WO2008071597A1

WO2008071597A1 - Fuel supply system for an internal combustion engine and related control method

Info

Publication number: WO2008071597A1
Application number: PCT/EP2007/063258
Authority: WO
Inventors: Olivier Ponsonnaille; Bertrand Carre; Richard Roth
Original assignee: Renault S.A.S
Priority date: 2006-12-12
Filing date: 2007-12-04
Publication date: 2008-06-19
Also published as: EP2102476B1; FR2909724A1; EP2102476A1; FR2909724B1

Abstract

The invention relates to a fuel supply system for an internal combustion engine comprising a pump system (5) provided an injection rail (4) and a fuel tank (6), with at least one actuator (9) the opening of which is controlled by a signal of the pulse modulation type, in order to carry out successive adjustments of the opening of said actuator using a control magnitude depending on a cycle ratio setpoint (RCO), and an electronic control unit (11). The electronic control unit further includes auxiliary means (42) for increasing the amplitude of the oscillations of said control magnitude about an average value if the value of said cycle ratio setpoint is comprised in a given critical zone.

Description

Fuel supply system for an internal combustion engine and corresponding control method.

The invention relates to the field of gasoline or diesel engines, with n cylinders, equipped with an injection system (also called fuel supply system), and in particular the injection systems involving at least one actuator. (or regulator) the flow and / or the pressure of the fuel. More specifically, the invention relates to gasoline or high pressure diesel direct injection systems, commonly known as "Common Rail" ("Common Rail") systems.

Conventionally, if there is a difference between the desired rail rail pressure and the actual measured pressure, the actuator is controlled to adjust the fuel flow through the pump (as part of a flow actuator) so that the measured pressure reaches the desired pressure.

In order to do this, the actuators are generally controlled using PWM (Pulse Modulated Modulation) type control signals (in English). A control signal may be a control current proportional to the electromotive force allowing the displacement of the actuator.

However, the reaction time of the actuator relative to the control can be particularly long because, for example, microblocking of internal parts due to the misalignment of the internal parts of the actuator, or to the wear of the guide of these pieces.

No. 4,020,654 discloses a process for improving the properties, in particular dynamic properties, of this regulation of the flow (or pressure) of the fuel by modifying the control signal of the actuator. However, this method requires, on the one hand, modifying the calculation unit of the control signal and, on the other hand, adding a device measuring the arrangement of the actuator member to be adjusted, which is particularly expensive and cumbersome.

The invention aims to provide a solution to this problem.

An object of the invention is therefore to propose a fuel supply system, in which the integrated actuator (s) within the pump are particularly responsive to the control signal for regulating the flow (or pressure) of the fuel.

For this purpose, a first aspect of the invention relates to a fuel supply system for an internal combustion engine, comprising a pump system arranged between an injection ramp and a fuel tank, with at least one actuator whose opening is controlled by a pulse modulation type signal, so as to make successive adjustments of the opening of said actuator by means of a control variable according to a cycle ratio setpoint.

This cycle report setpoint reflects the duration during which the control signal of the opening of the actuator is applied.

The power system also includes an electronic control unit.

According to a general characteristic of this first aspect of the invention, said electronic control unit further comprises auxiliary means capable of increasing the amplitude of the oscillations of said control variable around an average value, if the value of said Cycle report setpoint is included in a given critical area. In other words, if the cycle ratio setpoint reaches too large or too low values, the amplitude of the oscillation of the control variable of the actuator is increased. Thus, the actuator is no longer maintained in a fixed position, but is continuously moving around an average position, which greatly reduces the reaction time of the actuator.

Indeed, the inventors have observed that the reaction time was particularly long when the control variable was weakly oscillating. This case corresponds to particularly high or particularly low opening duty cycle values.

More precisely, as mentioned above, the variation

(oscillations) required of the electro-motor force around its mean value to ensure the displacement of the actuator is directly proportional to the changes in the control current which is the current flowing through the coil of the actuator.

Indeed, the relation between the electromotive force and the current passing through the coil of the actuator is written:

Fem = kIpwM, with:

Fem: the electromotive force, k: a coefficient of proportionality, I _P W _M : the current flowing through the coil of the actuator. In order for the actuator to move, the oscillations of the electromotive force must be greater than the different friction forces (dry friction or Stribeck effect).

