FR3051569A1 - METHOD AND SYSTEM FOR MONITORING AN ELECTROMAGNETIC VALVE ACTUATOR OF A THERMAL MOTOR WITH AN OPTIMIZED DAMAGE LAW - Google Patents

Abstract

The invention relates primarily to a method of controlling an electromagnetic actuator (9) for a thermal engine valve without a compression spring, comprising: a step of calculating a position command (Zref (k)); a step of calculating a pre-command (PRE), - a step of calculating a regulation term (REG) on a difference between said position command (Zref (k)) and a measured position (Zmes (k) )), - a step of calculating a command of said actuator (9) from the sum of said pre-command (PRE) and said regulation term (REG), and - a step of controlling said actuator (9) from the calculated command, characterized in that said method further comprises a step of limiting a slope (Lim P) of said position setpoint (Zref (k)).

Description

METHOD AND SYSTEM FOR CONTROLLING AN ELECTROMAGNETIC VALVE ACTUATOR OF A THERMAL MOTOR IN LAW

OPTIMIZED ACCOMMODATION

The present invention relates to a method and a control system of an electromagnetic valve actuator of a thermal engine optimized docking law. The invention finds a particularly advantageous application for automotive thermal engines.

The methods in the state of the art concerning the control of actuators with moving magnets and fixed coils are based on the follow-up of the instructions at intermediate levels of positions of the levies of the valves as well as on depreciation strategies. specific, related to the need to damp a moving mass in general of the order of a few hundred grams, consisting of the movable elements, the valves, and a compression spring for energy storage.

The defects of these methods are directly related to the configuration of actuators with moving magnets and storage springs, insofar as the models are non-linear. As a result, there is a lack of precision in the monitoring of the valve lift setpoint. These methods also make it difficult to meet acoustic noise level criteria (slamming) during the movement of the permanent magnets and when docking the valves on the valve seats during closure.

The invention aims to effectively remedy at least one of these disadvantages by providing a method of controlling an electromagnetic valve actuator of a heat engine without compression spring, comprising: - a step of calculating a setpoint of position, - a step of calculating a pre-command, - a step of calculating a regulation term on a difference between the position command and a measured position, - a step of calculating a command of the actuator from the sum of the pre-order and the regulation term, and a step of controlling the actuator from the calculated command, characterized in that the method further comprises a step of limiting a slope of the position command.

The invention thus optimizes the energy consumed by the actuator and take into account the noise constraints when docking the valves on their seat (limiting the docking speed). In addition, the invention solves the problem of compliance with the speed criterion when docking the valve on its seat and the use of energy just needed in this phase of docking and closing the valve.

According to one implementation, the slope of the position command is variable depending on a engine speed.

According to one embodiment, the slope of the position setpoint varies from a point dependent on the engine speed, the slope gradually varying at this point to a limit of authorized slope abutment.

According to one implementation, the limit of permitted slope abutment depends on the engine speed.

According to one implementation, the position setpoint is calculated according to a parameterizable profile adapted to the operation of the heat engine.

According to one implementation, the pre-order is calculated from, in particular, the position setpoint and a pressure measured at the valve.

The invention also relates to a control system of an electromagnetic valve actuator of a heat engine without a compression spring adapted to implement the method as defined above.

The invention will be better understood on reading the description which follows and the examination of the figures that accompany it. These figures are given for illustrative but not limiting of the invention.

