GB2344901A

GB2344901A - Control signal preparation for an engine fuel feed sytem

Info

Publication number: GB2344901A
Application number: GB9929319A
Authority: GB
Inventors: Andreas Kellner
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 1998-12-16
Filing date: 1999-12-10
Publication date: 2000-06-21
Anticipated expiration: 2019-12-10
Also published as: DE19857972B4; GB2344901B; DE19857972A1; JP2000179390A; GB9929319D0

Abstract

A method of preparing a control signal for a fuel feed system of an internal combustion engine, especially an engine with a common rail system in which at least one pump conveys fuel into a pressure storage device. The method detects fuel pressure in the pressure storage device and filters a signal PE indicative of the detected pressure using a filter, the behaviour of which varies with the detected pressure PE, to provide a filtered signal PA which is used in forming the control signal. The filter functions to reduce variations in the pressure signal, and preferably causes rapid changes in the filtered signal PA when signal PE rises, slow changes in the filtered signal PA when signal PE falls normally, and fast changes in the filtered signal PA when signal PE falls substantially. The filter may be analogue, as shown, or digital.

Description

2344901 CONTROL SIGNAL PREPARATION FOR AN ENGINE FUEL FEED SYSTEM The

present invention relates to a method of and means for preparing a control signal for a fuel feed system of an internal combustion engine.

In DE 195 48 278 there are disclosed a method and a device for the control of an internal combustion engine, in particular a method and a device for the regulation of the pressure in a pressure storage device of a common rail system (CR system). Usually, the duration of the drive control of the injectors in such a CR system is preset in dependence on the quantity of fuel to be injected and the pressure in the storage device. For this purpose, the pressure in the storage device is detected synchronously with the rotational speed. The pressure regulation takes place in a fixed time raster. For this purpose, the rail pressure detected synchronously with the rotational speed is scanned in time synchronisation.

In addition, it is known from DE 197 35 561 that the pressure values can be scanned at fixed time intervals. Accurate quantity values for the control of the injected quantity of fuel are achieved only when the pressure of the fuel during the injection is known. Any inaccuracy in the pressure measurement can lead to a quantity error and thereby to a poorer emission behaviour of the internal combustion engine.

There is thus a need for reduction in such quantity errors and thus improvement in engine emission behaviour.

According to a first aspect of the present invention there is provided a method for the control of an internal combustion engine, in particular an internal combustion engine with a common rail system, in which at least one pump conveys fuel into a pressure storage device, wherein a sensor signal is detected, which characterises the fuel pressure in the pressure storage device, characterised in that, starting from the sensor signal, filtering means presets a filtered sensor signal, wherein the behaviour of the filtering means is presettable in dependence on at least the sensor signal.

Preferably, the behaviour of the filtering means is presettable in dependence on the change in the sensor signal or in dependence on the sensor signal and the filtered sensor signal. The behaviour of the filtering means can be presettable differently for an increasing and for a decreasing sensor signal. For preference, a rapid change in the 2 filtered sensor signal is possible in the case of the increasing sensor signal and a slow change in the filtered sensor signal is possible in the case of the decreasing sensor signal. In addition, a rapid change in the filtered sensor signal can be possible in the case of greatly decreasing pressure. Moreover, the behaviour of the filtering means can be presettable in dependence on operating characteristic magnitudes, in particular in dependence on the rotational speed, of the engine.

According to a second aspect of the present invention there is provided a device for the control of an internal combustion engine, in particular an internal combustion engine with a common rail system, in which at least one pump conveys fuel into a pressure storage device, wherein a sensor detects a sensor signal which characterises the fuel pressure in the pressure storage device, characterised in that, starting from the sensor signal, a filtering means presets a filtered sensor signal, with means which preset the behaviour of the filtering means in dependence on at least the sensor signal.

Due to a filtered signal being formed by filtering means starting from the signal which characterises the pressure in the storage device and the behaviour of the filter means being dependent at least on the pressureindicative signal, a pressure value can be obtained which comes very close to the actual pressure at the beginning of fuel injection from the common rail system.

It is particularly advantageous if the filtering takes place in dependence on change in the pressure-indicative signal. Thereby, the filtered signal can react very rapidly to pressure rises and pressure collapses have an only insignificant effect. It is also particularly advantageous in that case if the filtered signal reacts rapidly to strong pressure collapses. The signal can thereby follow changes in target value very rapidly.

