GB2313679A

GB2313679A - Determination of a setting magnitude for an engine operating parameter

Info

Publication number: GB2313679A
Application number: GB9710779A
Authority: GB
Inventors: Dietbert Schoenfelder; Kai-Lars Barbehoen
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 1996-05-29
Filing date: 1997-05-23
Publication date: 1997-12-03
Anticipated expiration: 2017-05-23
Also published as: GB2313679B; JPH1054290A; GB9710779D0

Description

DETERMINATION OF A SETTING MAGNITUDE FOR AN ENGINE OPERATING PARAMETER The

present invention relates to a method of and determining means for determining a setting magnitude for setting means influencing an operating parameter of an internal combustion engine.

In DE-OS 34 00 711 (US-A 4 638 782) there is disclosed a method and a device for control and/or regulation of a setting member, in particular for the influencing of the metering of fuel in an internal combustion engine. The deviation of an actual value from a target value is ascertained and a regulator presets, starting from the deviation, a setting magnitude for action on the setting member. For high rotational speeds, regulation is carried out by means of a regulator preferably with proportional-integral behaviour. In other operational states, in particular at lower rotational speeds, the setting member is merely controlled.

In order to be able to cope with the high demands, in particular on the dynamic range, in such regulators, complex regulating loops are required. One possibility of meeting these demands is by way of regulators of variable structure. In this case, switching-over is carried out between different regulator structures under certain presuppositions. Thus, for example, switching over can be between a proportional-integral regulator and a proportional regulator or a pure control device.

The switching-over operations entail problems. On the switchingover from one regulator type to another, undesired transient states frequently arise. For example, when switching over from a proportional regulator with a high amplification factor, which can function as a large signal regulator, to a finely applied proportional-integral regulator, which can function as a small signal regulator, the integral component must be preset in suitable manner to avoid an undesired behaviour of the regulator, in particular oscillations. 5 There thus remains a need to achieve a good regulating behaviour, in all operational states, in a method and determining means for determining a setting magnitude for use in regulation of an internal combustion engine. According to a first aspect of the present invention there is provided a method of regulating an internal combustion engine, in particular for influencing the instant of fuel injection in an internal combustion engine, comprising the steps of determining a deviation starting from a target value and an actual value and, starting from the deviation, presetting a setting magnitude, by way of a regulator, for action on a setting member, wherein a first regulator structure is active in the presence of first operational states and a second regulator structure is active in the presence of second operational states, characterised in that a starting value for initialisation of the second regulator structure is presettable in dependence on at least the target value and the actual value on transition from the first regulator structure to the second regulator structure. Such a method may have the advantage that a good regulating behaviour results in all operational states and no oscillations occur on transition between the individual regulator structures. 25 Preferably, the starting value for the initialisation of the second regulator structure is presettable in dependence on at least the up-todate target value and the up-to-date actual value on transition from the first to the second regulator structure. For preference, the starting value is presettable starting from the target value weighted by a first weighting factor and the actual value weighted by a second weighting factor. The weighting factors are preferably presettable starting from the deviation and/or a maximum regulation range. For preference, the first regulator structure is formed as a large signal regulator and the second regulator structure as a small signal regulator. Expediently, the second regulator structure includes a regulator with at least an integral portion.

Preferably, on transition from the first regulator structure to the second regulator structure, the starting value for the initialisation of the second regulator structure is presettable in dependence on at least the up-to-date target value and the deviation of the integral portion from the target value on transition from the second regulator structure to the first regulator structure.

The starting value is, for preference, preset according to the formula IA = WF S + (1 - WF) I, wherein WF is the first weighting factor, ( 1 - WF) i s the second weighting factor, S is the target value and I is the actual value.

Preferably, the first operational states are present when the amount of the deviation is greater than the maximum regulation range and the second operational states are present when the amount of the deviation is less than the maximum regulation range.

According to a second aspect of the invention there is provided a device for the regulation of an internal combustion engine, in particular for influencing the instant of the injection of fuel into an internal combustion engine, wherein a regulating deviation is determi nab 1 e starting from a target value and an actual value and a regulator starting from the deviation presets a setting magnitude for action on a setting member, with switching means for activating a first regulator structure in the presence of first operational states and a second regulator structure in the presence of second operational states, characterised in that means are provided for presetting a starting value for the second regulator structure in dependence on at least the target value and the actual value on transition from the first regulator structure to the second regulator structure.