To do this, the current in the coil must have sufficiently large amplitude variations (or oscillations). If this is not the case the electro-motor force fails to "fight" against friction and response time of the actuator increases. This situation is illustrated in FIGS. 9 and 10. In these figures, it can be seen that for a 50% cycle ratio setpoint (curves referenced RCO501 and RCO502), the amplitude of the control current (curves referenced I _PWM 501 and I _PWM 502), which is proportional to the corresponding electromotive force, is sufficient to cause a continuous displacement of the actuator around an average position. The opposing forces (friction forces) are thus permanently overcome and if the setpoint of the actuator changes, it will quickly reach the value corresponding to the new setpoint.

On the other hand, if the cycle ratio setpoint is greater (90% in FIG. 9, curve referenced RCO90) or lower (10% in FIG. 10, curve referenced RCO1 O) at a certain threshold, the amplitudes of the control currents (respectively referenced I _PWM 90, and I _PWM IO) are no longer sufficient. The actuator is therefore at a standstill because the variations of the electromotive force (corresponding to the control current) around the average value are no longer sufficient to overcome the friction forces.

In this situation, if the actuator displacement instruction changes, it will first have to overcome the friction forces before reaching this new setpoint. The response time of the actuator is therefore considerably increased.

The invention thus makes it possible to prevent this situation from occurring, by causing under certain conditions the increase of the amplitude of the oscillations of the control signal of the actuator, and consequently the increase of the oscillation of the electromotive force. corresponding.

According to a preferred embodiment, said auxiliary means are capable of increasing the amplitude of the oscillations of said control quantity by decreasing the frequency of the control signal.

In this case, the system may further comprise first storage means capable of storing the value of a lower limit of the frequency of the control signal, determined according to the characteristics of the fuel supply system of the engine.

The system may also comprise second storage means capable of storing the value of an upper limit of the frequency of the control signal, lower than the natural frequency of said actuator.

According to one embodiment, wherein said pump system comprises an auxiliary pump called "low pressure" coupled between the actuator and said fuel tank. Said auxiliary means may then be able to adjust the frequency of the control signal of the actuator so that it is out of the vicinity of an integer value of the discharge frequency of said low pressure auxiliary pump.

When said pump system comprises an auxiliary pump called "high pressure" coupled between the actuator and the injection rail, said auxiliary means can also be adapted to adjust the frequency of the control signal of the actuator so that it out of the vicinity of an integer value of the suction frequency of said high pressure auxiliary pump. Moreover, said auxiliary means may also be able to adjust the frequency of the control signal of the actuator so that it is different from the natural frequency of the actuator relative to its environment (mechanical, hydraulic). In the case where the pump is coupled to said fuel tank via a return loop comprising a mechanical pressure limiter. Said auxiliary means may then be able to adjust the frequency of the control signal of the actuator so that it is different from the natural frequency of said mechanical pressure limiter relative to its environment (mechanical, hydraulic).

According to another embodiment, the system may also comprise at least one fuel temperature sensor, said auxiliary means then being able to oscillate said quantity also if the value of the fuel temperature is lower than a predetermined critical value.

For example, at least one of the actuators of the pump is a flow and / or pressure actuator. According to a second aspect of the invention there is provided a method of controlling a fuel supply system for an internal combustion engine comprising a pump system disposed between an injection ramp and a fuel tank, said method comprising controlling the opening of at least one actuator of the pump system, by means of a pulse - type type of control variable, so as to make successive adjustments of the opening of said actuator by function of a cycle report setpoint.

According to a general characteristic of this second aspect of the invention, the amplitude of the oscillations around an average value of said control quantity is increased, if the value of said cycle ratio setpoint is included in a given critical zone. Other advantages and characteristics of the invention will appear on examining the detailed description of an embodiment and a non-limiting embodiment, and the appended drawings, in which: FIG. schematically an internal combustion diesel engine;

FIG. 2 schematically illustrates the piston of the actuator of the fuel supply system associated with the engine represented in FIG. 1; FIG. 3 illustrates the evolution of the control current of the piston according to the opening duty cycle;

FIG. 4a illustrates an example of a simplified characteristic curve of a flow actuator for a given pump speed;

FIG. 4b illustrates an example of a simplified inverse characteristic curve of a flow actuator for a given pump speed;

FIG. 5 illustrates the elaboration of the control of the piston of the actuator according to the invention;

FIG. 6 illustrates more precisely the pump system inserted into the fuel supply system of the engine.