Figure 1 is a perspective view of an electromagnetic valve actuator of a heat engine implementing the method according to the present invention; Figure 2 is a schematic representation of the modeling of the electromagnetic actuator according to the present invention; Figure 3 is a schematic representation of a control model of a position command according to the present invention; Figure 4 is a graphical representation of a valve lift setpoint profile with speed limitation according to the present invention; FIGS. 5a, 5b, and 5c are graphical representations illustrating a control of the position of the valve with a regulator, a non-linear pre-control without pressure measurement, and with the modified setpoint, for an engine speed. 6000 rpm; FIGS. 6a to 6d are graphical representations illustrating a control of position of the valve position with a regulator, a non-linear pre-control without measurement of the pressure, and the modified setpoint, for an engine speed of 6000 revolutions / min, with a threshold position of 0.5 mm; FIGS. 7a to 7d are graphical representations illustrating a control of the position of the valve with a regulator, a non-linear pre-control without pressure measurement, and the setpoint modified with the variable slope limiter, for a engine speed of 6000 rpm and with a slope variation time of 0.75 ms; FIGS. 8a, 8b and 8c are graphical representations illustrating a control of the position of the valve with a regulator, a non-linear pre-control without measuring the pressure, and the setpoint modified with the variable slope limiter. , for a revs of 6000 rpm and with a slope variation time of 0.8 ms; FIGS. 9a and 9b are graphical representations illustrating a control of the position of the valve with a regulator, a non-linear pre-control without measuring the pressure, and the setpoint modified with the variable slope limiter, for a engine speed of 1000 rpm; FIGS. 10a to 10d are graphical representations illustrating a follow-up of the position of the valve position with a regulator, a non-linear pre-command without measurement of the pressure, and the setpoint modified with the variable slope limiter, for a engine speed of 1000 rpm and a threshold position of 0.2 mm; FIGS. 11a, 11b, and 11c are graphical representations illustrating a control of position of the valve position with a regulator, a non-linear pre-control without measurement of the pressure, and the modified setpoint with the variable slope limiter at a revs of 6000 rpm and a stop at 0.1 mm; FIGS. 12a to 12e are graphical representations illustrating a control of the position of the valve position with a regulator, a non-linear pre-control without measurement of the pressure, and the modified setpoint with the variable slope limiter and the standardization lifting, for a revving speed of 6000 rpm and a stop at 0.1 mm; [0025] FIGS. 13a to 13f are graphical representations illustrating a control of the position of the valve with a regulator, a non-linear pre-control without pressure measurement, and the modified setpoint with the variable slope limiter, for a engine speed of 6000 rpm and a stop at 0.1 mm; FIGS. 14a, 14b, and 14c are graphical representations illustrating a control of the position of the valve with a regulator, a non-linear pre-control without measuring the pressure, and the modified setpoint with the variable slope limiter. for a revs of 1000 rpm, a stop at 0.1 mm, and a plating command equal to -IV; FIGS. 15a to 15d are graphical representations illustrating a control of the position of the valve position with a regulator, a non-linear pre-control without measurement of the pressure, and the modified setpoint with the variable slope limiter, for a engine speed of 1000 rpm, a stop at 0.1 mm, and a veneer control adapted to the engine speed.

FIG. 1 shows an actuator to be driven 9 comprising a fixed portion of magnetic circuit having permanent magnets in the assembly 10. Movable assemblies 11 and 12 consist of valves 14, composite hollow carcasses 16 and 17. which carry coils 20, 21 and 22, 23. In this type of actuator design 9, the coils are on moving assemblies 11, 12 of small masses, for example of the order of 30 g per moving assembly 11 or 12. This actuator 9 does not have a mechanical spring with a kinetic energy storage function. It is not necessary to control the movements to manage the compression of mechanical elements. Reference may be made to PCT / FR2015 / 052441 and PCT / FR2015 / 052440 for details on the configuration of the actuator 9.

As is modeled in FIG. 2, the control principle of the actuator 9 is based solely on the variation of the intensity of the current I that passes through the coils:

F = B (z) IT

Where F is the force in Newton N, B (z) is the magnetic field generated in Tesla T as a function of the z position in millimeter mm, I is the current intensity in Ampere A, and L is the length of the coils in millimeter mm.

The general principle of the control law is achieved as follows. As shown in FIG. 4, the valve lift setpoint profile 14 to be followed is determined from the opening and closing cycles of the valves 14 necessary for the admission of air or the exhaust of burned gas. , the internal recycling of burned gas IGR (or "Internai Gas Recirculation" in English) or cylinder deactivation.