Examples of the method and embodiments of the determining means will now be more particularly described with reference to the accompanying drawings, in which:

is Fig. 1 is a schematic block diagram of determining means embodying the invention; Fig. 2 is a diagram of signals issued in the course of functioning of the determining means; Fig. 3 is a flow diagram illustrating steps in performance of a first method exemplifying the invention; and Fig. 4 is a flow diagram illustrating steps in performance of a second method exemplifying the invention.

Referring now to the drawings there is shown in Fig. 1 a device for determining a setting magnitude for use in regulation of an injection adjuster of a compression-ignition internal combustion engine. The device is not, however, restricted to regulation for this purpose, but can be used for the regulation of other magnitudes in, in particular, the operation of internal combustion engines. In the case of the device shown in Fig. 1, an engine (not shown) receives fuel from a fuel pump. The start of fuel conveying or of fuel injection can be controlled by means of an injection adjuster 100, 5 which is referred to in the following as a setting member 100. The setting member 100. is acted on by a signal from a setting magnitude presetter 110. The setting member 100 assumes a certain position in dependence on a drive control signal provided by the presetter 110. This, in turn, has the consequence that a certain value 10 results for the start of injection or of conveying. The setting magnitude presetter 110 is acted on by the output signal ST of a switching means 120. The switching means 120 selects either the output signal of a junction 125 or the output signal of a large signal regulator 140 and passes this on to the presetter 110. The junction 125 15 interlinks the output signal of a proportional component regulator part 130 and an integral component regulator part 135. The output signal of a switching means 150 is conducted to the two regulator parts 130 and 135, which together form a proportional-integral regulator (also termed a small signal regulator). 20 The switching means 150 passes the output signal of a zero value presetter 155 or the output signal of a junction 160 to the proportionalintegral regulator. The output signal of the junction 160 is also passed to a large signal regulator 140. The output signal 5 of a target value presetter 170 is present with 25 positive sign at the junction 160 and a signal I is present with negative sign at the junction 160. The signal I is an actual setting magnitude and the signal 5 is a target value of the regulating loop.

In addition, a control 180 is provided, which processes the signals S and 1 and optionally also further input magnitudes. The control 180 acts on the switching means 150 and 120 by drive control signals and acts on the regulator part 135 by a further signal.

Usually, the switching means are disposed in the positions shown in solid lines. In that case, the device operates as a small signal regulator. The target value S is compared with the actual value I at the junction 160 and the thus formed deviation is fed to the regulator 130, 135. Starting from this deviation, the regulator parts 130 and 135 each form a respective one of two components for the setting magnitude, the components being combined at the junction 125. The setting magnitude ST thus formed is passed by way of the switching means 120 to the presetter 110. This translates the output signal of the regulator into a drive control signal for action on the setting member 100. Preferably, the setting member 100 is acted on by a keying ratio dependent on the deviation. This has the consequence that the setting member assumes a certain position and the injection thereby begins at a certain instant.

The position of the setting member 100 or a signal indicative of the start of conveying and/or of injection can be used as the actual signal 1. Accordingly, the target value presetter 170 must preset a corresponding target value S of equal dimension. The target value presetter 170 presets the target value S in dependence on different operational characteristic magnitudes.

If large deviations arise, which is the case when, in particular, the target value S changes significantly, the setting member needs a certain time until it has reached the new target value. In that case, the large signal regulator 140 comes into action. The control 180 actuates the switching means 150 and 120 in such a manner in dependence on the deviation that they pass over into the positions shown in dashed lines. This means that the integral regulator part 135 and the proportional regulator part 130 are acted on by the zero value presetter 155 by such a signal that they maintain their states. The deviation is fed to the large signal regulator 140. Its output signal is then passed to the presetter 110.

When the deviation is greater than a certain value, a maximum value for the setting magnitude is preset and when the deviation is smaller than a negative threshold value, a negative maximum value is preset. The large signal regulator 140 acts substantially as two-point regulator. In the range between the minimum value and the maximum value, the regulator parts 130 and 135 are effective.

The transition from the large signal regulator 140 to the small signal regulator 130, 135 entails problems, thus in the transition from the dashed line position to the solid line position of the switching means 150 and 120. During this transition, the integral component must be preset at a suitable starting value.

Different signals, namely the target value 5, the actual value 1, the setting magnitude ST and the proportional component IAN of the regulator part 135 are entered as a function of time t in Fig. 2. The target value 5 is shown by a chain-dotted line, the actual value I by a dashed line, the setting magnitude ST, i.e. the input signal of the presetter 110, by a solid line and the integral component IAN by a dotted line.

At the instant tl, all four signals have constant values. The switching means 120 and 150 assume their positions shown by solid lines.