Diesel ;

FIG. 7 illustrates in greater detail the high pressure elements of the pump shown in FIG. 6;

FIG 8 illustrates an embodiment of the method according to the invention; and

FIGS. 9 and 10 illustrate the variations of the control current of the actuator according to the prior art.

FIG. 1 is very schematically represented an internal combustion diesel engine referenced la, fed with fuel via a Ib supply system (of course, the invention can also be applied to the gasoline engine). In this example, the engine includes four cylinders. The fuel supply system Ib of this engine comprises four injectors referenced 2, each connected by a high pressure hose 3 to the battery 4 called "the rail".

The fuel supply system Ib also comprises a pump 5, which draws fuel from the tank 6 of the vehicle via a low pressure circuit 7. The function of the pump 5 is to compress the fuel taken from the tank 6 of the vehicle and push it back into the rail 4.

As will be seen in more detail below, the pump 5 is composed of a low pressure part (transfer pump) and a high pressure part (high pressure pump). Note that the pump 5 is also connected to the tank 6 via a return circuit 8. This return circuit 8 serves to regulate the leakage rate of the high pressure part, as well as the lubrication and cooling rate.

A flow actuator 9 (or flow valve) is integrated with the low pressure side pump 5. This flow actuator 9 makes it possible to adjust the amount of fuel that will be sent to the rail 4, to which the pump 5 is connected via a connection 10.

In addition, the feed system Ib may also be equipped at the outlet of the pump 5 on the side of the connection 10, or directly on the rail 4, with a pressure actuator also called a discharge actuator. For simplification purposes, this pressure actuator is not shown.

Finally, the engine 1a and its power supply system Ib are controlled by an electronic control unit or computer 11. This computer 11 comprises conventional components, such as microprocessors, EEPROM type hard memories and RAM type buffers.

Furthermore, it receives input information 12 via a connection 13. This information 12 comes from different sensors placed on the engine and its ancillary systems, such as the injection system, the air supply system, etc. .

The computer 11 then processes the input data 12 to define or calculate control levels 14 outputted from the computer 11 via a connection 15. The control levels 14 are sent to the various actuators that participate in the control of the ancillary systems and therefore of the motor. More particularly, these control levels are transmitted via a connection 16 to the injectors 2, and a connection 17 to the actuator 9.

More specifically, the information 12 transmitted to the computer 11, such as, for example, the engine coolant temperature, the engine speed, the engine oil temperature, the air temperature in the engine, the pressure turbocharger air, the position of the accelerator pedal, etc. , are processed via functions or maps stored in an EEPROM type memory, for example.

These maps make it possible to define the value of the desired PCONS pressure in the rail 4. The control levels 14 are then calculated as a function of this pressure value PCONS.

Conventionally, the value of the PCONS pressure setpoint is compared with the value of the effectively measured pressure PMES in the rail 4. This value PMES is delivered via a connection 18 to the computer 11, in parallel with the information 12.

According to the pressure difference ΔP identified (with ΔP = PCONS - PMES), the lial controller 'for example a software corrector of the PID (Proportional Integral Derivative) type, adjusts the control signal of the flow actuator 9 so that the measured pressure PMES rejoins the PCONS pressure setpoint. If the difference ΔP is positive, the controller 11 acts to increase the flow (in the case of a flow actuator). Conversely, if ΔP is negative, the controller acts to decrease the flow.

Generally, actuator control signals are PWM (Pulse Modulated Modulation) type (PWM). The control signal is characterized by its nominal voltage U _P W _M and its frequency F _P W _M - The percentage of the duty ratio (RCO) is between 0 and 100%.

This translates the duration during which the voltage is applied

UPWM over the TPWM period (TPWM equals 1 / FPWM) - The intensity of the average current I _PWM created by the PWM type control signal is then proportional to the percentage of the duty cycle.

BCR.

In a simplified manner, as already mentioned above, it can be said that the average current I _PWM created by the control signal generates an electromagnetic force Fem as shown in FIG. 2.

This electromagnetic force Fem is proportional to the percentage of the opening duty ratio RCO, which allows the displacement of an internal piston 20 (or needle) of the flow actuator 9. The position of the piston 20 determines the passage section ( and therefore the flow rate) of the fuel entering the pump via inlets 21 and 22.

The fuel leaves through an opening referenced 23.