The control device in position z of the valves 14 to respond to these various problems is as follows (see FIG. 3): From the crankshaft angle, the position reference Zref (k) is calculated according to a parametric profile adapted to the operation of the engine and introducing more or less crossing between intake and exhaust.

The control of an actuator 9 is calculated as the sum of two terms: - a precommand PRE which is obtained from the position reference Zref (k) and the pressure Fpres (k) measured at the valve 14 by "reversing" the transfer function between the control, the pressure Fpres (k) and the position z of the valve 14 and limiting the slope of the position setpoint Zref (k), which varies according to the engine speed Nmot (k ). The slope varies progressively from a point dependent on the engine speed Nmot (k) up to a limit of permitted slope abutment also dependent on engine speed Nmot (k); a regulation term REG on the difference between the position reference Zref (k) and the measured position Zmes (k). The position setpoint Zref (k) may optionally be filtered to request the regulation a physically attainable setpoint tracking. This regulation term makes it possible to compensate for any deviations related to modeling errors or parameter variations (robustness of the setpoint tracking and stability).

When the control of the actuator 9 is calculated, the actuator 9 is controlled from this command calculated according to a control voltage U (k).

As for the physical PRE pre-order, the model of the actuator 9 is governed by the following equations: - equation of the electrical part (generation of the current I from the voltage and the counter-electromotive force)

- equation of the electromechanical part (transformation of current I into electromagnetic force)

Where B (z) is approximated by a polynomial (in practice of order 2) in z which can be mapped.

- equation of dynamics Where:

Ppres_gazit) is a signal supposedly estimated or measured,

is a viscous friction,

) is a restoring force of the spring, m is the mass of the moving part of the actuator 9.

The inverse model of the actuator 9 is therefore directly deduced from these equations by expressing U (t) as a function of z (t) and Fpres_gaz (.0 and by discretizing the derivatives at a period Te worth, for example, 50 μο The following equations are then obtained:

The pre-command PRE is therefore a function of (kTe), Fpresg ^^ (k * Te) and their numerical derivatives and has for expression: - we set HkTe) = Kpi {zikTe), F {kTe)) la map obtained from FQcTe) = Kp {zikTe), l {kTe)) by inversion.

In order to limit the noise during the docking of the valve on its seat, the slope of the setpoint Cons is limited to a negative value depending on the engine speed Nmot (k) when the position is less than a threshold about equal to 0.2 mm.

The valve lift setpoint profile then has the shape shown in FIG. 4.

So as to ensure a follow-up of the Cons instruction to the approach of the stops, one can achieve a first approach by limiting the slope Lim_P fixed in the docking area. As shown in FIGS. 5a to 5c, the control law is applied with the regulator REG and the nonlinear PRE pre-control without measurement of the pressure Fpres (k), and with the modified setpoint, for a motor speed Nmot (k) 6000 rpm. Note that Cons denotes the valve lift setpoint, Lev designates the valve lift, V_Act designates the speed of the actuator, S_Vit designates a speed threshold of the actuator, and PJmp designates a point of impact of the actuator. the actuator. The control law respects the speed constraint of the imposed lifting, higher than -0.4 m / s, but does not follow the imposed instruction. Respect for this constraint is therefore not guaranteed.

As shown in Figures 6a to 6d, to ensure that the setpoint tracking is guaranteed before reaching the stop, it must increase the value of the threshold position. If the error on the stop is 0.1 mm at maximum, a threshold position equal to 0.5 mm is assumed, and the control law is applied with the regulator REG, the pre-order PRE nonlinear without measurement of the pressure Fpres (k ), and the modified setpoint, for an engine speed Nmot (k) of 6000 rev / min. Note that Com FF designates the command of the actuator with a predictive system ("feedforward" in English), Corn T designates the total control of the actuator, and E_Bou denotes a loop difference. The setpoint tracking is performed satisfactorily at 0.1 mm from the stop and the constraint on the speed is therefore respected, that is to say less than -0.4 m / s at the point of impact PJmp.