At the instant tl, the target value drops to a substantially lower value. Since the actual value remains at its old value, this has the consequence that the deviation is substantial. The control 180 recognises this and controls the drive of the switching means 150 and 120 in such a manner that these adopt the positions indicated by dashed lines. This, in turn, has the effect that the large signal reflector 140 becomes active. This means that the setting signal rises to a very high value.

The integral component remains at its old value, since the regulator 130, 135 no longer has any effect. Between the instants tl and t2, the actual value I drops substantially due to the large setting magnitude ST. At the instant t2, the actual value reaches such a level that the deviation is smaller than the value at which the large signal regulator 140 is active. This means that the regulator 130, 135 is activated again at the instant t2 by the switching means 120 and 150 being brought into their basic positions.

The integral component must be initialised by a suitable value at this instant. This means that the integral component passes in one step to its new starting value at the instant t2.

After the instant t2, the actual value drops further and approaches the target value S. The setting magnitude ST drops rapidly over a short time and then approaches its new end value. The same applies to the integral component. After the instant t3, all four values have reached their new end values.

The steps of a method exemplifying the invention are illustrated by the flow diagram of Fig. 3. The control detects the target value S in a first s tep 300 and the actual value I in a step 310. In a step 320, the deviation R is computed starting from the target value S and the 9 actual value 1.

A subsequent interrogation step 330 checks whether the amplitude of the deviation JR1 is greater than a threshold value SW. The threshold value SW corresponds with the maximum possible regulating range. If this is the case, the drive control signal is passed to the switching means 120 and 150 in a step 340 and has the effect that these assume their positions shown in dashed lines. In this case, the large signal regulator 140 is active for a deviation which is large in terms of amplitude. Subsequently, a marker M is set to 1 in a step 350.

Subsequent to the step 350, a renewed program sequence takes place starting with the step 300.

If the interrogation step 330 shows that the deviation is not creater than the threshold value SW, then a step 360 follows, which checks whether the marker M is set at 1. If this is not the case, which means that the regulator 130, 135 was active during the preceding program sequence, a new program sequence takes place immediately beginning with the step 300. If the marker M was set at 1, which means that the large signal regulator 140 was active during the preceding program sequence, then a starting value IA for the integral component 135 is preset in a step 370 as function F from the target value S, the actual value I and a magnitude WF. Subsequently, in a step 380, a drive control signal is passed to the switching means 120 and 150, which has the effect that these assume their positions shown in solid lines. In this case, the small signal regulator 25 130, 135 is active for a deviation which is small in terms of amplitude. The starting value for the integral component IA is determined by the following formula:

IA = WF S + (1 - WF) I, in which WF is a first-weighting factor and (1 WF) is a second weighting factor. In the simplest case, a constant value, which is determined within the scope of the application and used constantly in operation, is used as the weighting factor WF. In a more involved variant, the weighting factor is obtained from a characteristic values field in dependence on different operational characteristic magnitudes. The values S and 1 are an up-to-date target value and an up-to-date actual value, respectively. The computation of the starting value IA for the integral component IAN takes place starting from these up-to-date values.

In a particularly advantageous version, the weighting factor WF is preset in dependence on the target value S, the actual value I and the threshold value SW. The threshold value SW corresponds with the deviation at which switching-over from the small signal regulation to large signal regulation is carried out.

This means introducing a weighted deviation WF, which describes the degree of the deviation referred to the maximum range of action of the small signal regulator. The integral component is preset at a value dependent on the target value and the actual value in dependence on this weighting. Thus, an optimum starting value of the integrator of the integral component results in dependence on the up-to-date operational conditions. It is ensured by the choice of the weighting factor that an entry into the new regulator in the case of up-to-date maximum deviation leads to a presetting of the integral component with maximum permissible setting component, i.e. greatest dynamic range, of the entire system. On the other hand, it is ensured that entry into the new regulator in the case of negligible deviation leads to a gentle activation of the setting magnitude drive control. All intermediate states are adapted continuously, which results in a largely step-free manner of procedure.

A further method exemplifying the invention is illustrated in the flow diagram in Fig. 4. The control detects the target value S in a first step 400 and the actual value I in a step 410. In a step 420, it computes the deviation R starting from the target value S and the actual value I.

The subsequent interrogation 430 checks whether the amplitude of the deviation JR1 is greater than a threshold value SW. If this is the case, a drive signal is passed to the switching means 120 and 150 in a step 440 and has the effect that these assume their positions shown in dashed lines. In this case, the large signal regulator 140 is active for a deviation large in terms of amplitude. In a step 450, a marker M is set to 1. Subsequently, the difference between the integral component and the target value S is ascertained in a step 455 and stored as value RA. Subsequent to the step 455, a new program sequence takes place starting with a step 400.