A friction force Ff opposes the electromagnetic force Fem. This friction force Ff results in example of a misalignment of parts or wear of the guidance of parts.

As represented in the upper part of FIG. 3, the percentage of the opening duty ratio RCO applied to the modulating control signal of the actuator makes it possible to establish a more or less large average current I _P W _M.

The upper part of Figure 3 corresponds to an opening duty cycle of 30% and the lower part to an opening duty cycle of 90%. At this average current I _P W _M corresponds the average electromagnetic force Fem, which makes it possible to control the displacement and the position of the piston of the actuator, thus fixing the desired passage section for the flow of the fuel (more or less important flow ). By way of example, a simplified characteristic curve of a flow actuator is shown in FIG. 4a. This characteristic corresponds to a given pump speed, knowing that the pump is rotated by the motor with a drive ratio R. Thus, the higher the flow demand (in liters / minute), the higher the duty cycle. RCO opening (for a fixed PWM amplitude modulation frequency) to be applied will be important.

There are so-called inverse characteristics, such as that shown in FIG. 4b, where, unlike that represented in FIG. 4a, the lower the flow demand, the greater the opening RCO duty cycle to be applied.

Thus, as illustrated in Figure 3, when the RCO is low

(in this example 30%), the average current I _P W _M corresponding to the UW _M voltage for a given RCO, oscillates around an average position.

As explained above, in the general case, if the duty cycle RCO is particularly high or particularly low, the average current stabilizes and the flow actuator will have a rather fixed position. This considerably increases the reaction time of the actuator during a change of operating point of the engine (by action of the driver on the accelerator pedal, for example) due to the difficulty of overcoming the friction force Ff.

Therefore, when the opening duty ratio RCO reaches a high value (in this example 90%) or low (in this example 10%), the invention makes it possible to make the current oscillating, with a sufficient amplitude of oscillation. A particularly advantageous solution for a PWM type signal is to increase the _PWM period T _PWM control signal (as illustrated on the lower part of Figure 3 for RCO = 90%), which therefore makes the average current I _P W _M oscillating again. The electromagnetic force Fem generated is then also oscillating. Thus, the position of the piston of the actuator is no longer maintained at a fixed position, but is set in motion around a mean position. The generation of these micromovements of the piston of the actuator by variation or adaptation of the frequency of the PWM type control signal in the flow zone or areas deemed critical (corresponding to a high or low opening ratio RCO) allows to greatly reduce the reaction time because of the increased facility to overcome the frictional force Ff. The invention makes it possible to reduce the setpoint overruns of the pressure within the rail 4.

An embodiment of the actuator control method implemented within the control unit 11 is shown in FIG. 5.

First, the control unit 11 stores a first mapping 30, called open loop mapping, for determining the positioning of the flow actuator according to a DC fuel flow setpoint and the engine speed. of the vehicle Nm respectively delivered at the input of the map 30 via connections 31 and 32. The mapping 30 makes it possible to develop a nominal current setpoint CCN for the control of the flow actuator.

Furthermore, another mapping 33 is able to develop the pressure set point of the rail PCONS as a function of the fuel flow setpoint DC delivered via a connection 34, and the engine speed Nm delivered via a connection 35 as well as the parameters 12 delivered. via the connection 13.

The instruction PCONS is delivered to a comparator 36 which also receives via the connection 18 the pressure PMES measured at the level of the rail 4.

The difference .DELTA.P between the two pressure values PCONS and PMES is delivered, via a connection 36a, to a corrector 37 (for example a corrector of the PID type) which generates a corrective current setpoint ICOR.

This instruction ICOR is delivered to an adder 38 via a connection 39. This adder 38 also receives on another input via a connection 40 the rated nominal current CCN. The output of the adder 38 is connected via a connection 41 to auxiliary means 42 whose role is to determine the frequency F _PWM of the control signal of the actuator and the percentage of the duty cycle. RCO, applied to this actuator.

For this purpose, the auxiliary means 42 also receives the engine speed Nm via a connection 43, the fuel temperature TC via a connection 44, and the value of the fuel flow DC via a connection 45. The role of these variables TC and DC are explained in more detail below.