It is possible to carry out a second approach by limiting the slope Lim P variable in the docking zone. As illustrated in FIGS. 7a to 7d, in order to obtain a more monotonous variation of the position, the shape of the setpoint, at the entry of the docking zone, is modified so as to cause the initial slope Penteinn to be reversed to the desired final slope. a given time set lower than the time required to lift the valve Lev to reach the stop with a safety gap of 0.1 mm. We choose a time in practice of 0.75 ms. For this we apply a variable slope limiter:

The control law is applied with the regulator REG, the nonlinear PRE pre-order without measuring the pressure Fpres (k), and the modified setpoint Cons_Lim with the variable slope limiter, for a motor speed Nmot (k) 6000 rpm. The setpoint tracking is performed satisfactorily at 0.1 mm from the stop, and the constraint on the speed is almost respected, less than -0.5 m / s at the point of impact PJmp. In addition, the variation of the lift of the valve Lev is smoother than with a fixed slope.

As is illustrated in FIGS. 8a, 8b and 8c, in the case of a time chosen of 0.8 ms with the same approach, the control law is applied with the REG regulator, the non-linear PRE pre-control without measurement of the pressure Fpres (k), and the modified setpoint Cons_Lim with the variable slope limiter, for an engine speed Nmot (k) of 6000 rev / min. The setpoint tracking is satisfactorily performed at 0.1 mm from the stop and the constraint on the speed is respected, less than -0.4 m / s at the point of impact PJmp. In addition, the variation of the lifting of the valve Lev is smoother than in the previous case. This validates the method at 6000 rpm.

As shown in FIGS. 9a and 9b, the control law is applied with the regulator REG, the non-linear pre-order PRE without measuring the pressure Fpres (k), and the modified setpoint Cons_Lim with the limiter of variable slope. for a Nmot (k) engine speed of 1000 rpm. The setpoint tracking is satisfactorily performed at 1 mm from the stop and the constraint on the speed is respected, less than -0.05 m / s at the point of impact PJmp. In addition, the variation of the lifting of the valve Lev is smoother than in the previous case. This validates the method at 1000 rpm. However, the slope does not need to be limited to 1 mm from the stop and the threshold position must therefore be adapted according to the engine speed Nmot (k).

As shown in FIGS. 10a to 10d, the control law is applied with the regulator REG, the pre-order PRE non-linear without measurement of the pressure Fpres (k), and the modified setpoint Cons_Lim with the limiter of variable slope, for an engine speed Nmot (k) of 1000 rev / min and a threshold position equal to 0.2 mm. The setpoint tracking is satisfactorily performed at 0.1 mm from the stop and the constraint on the speed is respected, less than -0.05 m / s at the point of impact PJmp. The instruction is limited much later, which allows to better respect the initial instruction. This validates the method.

Below is a learning strategy App-mes followed by normalization of the position of the stop, and a plating on the stop in an operation of the actuator 9 open loop.

Beyond the setpoint monitoring with a limitation of the slope Lim P near the stop, it is necessary to add a specific device for plating the valve Pla_S on the stop. Indeed, if we assume that the stop is at the position 0.1 mm instead of 0 mm, and if we apply the control law with the limiting of the slope Lim_P setpoint, it is observed that the applied command decreases until you reach a minimum value of -48 V if you wait long enough. This creates unnecessary energy consumption, this effort not being necessary for plating on the abutment, and excessive force on the abutment can eventually damage it.

As shown in FIGS. 11a-11c, the control law is applied with the regulator REG, the pre-order PRE nonlinear without measurement of the pressure Fpres (k), and the modified setpoint Cons_Lim with the limiter of variable slope, for a Nmot (k) engine speed of 6000 rpm and a stop at 0.1 mm. The E_Bou loop offset at the stop is 0.1 mm and the control decreases to -20 V. In addition, the maximum loop error E Bou due to the error on the stop value is 0.7 mm.

