetermined, an outside air temperature within a predetermined outside air temperature range, an atmospheric pressure within a predetermined atmospheric pressure range, a motor operating mode other than a degraded mode, a starting phase, a catalyst heating mode or any other mode outside the usual operating range of the engine, - the stability conditions attesting that one or more engine operating parameters have been stable for a predetermined duration, this or these parameters being chosen individually or in combination between a flow of air entering ns the engine, a engine speed, a position of the actuators of the intake line to be adapted, a measurement of one or more sensors such as flowmeter, air intake pressure sensor or oxygen sensor.
    Advantageously, on the engine operating point selected, it is determined the respective sensitivities of the behavior models to be adapted according to the differences between each real measurement and its associated model, an application of an adaptive on an associated estimation model. being corrected according to the respective sensitivity of the model, with a determination of respective sensitivities of the cylinder flow at the positions of the intake and exhaust phase shifters and respectively at a maximum position of the valve lift, a sensitivity of the throttle flow to a throttle passage section, respective sensitivities of the richness at the injector gain and the injector offset, the sensitivities being determined by mathematical derivative of the system state equations, by calculating a local variation around a operating point or by identification during an initial phase of development of the m otor.
    Advantageously, it is worked by convergence of a respective parameter for the determination of each adaptive, each converged parameter taking into account the previously converged parameters for operating points, and it is calculated on the selected operating point:
    being the sensitivity of the cylinder flow to the position of the intake phase shifter for the operating point then in force,
    being the sensitivity of the cylinder flow to the position of the exhaust phase shifter for the operating point then in force,
    the sensitivity of the cylinder flow rate to the maximum position of the valve lift for the operating point then in effect, the cylinder flow deviation uncorrected for the operating point then in force, parameter _convergé_vvta and parameter_convergé_vvte being the parameters of position respectively of the intake phase shifter and the exhaust phase shifter which are converged on the operating points seen so far, and - a corrected butterfly flow gap ε_Qpapillon ^ corrected according to the following equation
    the throttle flow sensitivity at the throttle passage section for the operating point then in effect, ejQpapillon, the uncorrected throttle gap for the operating point then in effect, parameter_converged_section being the section parameter of the throttle passage converged on the operating points seen so far.
    Advantageously, a difference between a real measure of wealth and a wealth directive is corrected to take into account only a drift of the injector model, the uncorrected wealth gap, e_Richesse, being reduced to a wealth gap. injector ε_Richesse_injecteur and calculated according to the following equation taking into account an estimated flow rate in the cylinder Flow_estimé_cylindre and a flow measured in the cylinder Flow_mesuré_cylindre:
    a corrected wealth difference ε_Richesse _corrigé being calculated as a function of the injector richness difference ε _Injector_Richesse according to the following equation:
    being the sensitivity of the richness to the injector gain for the operating point then in force,
    the sensitivity of the richness to the injector offset for the operating point then in effect, parameter _converged_Gain_injector being the parameter of injector gain converged on the operating points seen so far and parameter_convergé_Offset_injecteur being the parameter of injector offset converged for the operating points seen so far.
    Advantageously, it is determined a respective weight for the point of operation then in force that each parameter on its associated deviation, - with the weight of the position of the intake camshaft dephaser Weight vvta_Qcylindre on the flow in the cylinder:
    - with the weight of the position of the exhaust camshaft phase shifter Weight_vse_Qcylinder on the flow rate in the cylinder:
    - with the weight of the valve lift position Weight_Levee_Qcylinder on the cylinder flow rate:
    being the sensitivity of the position of the exhaust phase shifter to the cylinder flow on the selected operating point,
    being the sensitivity of the maximum position of lifting of the valve to the cylinder flow on the selected operating point,
    being the sensitivity of the position of the intake phase-shifter to the cylinder flow rate at the selected operating point, - a weight on the corrected butterfly difference gap ε_Qpapillon_corrected being equal to 1 or Weight _ Section _Qpapillon = - and the deviation of corrected richness ε_Richesse_corrected being distributed on the injector gain Gaininjecteur and injector shift Offset injector with as weight of the injector gain on the wealth Weight_gain_injection_Richesse:
    and as the weight of the injector shift on wealth:
    being the sensitivity of the injector gain to the richness at the selected operating point,
    the sensitivity of the injector shift to the richness at the selected operating point.
    Advantageously, it is deduced from the weights on the operating point in force a value of the parameters to be adapted to absorb the deviations for the respective position parameters of the intake phase shifter Parameter_vvta, the escapement phase-shifter Paramètre_vvte, de the valve lift Parameterjevee, the throttle section parameter Parameter_section, the injector gain Parameter_injector_gain parameter and the injector offset parameter Injector_shift_parameter the following formulas:
    Advantageously, for any converged parameter for a n + 1 nth loop Paramètre_convergé_x_n + 1, this converged parameter is computed with respect to a converged parameter of the previous n loop Paramètre_convergé_x_n on another operating point as a function of the raw parameter of the point operating cost in effect Paramètre_x_n + 1 according to the following equation:
    Parameter __converted_x_n + 1 = Parameter_converted_x_n + Parameter_x_n + 1.ff is a filtering factor excluding a measurement error on the associated multiplicative parameter, the parameter x being the position parameter parameter of the input phase-shifter, the phase-shifter, the valve lift, butterfly section, gain and injector offset, and as a function of the convergence of the measured parameter when a stability criterion of the parameter function of a number of successive converged parameters does not differ from a percentage predetermined is filled, it is estimated that the converged parameter has completed its convergence and the thus converged parameter is taken as adaptive for the associated actuator.