The auxiliary means 42 derives from all these variables the frequency modulated control signal F _P W _M , and the duty cycle RCO. This control signal is transmitted to the actuator 9 via the connection 15. However, when the frequency of the control signal F _P W _M is modified, in particular for a high or low percentage of RCO, the choice of the frequency must take into account characteristics of the components of the system, to avoid creating instability of the pressure in the rail 4. In fact, contrary to what has been described in the document DE 1 033 04665, the inventors have observed that the presence of The oscillations at the pressure of the rail 4 are not explained solely by the impact of the frequency of the flow of the fuel leaving the rail, but also by the impact of the frequency of the flow entering the rail. , on which the actuator 9 directly influences the frequency F _P W _M of its control signal.

To understand this phenomenon, reference is made to FIG. 6, which illustrates more precisely the pump system 5. The latter comprises a booster pump 50 or a low-pressure pump and a high-pressure pump 51. The flow actuator 9 is disposed between these two pumps 50 and 51, to which it is respectively connected via connections 52 and 53. Optionally, a pressure actuator (not shown for simplification purposes) could be installed at the outlet of the high pressure pump on the connection 10.

The pressure between the flow actuator 9 and the high pressure pump 51 is called P2. The pressure between the booster pump 50 and the flow actuator 9 is called P1.

A mechanical pressure limiter referenced 54 is connected between the connection 8 and the node common to the booster pump 50 and the flow actuator 9.

The limiter 54 is connected to this common node via a connection 55. Note Ql the flow delivered by the booster pump, Q2 the flow rate through the pressure limiter 54, Q3 the flow rate through the flow actuator 9 and Q4 the flow rate admitted by the high pressure pump

51.

For a given pump regime, the volume admitted by each element of the pump system 5 is directly determined by the flow rate

Q4 admitted by the high-pressure pump 51 (considered throughout the duration of admission). This flow Q4 is also directly related to the effective supply pressure P2, both from a static and dynamic point of view (mean value and oscillation frequency). Different interactions exist between the pressure P2 and the flow Q4. In order to identify them, reference is made to FIG. 7 which illustrates in greater detail the high-pressure pump 51.

The high-pressure pump 51 receives as input the setpoint DPS corresponding to the displacement of the piston of the flow actuator 9. This DPS instruction has a direct impact on the flow Q3 passing through the flow actuator 9 and thus on the effective supply pressure P2. An intake valve 60 is connected to a pumping element 61, via a connection 62. The output of the pumping element 61 is connected to another intake valve 63 via a connection 64.

The other intake valve 63 is connected to the rail 4 via the connection 10.

Different interactions exist between the effective supply pressure P2 and the admitted flow rate Q4 by the high - pressure pumping element 51.

When the oscillation frequency of the supply pressure P2 and the pumping frequency (given by the multiplication between the number of elements contained in the high - pressure pump system 51 and the speed of this same high - pressure pump system. pressure 51) are equal (or when they are integer values of each other), the flow rates, and therefore the volume admitted at each stroke of the pump system 5, are stable or repeatable. The average value of the rail pressure (at the level of the elementary pressure discharge cycle) is thus stable. The volume admitted to each shot, then depends on the phasing between the variations of the pressure and the pumping cycle. Indeed, the openings / closures of valves occur at different times in the oscillation cycle of the effective supply pressure P2. When the oscillation frequency of the supply pressure P2 and the pumping frequency are very close (or when they are almost integer values of each other), the flow rate, and thus the volume admitted to each shot Q4, has a low frequency variation. The value of this low harmonic frequency is directly related to the ratio between the two frequencies: we find this low frequency variation in the value of the pressure of the rail 4.

When the oscillation frequency of the supply pressure P2 and the pumping frequency are not integer values of each other, the values taken by the admitted flow rate Q4 at each stroke are fairly dispersed. However, the average value is stable over a much lower cycle number than in the previous case. In view of the foregoing, it appears that the oscillation frequency of the pressure P2 is particularly related to the frequency of oscillation of the flow Q4, itself directly related to the number of elements that comprises the high pressure pump 51 and at the speed of the pump system 5. In addition, the oscillation frequencies of the pressure P2 are related to the oscillation frequency of the flow Q3 and flow Q1.

The frequency of oscillation of the flow rate Q3 depends, of course, on oscillations of the pressure difference across the flow actuator (P2 - P1), but also on oscillations in the passage section of the flow actuator, which are directly related to the frequency of the pulse modulation type control signal.

Finally, the frequency of oscillations of the flow Ql is directly related to the number of elements of the low-pressure pump 50 and the speed of the pump system.