To overcome the delay created by the loop gap E Bou due to the variation of the position of the stop, the following device is used: - we learn before the start of the heat engine the abutment position of the valve val_butee ; the measured position Znew (k) is recalculated from the measured value Zmes (k) by normalizing the position of the stop at 0 mm. For this, we apply the formula:

Znew (k) = Zmes (k) -val_butee [0049] As is illustrated in FIGS. 12a to 12e, the control law is applied with the regulator REG, the pre-order PRE nonlinear without measurement of the pressure Fpres (k ), and the modified setpoint Cons_Lim with the variable slope limiter and the normalization of the lifting of the valve Lev, for an engine speed Nmot (k) of 6000 rev / min and a stop at 0.1 mm. Since the setpoint Cons and the normalized position Lev Norm are equal, the initial command no longer takes negative values, which has the effect of making the setpoint tracking at the beginning of the cycle more dynamic and thus reducing the loop gap E Bou maximum at 0.18 mm. On the other hand, when the valve is in abutment, the control no longer goes to the value of -20 V but oscillates around 0 V. The convergence towards the setpoint abutment is very slow, as shown in Figure 12c.

So as to avoid this oscillation without applying excessive force which generates unnecessary consumption, it switches to an open loop control when the setpoint is decreasing and less than 0.1 mm. As shown in FIGS. 13a to 13f, the control law is applied with the regulator REG, the nonlinear PRE pre-order without measurement of the pressure Fpres (k), and the modified setpoint Cons_Lim with the variable slope limiter, for a Nmot (k) engine speed of 6000 rpm and a stop at 0.1 mm. The gap E Bou loop abutment is 0.1 mm and the command is then worth the plating value -1 V. It is observed that the docking is fast and without oscillation of the control whose value is constant at -1 V. The maximum loop gap E Bou is less than 0.18 mm, and the constraint of -0.4 m / s of docking speed is respected.

As illustrated in FIGS. 14a to 14c, the control law is applied with the regulator REG, the pre-order PRE non-linear without measurement of the pressure Fpres (k), and the modified setpoint Cons_Lim with the limiter of variable slope, for an engine speed Nmot (k) of 1000 rpm and a stop at 0.1 mm. The veneer control is chosen equal to -1 V. With a veneer control applied for a reference value less than or equal to 0.1 mm, the behavior is satisfactory outside the speed constraint which is not respected (-0.05 m / s).

To solve this problem, the value of the set point below which the plating command is applied must be adapted to the engine speed Nmot (k). For a motor speed Nmot (k) of 1000 rpm, a position of 0.01 mm is chosen. The docking speed constraint is then respected, as shown in FIGS. 15a to 15d.

Claims (7)

claims 1. A method for controlling an electromagnetic actuator (9) for a valve (14) of a heat engine without a compression spring, comprising: a step of calculating a position command (Zref (k)), a step for calculating a pre-command (PRE), - a step of calculating a regulation term (REG) on a difference between said position command (Zref (k)) and a measured position (Zmes (k)) a step of calculating a command of said actuator (9) from the sum of said pre-order (PRE) and said regulation term (REG), and a step of controlling said actuator (9) from the calculated control, characterized in that said method further comprises a step of limiting a slope (Lim P) of said position setpoint (Zref (k)). 2. Method according to claim 1, characterized in that said slope of said position command (Zref (k)) is variable as a function of engine speed (Nmot (k)). 3. Method according to claim 2, characterized in that said slope of said position setpoint (Zref (k)) varies from a point dependent on said engine speed (Nmot (k)), said slope gradually varying at this point. up to a limit of permitted slope in abutment. 4. Method according to claim 3, characterized in that said limit of permitted slope abutment depends on said engine speed (Nmot (k)). 5. Method according to any one of claims 1 to 4, characterized in that said position setpoint (Zref (k)) is calculated according to a parameterizable profile adapted to the operation of said engine. 6. Method according to any one of claims 1 to 5, characterized in that said pre-command (PRE) is calculated from said position setpoint (Zref (k)) and a pressure (Fpres (k)). )) measured at said valve (14). 7. Control system of an electromagnetic actuator (9) for a valve (14) of a heat engine without a compression spring adapted to implement said method as claimed in any one of claims 1 to 6.