    The invention also relates to an assembly of a powertrain and its engine control unit, the powertrain comprising a motor, an exhaust line, an air intake line with a throttle body and its actuator, an injection line with a fuel injector with its actuator in each cylinder of the engine, the engine being provided with camshaft shift actuators and variable valve lift actuators, characterized in that it implements such a method for a registration in the motor control of the behavior models of actuators, the intake line comprising a flow meter disposed at the throttle body and an air pressure sensor and the line of exhaust comprising an oxygen sensor, the engine control comprising behavior models and means for resetting the behavior models.
    The present invention therefore also provides a system for simultaneous learning actuator models that is to say, camshaft phase shifter intake, camshaft phase shifter exhaust, variable valve intake and exhaust valve, housing butterfly, injector, by operating a flowmeter sensor measuring the air flow at the throttle body, a pressure sensor measuring the pressure in the intake plenum and an oxygen sensor sensor measuring the richness at the exhaust. The present invention thus makes it possible to teach a complex system of actuators with sensors already present in the powertrain, which represents an economy of means.
    Other features, objects and advantages of the present invention will appear on reading the detailed description which follows and with reference to the accompanying drawing given by way of non-limiting example and in which: - Figure 1 is a flow diagram of the registration process according to the present invention.
    In general, the present invention relates to a method for resetting the behavior models of actuators of intake and injection lines of an internal combustion engine of a motor vehicle.
    The actuators can be grouped into two categories. The first category includes the air intake line actuators, namely a throttle valve actuator, camshaft phase shifter actuators, both intake and exhaust camshafts and actuators of the engine. Variable valve lift, both intake and exhaust. The second category comprises the actuator, or the actuators of the fuel injection line to one or respective injectors, with an injector for each cylinder of the engine.
    Sensors or measuring devices already present in the engine assembly taken in its broad sense including the inlet and exhaust lines of the engine are for the intake line a pressure sensor to the intake manifold , a flowmeter at the throttle body and for the injection line an oxygen sensor in the exhaust line and giving the wealth.
    According to the invention, the registration of the behavior models is done simultaneously for the actuators of the intake and injection lines in at least one operating point of the engine with: - for the shaft phase shift actuators to cams and variable valve lift, according to a difference between an actual flow measurement and a model for estimating a flow in at least one cylinder of the engine, - for the throttle body actuator, according to a difference between a measurement actual flow rate and a model for estimating the throttle flow, - for each injector actuator, according to a difference between a real measure of wealth and a richness setpoint.
    The general principle of the method according to the invention will be to take measurements via the sensors in a given zone or learning point with given operating conditions and specific areas or points of the engine operating field to deduce therefrom. temporary values of adaptive model registration.
    These temporary values will be updated from the information of each new measurement, by convergence, and will be applied to the behavior models of the actuators only when a specific criterion of convergence on the adaptives will be established, so that to modify the response of the actuator models only when the correction is deemed robust. Once the adaptive applied on the system, the models of behavior of the actuators will then be recaled closer to the real physics of these actuators.
    Referring to Figure 1, the method according to the present invention can take place in several steps. The first step referenced 1 relates to the selection of the operating point and the taking of measurement. According to an advantageous characteristic of the present invention, it can be observed each operating point of the engine. Once an operating point fulfills all the criteria, it is considered eligible and used in the process. It can then be memorized various parameters and measures to then treat them. It will be paid attention to the operating conditions and the stability of the point. It is indeed important for the method according to the invention to correctly select the points to avoid processing information that does not reflect the actual offsets between the actuator and its behavior model.
    The criterion on the operating conditions to determine whether the engine is in the nominal conditions to initiate an adaptation. The operating conditions may relate individually or in combination to a motor temperature within a predetermined motor temperature range, for example a temperature of 90 ° C, an outside air temperature in an air temperature range. a predetermined outside temperature, for example a temperature that is not too hot above 30 ° C and not too cold below 0 ° C, an atmospheric pressure being within a predetermined atmospheric pressure range, i.e., close atmospheric pressure at sea level, an engine operating mode other than a degraded mode. A degraded mode may be a start phase mode, a catalyst heating mode, or any other mode outside the usual operating range of the engine.
    The stability conditions are indicative that one or more operating parameters of the engine are stable for a predetermined time, for example of the order of one second. This or these parameters can be chosen individually or in combination between a flow of air entering the engine, a engine speed, a position of the actuators of the air intake line to be adapted, a measurement of one or sensors such as flowmeter, air intake pressure sensor or oxygen sensor.
    The second step referenced 2 in Figure 1 relates to a preparation of the treatment of the point. For each point selected during the first step, the process will prepare it before analyzing it and evolve an estimate of the correction of the models of behavior of the actuators as a function of the information provided by this point.
    Based on feedback from the analysis of the sources of dispersions, wear and clogging of the actuators, it is determined the parameters of the actuator models that are to be modulated via one or more elements. adaptive. Thus, by experience, it is considered that the model of behavior of the throttle body can be adjusted with respect to physics by modulating the estimate of the passage section for an opening of a given butterfly flap. An adaptive is therefore defined on the butterfly passage section for the registration of the behavior model of the throttle body.
    For phase shifters of intake and exhaust camshafts, it is considered that it is the phase shift value that must be able to modulate. An adaptive parameter is thus defined on the phase shift value for each type of phase-shifter, thus for both intake and exhaust.