Therefore, it follows from the aforementioned observations of the inventors, that the oscillations of the pressure of the rail 4 are due inter alia: the harmonics of the oscillations related to the discharge carried out by the low pressure pump 50 (related to the number of discharge elements that comprises the low pressure pump and the speed of the pump system 5); the harmonics of the oscillations of the pumping effected by the high-pressure pump 51 (related to the number of elements that the high-pressure pump comprises and the pumping rate); at the frequency F _P W _M of the pulse modulation type control signal of the actuator 9.

Greater stability is obtained when all these frequencies are integer values of each other.

Nevertheless, the value of the flow rate Q4 admitted by the high-pressure pump 51 also depends on the phasing of the oscillations, that is to say of the angular relationship existing between the periods of discharge between at least two outputs of the high-pressure pump. 51.

The choice of the frequency F _P W _M of the pulse modulation type control signal according to the speed of the pump system 5 must therefore fulfill the following conditions: the frequency F _PWM is not too high, that is to say less than the natural frequency of the flow actuator, to guarantee the movement of the latter, the frequency F _P W _M must not be too slow (the lower limit being determined according to characteristics of the system engine fuel supply), so that there is no impact on the rail pressure; the F _PWM frequency is not in the vicinity of an integer value of the suction frequency of the high-pressure pump (taking into account the speed of the pump); the F _PWM frequency must not be in the vicinity of an integral value of the discharge frequency of the low-pressure pump (taking into account the pump speed). Moreover, for reasons of stability of the components and therefore of the behavior of the pump system 5, it is added that the frequency

F _PWM of the _PWM control signal as a function of the pump speed shall also guarantee: not to be equal to the eigenfrequencies of the flow actuator (nor to the vicinity), so as to control the amplitude of the flow oscillations; not to be equal to the natural frequencies of the pressure limiter 54 (or in the vicinity), so as to control the amplitude of the oscillations of the pressure upstream of the flow actuator 9. The invention makes it possible to determine a frequency F _PWM of the pulse modulation type control signal, achieving the best compromise on all of these criteria. One mode of implementation is detailed below.

On the other hand, the inventors have observed that it is very advantageous to also take into account the fuel temperature conditions to modify the frequency F _PWM of the control signal of the actuator, especially in the flow zones deemed critical, when the opening cycle ratio RCO is particularly high or particularly low. Indeed, it is when the temperature of the fuel is particularly low that the reaction time of the actuator is the longest and therefore the most harmful. Thus, in order to minimize the wear that is generated by the generation of micromovements of the actuator pistons, the invention makes it possible to modify the frequency of the control signal when the fuel temperature is particularly low.

To do this, the fuel temperature information TC is supplied to the auxiliary means 42 (see FIG. 1) by means of a sensor placed either at the inlet of the pump 5 or on the return circuit.

8 between the tank of the vehicle 6 and the pump 5 (for simplification purposes, these sensors are not shown).

In order to implement what has been stated above, two maps F1 and F2 are stored in the computer 11. Each map is associated with a function that makes it possible to determine the frequency of the control signal F _PWM to be applied in engine speed function Nm.

The function associated with the mapping Fl gives the frequencies

F _P W _M predetermined (such as F _P W _M = Fl (Nm)), taking into account the specific characteristics of the components of the injection system in order to limit the pressure oscillations in the rail, according to the various conditions set out below. before.

The function associated with the mapping F2 gives the predetermined frequencies F _P W _M , (such that F _P W _M = F2 (Nm)) lower than those provided by the function associated with the mapping F1 and used when: the flow instruction DC fuel is above a calibratable critical threshold DCcr2 (eg 90%) corresponding to zones of high flow in the case of actuators with "normal" characteristic (corresponding to low flow zones in the case of actuators with so-called "inverse" characteristic), the setpoint of the fuel flow rate DC is below a calibratable critical threshold DCcrl (for example 10%) corresponding low flow areas in the case of actuators with "normal" characteristics (corresponding to zones of high flow in the case of actuators with "inverse" characteristics), and - when the temperature of the TC fuel is lower than a calibratable critical threshold TCcr (in the case where the engine fuel system includes fuel temperature sensors.

The lower frequencies obtained from the mapping F2 will then allow the generation of micromovements of the piston of the actuator around a mean position. These micromovements play in favor of a strong reduction of overshoots of the rail pressure setpoint in the areas considered critical. For this purpose, the flowchart of FIG. 8 can be implemented within the auxiliary means 42.