Examples of the method of embodiments of the signal preparation means of the present invention will now be more particularly described with reference to the accompanying drawings, in which:

Fig. 1 is a block circuit diagram of an engine fuel feed system incorporating control signal preparation means embodying the invention; Fig. 2 is a circuit diagram of analog filter means in the signal preparation means; 3 Fig. 3 is a circuit diagram of digital filter means in the signal preparation means; and Fig. 4a) to c) are diagrams showing different signals by way of illustration of operation of the signal preparation means.

Referring now to the drawings, there is shown in Fig. 1 those components of a fuel supply system of an internal combustion engine with highpressure injection as may be required for an understanding of an embodiment of the invention. The illustrated system is usually termed a common rail system and is illustrated only by way of example.

The system comprises a fuel supply container 100 connected with a preliminary conveying pump 110 by way of a duct. Fuel is fed by the pump 110 by way of a connecting duct to an admetering valve 120 controllable by way of a coil 136. The connecting duct between the pump 110 and the valve 120 is connected with the container '4100 by way of a lowpressure limiting valve 145. The valve 120 is connected by way of a further duct and a high-pressure pump 125 with a common rail 130, which is also termed a storage device. The rail 130 is connected with different injectors 131 by way of fuel lines and is connectible with the fuel supply container 110 by way of a pressure-limiting valve 135.

The part of the system between the exit of the high-pressure pump 125 and the entry of the press ure-1 imiting valve 135 represents a high-pressure region, in which fuel stands under high pressure. The pressure in the high-pressure region is detected by means of a sensor 140. The part of the system between the tank 100 and the high-pressure pump 125 represents a low-pressure region.

Associated with the fuel supply system is a control device 150, which acts on the injectors 131 by drive control signals A and also controls the coil 136 of the admetering valve 120. For this purpose, the output signal PE of the pressure sensor 140 and different output signals of further sensors 160, for example a rotational speed sensor, are evaluated in the control device.

The control device 150 comprises a signal filter 151, to which the output signal PE of the pressure sensor 140 is conducted. The filter 151 acts by signals PA, P] on a quantity 4 computation (QK) block 152 and a first input of an interlinking point 154. The output signal PS of a target value presetting device 153 is present at a second input of the interlinking point 154. The device 153 processes the output signal N of a rotational speed sensor 160 as well as that of the quantity computation block 152. The quantity computation block 152 acts on the 131 by the drive control signals A and the pressure computation, i.e. the filter 151, by a signal QK which indicates that an injection takes place. A pressure regulator 155 is acted on by the output signal of the interlinking point 154 and in turn controls the coil 136 of the admetering valve 120.

In operation, the fuel, which is situated in the supply container 100, is conveyed by the preliminary conveying pump 110 into the lower pressure region of the system. If the pressure in the low-pressure region rises to an impermissibly high value, the low-pressure limiting valve 145 opens and frees the connection between the output of the preliminary conveying pump 110 and the supply container 100.

The high-pressure pump 125 conveys the fuel from the low-pressure region into the highpressure region. The high-pressure pump 125 builds up a very high pressure in the rail 130. Usually, pressure values of about 30 to 200 bars are achieved in systems for applied ignition engines and pressure values of about 1000 to 2000 bars in compress ion-i g n ition engines. The fuel is admetered under high pressure to the individual cylinders of the engine by way of the injectors 131.

The pressure in the rail 130 or in the entire high-pressure region is detected by means of the sensor 140. The pressure in the high-pressure region can be regulated by means of the admetering valve 120, which is controlled by the coil 136. The valve 120 makes different conveyed quantities available to the high-pressure pump either in dependence on the voltage lying across the coil 136 or the current flowing through the coil 136.

Other setting members can be used for the regulation of the pressure P in the highpressure region. These are, alternatively to the admetering valve 120, an electrical preliminary conveying pump which is adjustable in respect of the conveyed quantity or a pressure-limiting valve 135, which is controlled by means of a coil.

The filter 151 prepares the sensor signal provided by the pressure sensor 140 and makes it available for pressure regulation on the one hand to the quantity computation block 152 and on the other hand to the comparison point 154. The block 152 computes the drive control signals A for action on the injectors 131 in dependence on the pressure P and the desired quantity of fuel to be injected.