    For the system of variation of the maximum height of the valves, it is considered that it is the maximum value of lift that must be modulated. It is therefore defined an adaptive lift height for each type of valve intake and exhaust, lifting height may be different for these two types of valve.
    The behavior model of the injector that is similar to a straight line can be recaled by modulating the estimate of its gain and its offset. We therefore define an adaptive on the gain and another on the offset.
    It also defines the differences that the process takes into account and seeks to diminish, namely a difference between the flow measurement, resulting from the flow meter and a cylinder flow rate estimate provided by the cylinder flow estimation model, a the difference between the flow measurement, from the flowmeter, and the butterfly flow estimate provided and the butterfly flow estimation model, a difference between the measurement of richness, resulting from the oxygen probe, and the richness setpoint.
    It will first be determined on the selected point the sensitivities of the parameters of the models to adapt to the differences between the measurement and the estimation models. That is, how much is the difference for a given variation of the parameter. This sensitivity makes it possible to make the link with what is measured on the system via the sensors and what must be modified on the model of behavior of each actuator by the adaptive.
    The sensitivities can be determined according to their complexity by several methods. There may be mentioned, for example, a mathematical derivative method of the system state equations, a method of calculating a local variation around the current point and a method by identifying the various sensitivities relative to the actuators of the motor assembly during the engine tuning phase.
    For example, by the first method, it is possible to determine the sensitivity of the flow difference, ie the flow rate measured by the flow meter minus the estimated flow rate by the butterfly model to the variation of the butterfly section Sectionjjapillon. The basis of this method is a known equation that models the behavior of the throttle body, that is to say the estimate of the flow passing at the Qpapillon butterfly, according to the following equation:
    P and PO being respectively the pressures at a given operating point downstream and upstream of the butterfly flap, T0 the temperature upstream of the butterfly flap and
    is a known modeling of the pressure drop at the butterfly. Deriving this equation with respect to the butterfly section, we obtain the following sensitivity:
    This equation allows for each operating point to determine the relationship between a butterfly section variation and the flow variation.
    For the second method, it can be applied, for example in the case of calculating the sensitivity of the cylinder flow at the position of the intake camshaft dephaser. In this case, the application of the first method is possible but proves relatively heavy because the writing of the mathematical derivative of the equation of the cylinder flow with respect to the position of the intake camshaft dephaser is complex. , this position intervening in multiple places in the modeling of the flow cylinder and can pose problem on the derivative of some modeled elements.
    For this specific sensitivity, it is therefore preferable to apply the second method which is to compare on the current point the cylinder flow rate from the modeling, noted Qcylindre in the equation below and that of the same model calculated on the same current point but by adding a bias on the entry for which one wishes to know the sensitivity, noted Qcylindre_biais_VVTA for an intake camshaft actuator in the equation which will be given below. In the case of the example, we introduce a bias on the position of the intake camshaft dephaser which will be noted bias_VVTA in the equation below.
    By the following formula, we obtain the sensitivity of the cylinder flow rate, noted SQcylinder in the equation below, to the variation of the position of the intake camshaft dephaser, noted SWTA
    It should be noted that the method may propose imposing the bias value in a fixed manner or dynamically variable depending on the target that one seeks to achieve, for example, based on the value of the adaptive in the course of calculation.
    The third method is more accurate but more difficult to implement. The third method is at the time of development of the engine to perform specific tests on a test engine placed on the test bench. The tests consist in making variations of the different parameters on each of the focusing points and in deducing from them the real sensitivity of the engine to the various parameters on the point. These sensitivities must then be memorized by operating point or operating zone in the motor control. Thus, when at the stage a point is selected, the method recalls this sensitivity stored for the current point.
    This third method is advantageous since we use the real sensitivities relative to the actuators of the engine, while in other processes we obtain the sensitivities of the model that may be less accurate. However, the additional effort required in the development phase and the complexity of the memory system of these sensitivities in the engine control that this third method is not necessarily preferred.
    From these methods, it is determined the different useful sensitivities of the parameters of the models to adapt to the differences between the measurement and the estimation models. The sensitivity of the cylinder flow is thus determined at the position of the phase shifter.
    , the cylinder flow sensitivity at the position of the exhaust phase-shifter
    the sensitivity of the cylinder flow at the maximum position of the valve lift
    the sensitivity of the butterfly flow to the butterfly passage section
    the sensitivity of the wealth to the injector gain
    , the sensitivity of the wealth to the injection gap
    The third step of the method according to the invention is referenced 3 in the figure and relates to the treatment of the operating point. An error calculation is performed. The purpose of this third step is to derive from the current point the information of registration of available models. The treatment of the point will be made from the sensitivities calculated on the current point and the measurement of the deviations.
    These differences are the following: e_Qcylinder or uncorrected cylinder flow rate difference between the flow measurement from the flowmeter and the flow from the cylinder flow estimation model, ε - QPaPiU ° n or uncorrected butterfly flow gap between the flow measurement, resulting from the flowmeter and the flow rate from the butterfly flow estimation model, ε-Wealth or unrefined correspondence between the measurement of richness, resulting from the oxygen probe and the richness setpoint.
    More precisely, the measurement of the wealth gap ε-Wealth can be a difference on the injector model but also a difference on the cylinder flow model. Indeed, the estimation of the air flow rate at the cylinder level by the cylinder flow rate model is converted into fuel flow which is then converted into control time of the injectors by the injector model. In the context of the present invention, it may be necessary for the differential to refer only to the drift of the injector model, the drift of the cylinder flow model being corrected by another part of the invention.