At start-up 100 of the vehicle, the flow rate of the fuel DC is compared with the critical values DCcrl and DCcr2 and the fuel temperature TC is compared with the critical temperature TCcr of the fuel, step 101. If the fuel flow is higher than the critical value associated DCcr2 or lower than the associated critical value DCcrl, and if the fuel temperature is lower than the associated critical value and if the fuel temperature is lower than the associated critical value, the frequency F _PWM of the actuator control signal is developed from the mapping F2, step 102.

Otherwise, the frequency F _P W _M is developed from the mapping F1, step 103. Once the development of the frequency F _{PWM is} carried out, if the engine is stopped, step 104, the process is terminated, step 105, otherwise, steps 101 to 104 are repeated.

Of course, the invention is not limited to the modes of implementation and embodiments presented above.

For example, the control of the actuator by varying the frequency F _PWM of the control signal also applies to the pressure actuators that can be integrated into the fuel supply system. In this case, the mimic diagram of the regulation loop represented in FIG. 5 is adapted to the case of a pressure actuator: for example, the maps 30 and 33 are no longer dependent on the engine speed Nm and on the flow setpoint of the engine. DC fuel.

Claims

Fuel system for an internal combustion engine comprising a pump system (5) disposed between an injection manifold (4) and a fuel tank (6), with at least one actuator (9) opening is controlled by a pulse modulation type signal, so as to make successive adjustments of the opening of said actuator by means of a control variable according to a cycle ratio setpoint (RCO) , and an electronic control unit (1 1), characterized in that said electronic control unit further comprises auxiliary means (42) capable of increasing the amplitude of the oscillations of said control variable around a value average, if the value of said cycle report setpoint is within a given critical area.

2- System according to claim 1, wherein said auxiliary means are capable of increasing the amplitude of the oscillations of said control quantity by decreasing the frequency (F _PWM ) of the control signal. 3. System according to claim 2, further comprising first storage means able to store the value of a lower limit of the frequency of the control signal, determined according to the characteristics of the fuel supply system of the engine. 4. System according to one of claims 2 or 3, further comprising second storage means capable of storing the value of an upper limit of the frequency of the control signal, less than the natural frequency of said actuator. 5 - System according to one of claims 2 to 4, wherein said pump system comprises a so-called low-pressure auxiliary pump (50) coupled between the actuator (9) and said fuel tank (6), said auxiliary means ( 42) being further adapted to adjust the frequency of the control signal of the actuator so that it is out of the vicinity of an integer value of the discharge frequency of said low pressure auxiliary pump.

6 - System according to one of claims 2 to 5, wherein said pump system comprises a so-called high-pressure auxiliary pump (51) coupled between the actuator (9) and the injection rail (4), said auxiliary means (42) being further adapted to adjust the frequency of the control signal of the actuator (9) so that it is out of the vicinity of an integer value of the suction frequency of said high pressure auxiliary pump. 7- System according to one of claims 2 to 6, wherein said auxiliary means (42) are further adapted to adjust the frequency of the control signal of the actuator so that it is different from the natural frequency of the actuator (9) with respect to its environment. 8- System according to one of claims 2 to 7, wherein the pump is coupled to said fuel tank via a return loop comprising a mechanical pressure limiter (54), and wherein said auxiliary means (42) are further adapted adjusting the frequency of the control signal of the actuator so that it is different from the natural frequency of said mechanical pressure limiter (54) relative to its environment.

9- System according to one of claims 2 to 8, further comprising at least one sensor of the fuel temperature, and wherein said auxiliary means are able to oscillate said magnitude if the value of the fuel temperature is below a predetermined critical value.

10- System according to one of claims 2 to 9, wherein at least one of the actuators of the pump is a flow actuator. 11 - System according to one of claims 2 to 10, wherein at least one of the actuators of the pump is a pressure actuator.

A method of controlling a fuel supply system for an internal combustion engine comprising a pump system disposed between an injection manifold and a fuel tank, said method comprising controlling the opening of at least one an actuator of the pump system, by means of a pulse modulation type control variable, so as to perform successive adjustments of the opening of said actuator as a function of a cycle ratio setpoint, characterized in that the amplitude of the oscillations around an average value of said control quantity is increased, if the value of said cycle ratio setpoint is included in a given critical zone.