The target value presetting device 153 computes the target value PS for the fuel pressure in the rail 130 starting from different operating parameters, such as the rotational speed N of the engine and the quantity of fuel to be injected. This target value PS is compared in the interlinking point 154 with the actual value P], provided by the filter 151. The pressure regulator 155 computes the drive control signal for action on the valve 120 in dependence on the comparison result.

The computation of the drive control signals in dependence on the pressure P takes place before each injection and at variable time intervals in dependence on the rotational speed. The intervals between these computations depend to a large extent on the rotational speed. The computation of the drive control signal for the valve 120 by the pressure regulator 155 takes place at a fixed time rate. This time rate is chosen so that the regulator can react immediately to changing target values PS and the new target value can be set as rapidly as possible.

The pressure in the rail 130 fluctuates greatly due to, on the one hand, the injection and, on the other hand, the pressure generation. The pressure values needed for quantity computation andlor for pressure regulation are scanned at certain fixed time intervals. This leads to substantial fluctuation in the pressure values used for the quantity computation. Large fluctuations in the pressure values have the consequence of large fluctuations in the quantity values. Moreover, the pressure value ascertained during the pressure detection differs greatly from the pressure during the injection. This leads to inaccurate injection.

To improve accuracy, the sensor signal is filtered and in particular in such a manner that the filtering takes place in different ways in dependence on the sensor signal. Preferably, the filtering takes place in dependence on change in the sensor signal. Change in the sensor signal is recognised starting from the difference between the input voltage and the output voltage of the filter. This means that the behaviour of the filter means is presettable in dependence on change in the sensor signal or in dependence on the sensor signal and the filtered sensor signal.

6 A first embodiment of the filter 151 is illustrated in Fig. 2. The pressure signal PE as an input voltage of the filter is applied by way of a first resistor R1 to the positive input of an operational amplifier OV1. The positive input of the amplifier OV1 is also connected with ground by way of a first capacitor Cl. The output of the amplifier OV1 is connected by way of a first diode D1 with the positive input of a second operational amplifier OV2, at the output of which the output voltage PA is present.

The first diode D1 has its anode connected with the output of the first amplifier OV1 and its cathode with the positive input of the second amplifier OV2. The cathode of the diode D1 is also connected directly with the negative input of the first amplifier OV1. In addition, the cathode of the diode D1 is connected with the positive input of the first amplifier OV1 by way of a third resistor R3 and a second diode D2. In that case, the cathode of the second diode D2 is connected with the positive input of the first amplifier OV1 and the anode of the diode D2 with the resistor R3.

The cathode of the diode D1 or the positive input of the amplifier OV2 is connected with ground by way of a second capacitor C2 and by way of a resistor R2. The output of the amplifier OV2 is connected directly back with the negative input of the amplifier OV2.

The mode of operation of the analog filter illustrated in Figure 2 can be divided into two phases. The resistance-capacitance filter consisting of the first resistor R1 and the first capacitor Cl filters out highfrequency disturbances of the sensor signal, which preferably have a time constant of the order of magnitude of 0.5 to 1 milliseconds.

In a first phase, in which the output voltage PA is lower than the input voltage PE, the first diode D1 is in its conductive state and the second diode D2 in its blocked state. The capacitor C2 is charged to the voltage of the capacitor Cl, i.e. the voltage across the second capacitor C2 corresponds with the filtered input voltage PE. The amplifier OV1 compensates for the voltage drop across the first diode D1 and the discharge current by way of the second resistor R2. The second amplifier OV2 is connected as a voltage follower and provides the output voltage PA, which corresponds with the voltage across the capacitor C2, to be usable without the circuit part upstream of the amplifier OV2 being influenced. The amplifier OV2 can be dispensed with in a simplified embodiment.

7 Thus, when the input voltage PE is greater than the output voltage PA, the output voltage follows the input voltage very rapidly. Only highfrequency disturbances are filtered out by the resistance-capacitance member consisting of the first resistor Rl and the first capacitor Cl, which is the case when the pressure signal rises. This means that a rapid change in the filtered signal is possible in the case of an increasing sensor signal. The behaviour of the filter 151 is settable in different manner for an increasing and for a decreasing sensor signal.