    It can thus be proceeded to the treatment of the measurement of the wealth gap ε_Richesse in a measure qe the injector richness difference ε_Richesse_injecteur from the cylinder flow estimation information Flow_estimé_cylindre and flow measurement from the flow meter Flow_mesuré_cylindre . Such a treatment can be, for example:
    These differences are not used directly: we will deduce from this difference seen on the current point, the level of correction that the process has already calculated from the points previously treated. The purpose of this calculation is to make the process work only on the rest to be corrected taking into account the correction that has already been identified, but not yet applied to the system.
    For this, we will use converged parameters that will be defined in the fourth step of the process. This convergence will be done in a loop with correction of a parameter converged by a previous converged parameter obtained by a previously noted operating point.
    The impact of these converged parameters on the deviations is evaluated via the sensitivity functions. Thus the cylinder flow difference to be used (ε_Qcylindre_corrigé) is the following
    The difference in butterfly flow to be used (ε_Qpapillon_corrected) is as follows:
    The wealth difference to be used (ε_Richesse_corrected) is the following:
    It will also proceed to the determination on a current point of the weights that each of the parameters have on each of the observed deviations. This gives an image of the distribution of the gap for each of the impacting parameters. Thus, the corrected cylinder flow differential ε_Qcylinder_corrected is distributed on the three parameters which are the position parameters respectively of the intake phase shifter and the intake and exhaust valve exhaust and exhaust phase shifter which are converged for the operating point then in effect.
    So we calculate on the current point a weight of the parameter relative to the position of the intake camshaft phase shifter Weight_vvta_Qcylinder on the flow in the cylinder
    also, a weight of the position of the exhaust camshaft phase shifter Weight_vse_Qcylinder on the flow in the cylinder:
    also, a weight of the valve lift position Weight_Levee_Qcylinder on the cylinder flow:
    being the sensitivity of the position of the exhaust phase shifter to the cylinder flow on the selected operating point,
    being the sensitivity of the maximum position of lifting of the valve to the cylinder flow on the selected operating point,
    being the sensitivity of the position of the intake phase shifter to the cylinder flow on the selected operating point, [0069] The corrected butterfly gap ε_Qpapillon_corrigé being only assigned to the butterfly section parameter, the associated weight to impose is equal to 1 and is not shared among several parameters. We thus have as weight of the parameter section on the butterfly flow:
    The corrected richness difference ε_Richesse _corrected is distributed on the injector Gain injector gain and injector Offset injector offset with the weight of the injector gain on the wealth Weight_gain_injection_Richesse:
    and as the weight of the injector shift on wealth:
    being the sensitivity of the injector gain to the richness at the selected operating point,
    the sensitivity of the injector shift to the richness at the operating point selected, [0071] From these weights, it is possible to deduce on the current point the value of the parameters to be adapted to absorb the differences. The principle consists in determining for each error, that is to say cylinder flow error, throttle valve error, richness error, the part due to each intake and exhaust camshaft actuator position parameter. , intake valve and exhaust valve lift, butterfly section, gain and injector offset and transform this part, using the associated sensitivity function, correction that must be made on the parameter.
    Obtained on an operating point a value of the parameters to be adapted to absorb the deviations for the respective parameters of the position of the phase shifter Parameterjwta, the phase shifter of the escapement Parameter_vte, the valve lift Paramètrejevee, the parameter section Throttle Parameter_section, Injector Gain parameter InjectorGenerator parameter and Injector offset parameter Injector_offset_component parameter according to the following formulas:
    The fourth step referenced 4 in Figure 1 relates to the evolution of the parameters. The purpose of this step will be to take into account the results of the previous step, that is to say the parameter values calculated on the current point to update the converged value of each of the parameters, it is ie the value of the consolidated parameter from all points seen so far. The dotted arrow back to the step referenced 3 from the fourth step illustrates the possible loop to perform a convergence of the parameters.
    For example, for any converged parameter for an n + 1 nth loop Parameter_converted_x_n + 1, this converged parameter can be computed with respect to a converged parameter of the previous n loop Parameter_converged_x_n on another operating point as a function of the raw parameter Parameter_x_n + 1 according to the following equation:
    Parameter_converged_x_n + 1 = Parameter_converged_x_n + Parameter_x_n + 1. ff being a filtering factor excluding an error of measurement on the associated multiplicative parameter, the parameter x being the position parameter parameter of the intake phase shifter, the exhaust phase shifter, the valve lift, the throttle section, the gain and injector offset.
    The filtering factor makes it possible not to take all the information parameter_x_n + 1 provided by the third step. Indeed, due to measurement errors, calculation inaccuracies, the information parameter_x_n + 1 may be tainted by a certain error. With this filtering element, depending on the operating conditions, it is possible to set the information part parameter_x_n + 1 that is used to update the Converged_Parameter_x_n + 1.
    It can thus be ensured that at the beginning of the life of the vehicle, when the manufacturing dispersions make that there are large deviations to catch up, the filter element passes a large proportion of the parameter_x_n + 1 ce which makes it possible to quickly make up for these large differences.
    By cons, as time goes by, we can consider that the bulk of the deviations will have been caught and the remaining gaps will be mainly due to fouling and wear of the actuators. These physical phenomena being generally slow in terms of evolution, we can make sure that the term "factor" strongly filters the evolution of parameter_convergé_x_n + 1 to follow the dynamics of evolution of these phenomena.