If the interrogation step 430 has the result that the deviation is not greater than the threshold value SW, then a step 460 is carried out.

In this step, it is checked whether the marker M is set at 1. If this is not the case, which means that the regulators 130 and 135 were active during the preceding program sequence, a new program sequence follows immediately beginning with the step 400. If the marker M was set at 1, which means that the large signal regulator 140 was active during the preceding program sequence, a starting value IA for the integral component 135 is preset in a step 470 as function F of the up-to-date target value S, the deviation RA of the integral component IAN from the target value S, which was stored on the transition from the small signal regulator to the large signal regulator. 5 Subsequently, in a step 480, a drive control signal is passed to the switching means 120 and 150 and has the effect that these assume their positions shown in solid lines. In this case, the small signal regulator 130, 135 is active for a deviation which is small in terms of amplitude. The starting value for the integral component IA is determined, in the simplest case, by the following formula:

IA = S + RA.

The difference between the integral component IAN and the target value S is ascertained in the transition from small signal behaviour into large signal behaviour and this value RA is stored. On the transition from large signal behaviour to small signal behaviour, the new starting value for the integral component IAN is ascertained starting from this stored value RA and the up-to-date target value S. This means that the starting value for the initialisation of the small signal regulator is presettable after the change from the large signal regulator to the small signal regulator in dependence on at least the up-to-date target value and the stored deviation of the integral component from the target value during the transition from the small signal regulator to the large signal lator.

regul It is particularly advantageous when the starting value is 25 ascertained by the formula IA = J S + (1 - WF) I + RA.

This means that the stored value RA is added to the starting value which was ascertained according to the procedure illustrated in Fig. 3.

Claims

- 14 CLAIMS

1 A method of determining a setting magnitude for setting means influencing an operating parameter of an internal combustion engine, the method comprising the steps of determining the difference between a target value and an actual value related to the parameter, determining the setting magnitude in dependence on the determined difference by way of either one of two regulators each active in the presence of a respective operational state, and determining a starting value for initialisation of one of the regulators in dependence on the target value and the actual value on transition to that regulator from the other regulator.

2. A method as claimed in claim 1, wherein the starting value is determined in dependence on target and actual values which are up-todate.

3. A method as claimed in claim 1 or claim 2, wherein the starting value is determined in dependence on target and actual values which are each weighted by a respective weighting factor.

4. A method as claimed in claim 3, wherein at least one of the weighting factors is determined in dependence on at least one of said determined difference and a given maximum regulation range.

5. A method as claimed in any one of the preceding claims, wherein said one regulator is a small signal regulator and said other regulator is a large signal regulator.

6. A method as claimed in any one of the preceding claims, wherein said one regulator has at least an integral component.

7. A method as claimed in claim 6, wherein the starting value is determined in dependence on the up-to-date target value and the difference between the integral component and the target value on transition to said other regulator from said one regulator.

8. A method as claimed in claim 3 or claim 4, wherein the starting value is determined by the formula WF S + (W - WF) 1 wherein S is the target value, I is the actual value, WF is a first weighting factor and 1 - WF is a second weighting factor.

9. A method as claimed in any one of the preceding claims, wherein said one regulator is active when said determined difference is below a given maximum regulation range and said other regulator is active when that difference is above the given maximum regulation range.

10. A method as claimed in any one of the preceding claims, wherein the parameter is the start of fuel injection by a fuel injection installation of the engine.

11. A method as claimed in claim 1 and substantially as hereinbefore described with reference to Figs. 1 to 3 or Figs. 1, 2 and 4 of the accompanying drawings.

12. Determining means for determining a setting magnitude for setting means influencing an operating parameter of an internal combustion engine, the determining means comprising means for determining the difference between a target value and an actual value related to the parameter, two regulators each operable to determine the setting magnitude in dependence on the determined difference, switching means for switching one regulator to be active in the presence of one operational state and the other regulator to be active in the presence of another operational state and means for determining a starting value for initialisation of one of the regulators in dependence on the target value and the actual value on transition to that regulator from the other regulator.

13. Determining means as claimed in claim 12, wherein the parameter is the start of fuel injection by a fuel injection installation of the engine.

14. Determining means substantially as hereinbefore described with 20 reference to Figs. 1 to 3 or Figs. 1, 2 and 4 of the accompanying drawings.