In a second phase, the output voltage PA is greater than the input voltage PE. This has the consequence that the first diode D1 is in its blocked state. Then, only two cases are to be distinguished. In a first case, the difference between the output voltage PA and the input voltage PE is smaller than the breakdown voltage of the diode D2. This means that the second diode D2 is in its blocked state. This has the consequence that the capacitor C2 is discharged by way of the resistor R2. The discharge time constant R2C2 is advantageously chosen so that short pressure collapses, which are caused by injection, do not become noticeable in the output voltage. However, it takes place so rapidly that different pressures of the individual injections come into effect. The discharge time constant is preferably chosen to be in a range between 300 and 800 milliseconds.

This means that when the input voltage PE is lower than the output voltage PA, the output voltage follows the input voltage slowly. High- frequency disturbances are filtered out by the resistance-capacitance member consisting of the first resistor Rl and the first capacitor C2, which is the case when the pressure signal decays. This means that a slow change in the filtered signal is possible in the case of a reducing sensor signal.

In a second case, in which the difference between the output voltage PA and the input voltage PE is greater than the breakdown voltage of the diode D2, the diode D2 conducts. The voltage of the capacitor C2 is now discharged additionally by way of the resistor R3 to the input voltage plus the breakdown voltage of the diode D2. The discharge time constant R3C2 is preferably chosen to be very short, in particular between 2 and 10 milliseconds. The diode D2 is selected so that its breakdown voltage is somewhat greater than the maximum occurring collapse of the input voltage PE in consequence of an injection.

8 This means that when the input voltage PE is very much smaller than the output voltage PA, the output voltage follows the input voltage very rapidly. High-frequency disturbances are filtered out by the resistancecapacitance member consisting of the first resistor Rl and the first capacitor Cl, which is the case when the pressure signal reduces very substantially. This means that a rapid change in the filtered sensor signal is provided for a greatly reducing sensor signal.

This has the advantage that the output voltage PA in the case of a rapid pressure decay, which is caused for example by a change in the target value for the pressure, is likewise reduced rapidly. Changes, which are due merely to the injection, have no effect on the measurement signal PA.

The filter is thus structured in such a manner that the signal PA reacts very rapidly to a rising pressure and a greatly failing pressure, such as usually arises in the case of a change in the target value. The output signal PA of the filter reacts only very slightly to small pressure changes such as caused by the injection.

It is particularly advantageous if the filter is realised in digital form. In that case, as shown in Fig. 3, the signal PE filtered by the resistance-capacitance member Rl, Cl is detected in a step 300. A subsequent interrogation step 310 checks whether the instantaneous input signal PE is greater than the output signal PA. If this is the case, a filtering procedure F2 is activated in a step 320. If the signal PE is not greater than the signal PA, it is checked in an interrogation step 330 whether the difference between the input signal PE and the output signal PA is greater than or equal to a threshold value SW. If this is the case, the step 320 is carried out. If this is not the case, a filtering procedure Fl is selected in a step 340.

If the pressure of the input signal PE is greater than the output signal PA, this occurs particularly when the pressure rises or when it drops very rapidly. A rapid pressure drop happens, in particular when the target value for the pressure changes to a lower value. In this case, the filtering procedure F2 is selected in the step 320 and the input signal PE is filtered accordingly in a step 350 in order to form the output signal PA. This filtering is carried out in such a manner that it reacts very rapidly to pressure changes.

9 The properties of the filtering procedures F1 and F2 can, in a particularly advantageous example, be preset in dependence on operational states of the engine. It is particularly advantageous of these properties of the filtering procedures F1 and/or F2 are preset in dependence on the rotational speed of the engine and/or a magnitude dependent on the rotational speed. Thereby, the properties can be adapted in very simple manner to the operational state of the engine. It is particularly advantageous if the filter time constants are proportional to the reciprocal of the rotational speed. The higher the rotational speed, the smaller the interval between the injections and the smaller the filter time constant. The filtering procedure is so carried out that it has the same effect for any rotational speed. This means that a rapid changeover time is provided for high rotational speeds and a slow changeover time is provided for low rotational speeds.

When the input signal is smaller than the output signal, i.e. the pressure drops, the interrogation step 330 recognises that the difference between the input signal and the output signal is smaller than a threshold value. In this case, the filtering procedure F1 is chosen in the step 340 and the input signal PE is filtered by the filtering procedure F1 in a step 360. The filtering procedure F1 is carried out in such a manner that the output signal reacts only slowly to changes in the input signal PE.