    Therefore, for each parameter, the parameters for the position of the intake phase-shifter Parameter_vvta, the exhaust phase-shifter Parameter_vte, the valve lift Parameterjevee, which can be differentiated in intake and exhaust, are calculated from the section parameter of the throttle Parameter_section. , of the Injector Gain Parameter Injector Gain parameter and the Injector Offset Parameter injector offset parameter, a converged parameter value. This converged parameter value indicates the correction necessary for the models of behavior of a respective actuator to better model the physics of each real actuator present on the system.
    The fifth step of the present invention, referenced 5 in Figure 1, relates to the application of adaptive. During this step, the evolution of each of the converged parameters calculated during the fourth step 4 will be analyzed and determined from an evolution criterion if the parameter has reached a stable value or not. As long as the value is not stable, it is considered that the correction that will bring the converged parameter is not sufficiently robust. It is expected that the conditions are realized to go back to the first step 1 and continue to update, from the measurement information, the converged parameter.
    On the other hand, if the converged parameter realizes the criterion, the converged parameter is considered sufficiently robust. It is therefore possible to correct the actuator behavior model from this converged parameter. We stop the evolution of this converged parameter and we memorize its value. It is then defined as an adaptive. This adaptive is applied to the actuator behavior model. This is shown by reference 6 with adap for selecting an adaptive.
    This model then estimates at most just the physical behavior of the actuator. The converged parameter can then be reset to 0 and the conditions are expected to be realized to go back to the first step 1 and restart a calculation sequence of the converged parameter. This makes it possible to continuously analyze the system deviations and continuously calculate corrections to be applied to the behavior models of the actuators.
    An example of an evolution criterion may be that no change is observed greater than A% of the converged parameter at the last three measurement points taken, A being able to be from 1 to 10% without this being the case. limiting.
    The invention also relates to a set of a powertrain and its engine control unit, the powertrain comprising a motor an exhaust line, an air intake line with a throttle body and its actuator , an injection line with a fuel injector with its actuator in each cylinder of the engine, the engine being provided with camshaft shift actuators and variable valve lift actuators, both in intake and 'in exhaust.
    According to the invention, this set implements such a method for a registration in the motor control of actuator behavior models. In this assembly, the intake line comprises a flowmeter disposed at the throttle body and an air pressure sensor and the exhaust line comprises an oxygen sensor, the engine control comprising behavior models and means of registration of behavior patterns.
    The invention is not limited to the described and illustrated embodiments which have been given only as examples.
    权利要求:
    Claims (10)
    [1" id="c-fr-0001]
    1. A method of resetting the models of behavior of actuators of air intake lines and fuel injection of a motor vehicle internal combustion engine, the engine being equipped with controlled actuators comprising phase shift actuators of an intake and exhaust camshaft and variable intake and exhaust valve lift actuators, a throttle valve actuator and an injector actuator for each engine cylinder, characterized in that the registration models of behavior are simultaneously performed for the actuators of the intake and injection lines in at least one operating point of the engine with: - for the actuators of phase shift of a camshaft and variable valve lift of intake or exhaust, according to a cylinder flow rate difference between an actual flow measurement and a flow estimation model in each cylinder, - for the throttle body actuator, according to a throttle flow gap between an actual flow rate measurement and a throttle flow estimation model, - for said at least one injector actuator, according to a difference in richness between a real wealth measurement and a deposit of wealth.
    [2" id="c-fr-0002]
    2. Method according to claim 1, wherein there is defined at least one adaptive for each actuator, said at least one adaptive being: for the throttle body actuator, an adaptive on a section for passage through the throttle, for camshaft phase shift actuators, adaptive to a phase shift value for each type of intake or exhaust phase shifter, - for variable valve lift actuators, an adaptive lift height for each type of the intake or exhaust valve, - for the injector actuator, its model of behavior resembling a line with modulation of an estimate of the gain and an offset, adaptive respectively on the gain and the shift.
    [3" id="c-fr-0003]
    A method according to claim 2, wherein said at least one operating point of the engine is selected according to operating conditions and stability of the point, the operating conditions relating individually or in combination to a temperature of the engine being in a range. a predetermined engine temperature, an outside air temperature within a predetermined outside air temperature range, an atmospheric pressure within a predetermined atmospheric pressure range, a motor operating mode other than a degraded mode , a start phase mode, a catalyst heating mode or any other mode outside the usual operating range of the engine, - the stability conditions attesting that one or more engine operating parameters have been stable for a predetermined duration, this where these parameters are chosen individually or in combination between a die bit of air entering the engine, an engine speed, a position of the actuators of the intake line to be adapted, a measurement of one or more sensors such as flowmeter, air intake pressure sensor or probe to oxygen.
    [4" id="c-fr-0004]
    The method according to claim 2 or 3, wherein, on the selected engine operating point, the respective sensitivities of the behavior models to be adapted according to the differences between each actual measurement and its associated model are determined, an application of a adapting to an associated estimation model being corrected according to the respective sensitivity of the model, with a determination of respective sensitivities of the cylinder flow at the positions of the intake and exhaust phase shifters and respectively at a maximum position of lifting of the valve, a sensitivity of the throttle flow to a throttle passage section, respective sensitivity of the richness to the injector gain and the injector offset, the sensitivities being determined by mathematical derivative of the system state equations, by calculation of a local variation around an operating point or by identification during a pha is the initial focus of the engine.
    [5" id="c-fr-0005]
    5. Method according to claim 4, wherein it is worked by convergence of a respective parameter for the determination of each adaptive, each converged parameter taking into account the previously converged parameters for operating points, and it is calculated: flow differential cylinder ε_Qcylinder_corrected according to the following equation:



    being the sensitivity of the cylinder flow rate to the position of the intake phase shifter at the operating point then in effect,

    being the sensitivity of the cylinder flow to the position of the exhaust phase shifter at the operating point then in effect,

    the sensitivity of the cylinder flow to the maximum valve lift position at the operating point then in effect, c_Qcylinder the uncorrected cylinder flow rate deviation at the operating point then in effect, _convergé_vvta parameter and the_conversion_conversion parameter being the position parameters respectively of the intake phase shifter and the exhaust phase shifter which are converged for the operating points seen so far, and - a corrected butterfly flow gap ε_Qpapillon_corrected according to the following equation:



    the sensitivity of the throttle flow to the butterfly passage section at the selected operating point then in effect, e_Qpapillon the uncorrected throttle valve pitch deviation at the currently selected operating point, parameter_converged_section being the section parameter of the converged throttle passage for operating points seen so far.
    [6" id="c-fr-0006]
    6. Method according to claim 5, in which: a difference between a real measure of richness and a richness setpoint is corrected to take into account only a drift of the injector model, the uncorrected wealth gap, e_Richesse, being reduced to an injector richness difference ε_Injector_Richesse and calculated according to the following equation taking into account an estimated flow rate in the cylinder Estimated cylinder_flow and a measured flow rate in the cylinder Measured_flow_cylinder:

    a corrected richness difference ε_Richesse _corrigé being calculated as a function of the difference in injector richness ε _Injector_Richesse according to the following equation: ε _ Wealth_ corrected = ε _ Injected_wealth_parameter_ parameter_ Gain_injectorx



    being the sensitivity of the richness to the injector gain on the selected operating point then in force,

    the sensitivity of the richness to the injector offset on the selected operating point then in force, parameter _converged_Gain_injector being the converged injector gain parameter for the operating points seen so far and parameter _convergé_0ffsetjnjecteur being the injector offset parameter converged for the operating points seen so far.
    [7" id="c-fr-0007]
    7. Method according to claim 6, wherein a respective weight is determined on the point of operation then in force that each of the parameters on its associated deviation, with the weight of the position of the camshaft dephaser d Admission Weight_vvta_Qcylinder on the flow in the cylinder:

    - with the weight of the position of the exhaust camshaft phase shifter Weight_vse_Qcylinder on the flow rate in the cylinder:

    - with the weight of the valve lift position Weight_Levee_Qcyïmdre on the cylinder flow rate:



    being the sensitivity of the position of the exhaust phase shifter to the cylinder flow on the selected operating point,

    being the sensitivity of the maximum position of lifting of the valve to the cylinder flow on the selected operating point,

    being the sensitivity of the position of the intake phase shifter to the cylinder flow rate at the selected operating point, - with the weight on the corrected butterfly difference gap ε_Qpapillon_corrigé a weight equal to 1 or Weight_Section_Qpapillon = - and the wealth gap corrected ε_Richesse _corrected being distributed on the injector Gain injector gain and the injector Offset injector offset with the weight of the injector gain on the wealth Weight_gain_injection_Richesse:

    and as the weight of the injector shift on wealth:



    being the sensitivity of the injector gain to the richness at the selected operating point,

    the sensitivity of the injector shift to the richness at the selected operating point,
    [8" id="c-fr-0008]
    8. Method according to claim 7, wherein it is deduced, from the weights on the operating point in force, a value of the parameters to be adapted to absorb the deviations for the respective position parameters of the intake phase shifter Parameter_wta, the Exhaust phase-shifter Parameter_veget, valve lift Parameterjevee, throttle section parameter Parameter_section, injector gain Parameter_injector_gain parameter and injector offset parameter Ignition_offset_value the following formulas:


    [9" id="c-fr-0009]
    9. Method according to any one of claims 5 to 8, wherein for any converged parameter for an n + 1th loop Paramètre_convergé_x_n + 1, this converged parameter is computed with respect to a converged parameter of the previous n loop Paramètre__convergé_x_n on a other operating point according to the raw parameter of the operating point in effect Paramètre_x_n + 1 according to the following equation: Parameter_convergé_x_n + 1 = Paramectreconvergé_x_n + Paramètrexn + 1.f fétant a filtering factor excluding a measurement error on the associated multiplicative parameter, the parameter x being the respective position parameter of the intake phase shifter, the exhaust phase shifter, the valve lift, the throttle section, the gain and the injector offset, and depending on the convergence of the parameter, when a criterion of stability of the parameter function of a number of successive converged parameters not differing from a predetermined percentage is filled, it is estimated that the converged parameter has completed its convergence and the thus converged parameter is taken as adaptive for the associated actuator.
    [10" id="c-fr-0010]
    10. A power train and its engine control unit together, the power train comprising an engine, an exhaust line, an air intake line with a throttle body and its actuator, an injection line of fuel with a fuel injector for each cylinder of the engine with its actuator, the engine being provided with actuators of phase shift of a camshaft and variable valve lift actuators, characterized in that it implements a method for resetting in the motor control actuator behavior models according to any one of the preceding claims, the intake line having a flow meter disposed at the throttle body and an air pressure sensor and the line exhaust system comprising an oxygen sensor, the engine control comprising behavior models and means for resetting the behavior models.
    类似技术:
    公开号 | 公开日 | 专利标题
    FR3052189A1|2017-12-08|METHOD FOR REPRESENTING MODELS FOR BEHAVIOR OF ACTUATORS OF INTAKE LINES AND INJECTION OF INTERNAL COMBUSTION ENGINE
    FR2929995A1|2009-10-16|METHOD AND APPARATUS FOR CONTROLLING AN INTERNAL COMBUSTION ENGINE
    EP1729001B1|2008-03-26|Estimation method using a non-linear adaptive filter of air-fuel mixture richness in a cylinder of a combustion engine
    WO2006114550A1|2006-11-02|Method for controlling a motor vehicle using a network of neurones
    EP1729000A1|2006-12-06|Method for estimating the air/fuel ratio in a cylinder of an internal combustion engine using an extended Kalman filter
    EP2748451B1|2016-02-24|Method and system for controlling the operation of a vehicle engine
    JP5120301B2|2013-01-16|Vehicle engine torque calculation device
    FR3057031A1|2018-04-06|METHOD FOR REPRESENTING MODELS FOR BEHAVIOR OF ACTUATORS OF INTAKE LINES AND INJECTION OF INTERNAL COMBUSTION ENGINE
    FR2910549A1|2008-06-27|Injector drift correcting method for e.g. diesel engine of vehicle, involves determining drift correction value of injector to be added to nominal set point value by utilizing indicated average torque estimation for cylinders
    FR2917462A1|2008-12-19|Injector's drift correcting method for e.g. Petrol engine, of vehicle, involves estimating minimal activation time using estimation of indicated average torque, and initializing each injector to zero value of efficient activation time
    EP2404047B1|2017-01-25|Method for processing a signal from a flow meter for measuring a gas flow in an internal combustion engine
    FR3073570B1|2019-10-11|PROCESS FOR CORRECTING ENGINE WEALTH
    FR2910550A1|2008-06-27|Injector drift correction method for e.g. motor vehicle's direct injection oil engine, involves comparing actual injected and set point fuel quantities, and determining correction to be carried out to set point by using torque estimation
    EP3215727B1|2020-07-08|Method of estimation of a intake gas throttle position for control of an internal combustion engine
    FR2917459A3|2008-12-19|Drift and dispersion correcting method for vehicle, involves comparing two air flows, and calculating correction to be subjected to first air flow that is measured by using air flow measuring device
    FR2910551A1|2008-06-27|Injector's drift correction method for e.g. spark ignition oil engine, of motor vehicle, involves determining correction to be carried out to set point quantity of fuel to be injected by injector, by using average torque estimation
    FR3055667A1|2018-03-09|METHOD FOR MANAGING THE POWER SUPPLY OF A THERMAL MOTOR, AND COMPUTER IMPLEMENTING SAID METHOD
    EP3619414A2|2020-03-11|Method for updating a dynamic of adjustment of an air-fuel ratio value to a target value in an engine
    EP3619412B1|2021-01-13|Method for filtering an air-fuel ratio signal produced by an engine exhaust sensor
    FR3052206A1|2017-12-08|CALIBRATION OF AN EMBARGO THERMAL ESTIMATOR FOR CLUTCH
    FR2917461A1|2008-12-19|Fuel drift correcting method for injector of motor vehicle's oil engine, involves estimating control voltage by estimating indicated average torque from characteristic quantity of rotation movement of engine
    WO2008043933A1|2008-04-17|System and method for monitoring the operation of an internal combustion engine with compensation for errors in the intake airflow measurement
    FR3062884A1|2018-08-17|CONTROL METHOD FOR CALCULATING THE SENSITIVITY OF A BEHAVIOR MODEL OF A MOTOR VEHICLE MOTOR ACTUATOR
    FR2934642A1|2010-02-05|Injector's dead time correction method for e.g. direct or indirect fuel injection type petrol engine, of motor vehicle, involves padding cartography of dead time or corrected minimum activation time based on fuel pressure of ramp
    FR2944561A3|2010-10-22|Method for adjusting regulator with state parameter i.e. particle filter/nitrogen oxide trap output gas temperature, in electronic control unit of internal combustion engine of motor vehicle, involves calculating parameters of corrector
    同族专利:
    公开号 | 公开日
    FR3052189B1|2018-06-15|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    EP0670419A2|1994-02-04|1995-09-06|Honda Giken Kogyo Kabushiki Kaisha|Air/fuel ratio estimation system for internal combustion engine|
    GB2362226A|2000-05-13|2001-11-14|Ford Global Tech Inc|Feed forward observer based control for estimating ic engine cylinder air charge|
    DE10224213C1|2002-05-31|2003-10-09|Siemens Ag|Regulating combustion air filling of internal combustion engine, involves tuning model using measurement and model values, deriving actuator element desired values using inverted version of tuned model|
    WO2013026970A1|2011-08-23|2013-02-28|Valeo Systemes De Controle Moteur|Method and system for controlling the operation of a vehicle engine|FR3095009A1|2019-04-09|2020-10-16|Psa Automobiles Sa|PROCESS FOR CORRECTING A RICH FUEL DURING A COLD START|
    WO2020245522A1|2019-06-04|2020-12-10|Psa Automobiles Sa|Method for correcting the richness of fuel during a cold start of a heat engine|
    WO2021205089A1|2020-04-07|2021-10-14|Psa Automobiles Sa|Method for correcting the richness of an air/fuel mixture supplying an internal combustion engine|
    法律状态:
    2017-05-22| PLFP| Fee payment|Year of fee payment: 2 |
    2017-12-08| PLSC| Search report ready|Effective date: 20171208 |
    2018-05-25| PLFP| Fee payment|Year of fee payment: 3 |
    2020-05-20| PLFP| Fee payment|Year of fee payment: 5 |
    2021-05-19| PLFP| Fee payment|Year of fee payment: 6 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1654948|2016-06-01|
    FR1654948A|FR3052189B1|2016-06-01|2016-06-01|METHOD FOR REPRESENTING MODELS FOR BEHAVIOR OF ACTUATORS OF INTAKE LINES AND INJECTION OF INTERNAL COMBUSTION ENGINE|FR1654948A| FR3052189B1|2016-06-01|2016-06-01|METHOD FOR REPRESENTING MODELS FOR BEHAVIOR OF ACTUATORS OF INTAKE LINES AND INJECTION OF INTERNAL COMBUSTION ENGINE|
    [返回顶部]