In Fig. 4a) to c), different signals are entered as a function of time t. The time intervals in which injection takes place are entered in Fig. 4a). The input signal PE of the filter is indicated by a solid line in Fig. 4b). The signal processed by the regulator or by the quantity computation block 152 is shown by a dashed line. The regulator 155 or the quantity computation block 152 processes a signal which is scanned at fixed time intervals. For this purpose, the value, which is present at a certain instant, is retained for the scanning time. The course of this signal is marked by a dashed line. In Fig. 4c), the filtered output signal PA is represented by a solid line and the corresponding scanned signal by a dashed line. A certain value X is marked by the zero line in Figs. 4b) and 4c).

The input signal PE fluctuates very substantially. This is caused on the one hand, by the injection, which produces a very large drop in the pressure. Moreover, the fuel conveying by the high-pressure pump 125 causes a brief rise in the fuel pressure each time. Due to the scanning of the value at preset time instants, a very strong fluctuation of the scanning value in Fig. 4b) results. This strong fluctuation leads to a strong fluctuation of the injected quantity of fuel. Fig. 4c) shows the filtered signal PA, which follows the rise more or less directly and the strong decay becomes effective only with delay. The fluctuation of the filtered signal is thus significantly less than in the case of the unfiltered signal.

11 1. A method of preparing a control signal for a fuel feed system of an internal combustion engine, comprising the steps of detecting the pressure of fuel in fuel storage means which is supplied with fuel by fuel conveying means of the system and delivers fuel for combustion in the engine, filtering a signal indicative of the detected pressure by filtering means, causing the behaviour of the filtering means to vary in dependence on the pressure-indicative signal and providing the control signal in dependence on the filtered signal.

2. A method as claimed in claim 1, wherein the behaviour of the filtering means is varied in dependence on change in the value of the pressureindicative signal.

3. A method as claimed in claim 1, wherein the behaviour of the filtering means is additionally varied in dependence on the filtered signal.

4. A method as claimed in any one of the preceding claims, wherein the behaviour of the filtering means is varied in different manner respectively for increase in the value of the pressure-indicative signal and decrease in the value of the pressure-indicative signal.

5. A method as claimed in any one of the preceding claims, wherein the behaviour of the filtering means is varied to cause a rapid change in the filter signal in response to increase in the value of the pressureindicative signal.

6. A method as claimed in any one of the preceding claims, wherein the behaviour of the filtering means is varied to cause a slow change in the filter signal in response to decrease in the value of the press ure-i ndicative signal.

7. A method as claimed in any one of the preceding claims, wherein the behaviour of the filtering means is varied to cause a rapid change in the fitter signal in response to substantial decrease in the value of the pressure-indicative signal.

8. A method as claimed in any one of the preceding claims, wherein the behaviour of the filtering means is varied in dependence on at least one operating parameter of the engine.

12 9. A method as claimed in claim 8, wherein the at least one parameter is engine rotational speed.

10. A method as claimed in claim 1 and substantially as hereinbefore described with reference to Figs. 1, 2 and 4 of the accompanying drawings.

11. A method as claimed in claim 1, 3 and 4 and substantially as hereinbefore described with reference to Figs. 1, 3 and 4 of the accompanying drawings.

12. Signal preparation means for preparing a control signal for a fuel feed system of an intemal combustion engine, comprising a pressure sensor for detecting the pressure of fuel in fuel storage means which is supplied with fuel by fuel conveying means of the system and delivers fuel for combustion in the engine, filtering means for filtering a signal indicative of the detected pressure, means for causing the behaviour of the filtering means to vary in dependence on the press u re-i rid icative signal and means for providing the control signal in dependence on the filtered signal.

13. Signal preparation means substantially as hereinbefore described with reference to Figs. 1, 2 and 4 of the accompanying drawings.

14. Signal preparation means substantially as hereinbefore described with reference to Figs. 1, 3 and 4 of the accompanying drawings.

15. A motor vehicle fuel supply installation including signal preparation means as claimed in any one of claims 12 to 14.

16. A system as claimed in claim 15, wherein the installation comprises a fuel injection system and the fuel storage means is a common rail of the injection system.