WO2019207542A1

WO2019207542A1 - Speed control method for an internal combustion engine

Info

Publication number: WO2019207542A1
Application number: PCT/IB2019/053449
Authority: WO
Inventors: Marcello CARO; Bruno Aimar; Aldo BRUCATO
Original assignee: Fpt Industrial S.P.A.
Priority date: 2018-04-27
Filing date: 2019-04-26
Publication date: 2019-10-31
Also published as: ES2951992T3; IT201800004932A1; EP3784895B1; CN112020603A; BR112020021988A2; CN112020603B; EP3784895A1

Abstract

A control method for an internal combustion engine, carried out by means of a feedback control of the engine speed, based on an error (Err) of said engine speed, calculated between a reference value (Ref) and a measured value (Speed) of the speed, wherein said control comprises a sum (S2) of a proportional contribution (P) and an integral contribution (I), the method comprising a step of adding a derivative contribution (D) only when a first sign of said error and a second sign of a derivative of the engine speed are reciprocally concordant.

Description

"SPEED CONTROL METHOD FOR AN INTERNAL COMBUSTION ENGINE"

Cross-reference to related applications

This patent application claims priority from Italian patent application no. 102018000004932 filed on 27/04/2018, the entire disclosure of which is incorporated herein by reference

Technical field of the invention

The present invention refers to the field of control methods for controlling the operating point of internal combustion engines .

State of the art

It is known that the accelerator pedal allows the driver to cause a variation in the rotation speed of the internal combustion engine.

This is evidently the consequence of a variation in the torque delivered by the internal combustion engine.

The accelerator pedal can therefore relate to the motor point, namely engine speed/torque, of operation of the internal combustion engine representing, with a relative inclination of said pedal, a required torque value or a target speed of the engine .

Engine speed and torque are evidently interrelated parameters; nevertheless, speed control entails very rapid responses by the internal combustion engine.

Speed control on the internal combustion engine is particularly delicate due to the intrinsic delay in response of the engine due to various factors, including the capacitive effect of the intake manifold, which presents a slow response. During a gear change or a manoeuvre, or the sudden insertion of a load, such as a pump or a vehicle compressor, or a sudden variation in the road gradient, maximum rapidity and precision of response is required of the speed control.

A speed control method is based on the fact of calculating a speed error relative to a reference value and consequently calculating the control variable necessary (torque) of the system controlled, in order to vary the response of the internal combustion engine, to reduce or annul said error. It is obviously a feedback control.

The typical controller used in the known art to control the engine speed is of PI (proportional, integral) type. In other words, the error is corrected, between a reference value and the measured or estimated value of the system output, via a proportional controller and an integral controller which, on the basis of said error, generate respective control signals. The sum of the respective outputs, i.e. the respective generated control signals, is applied to the system input to cause the system output to converge with the reference value, which in this case is a speed reference of the engine.

It would be extremely advantageous to use also a derivative contribution, providing a full PID, since the derivative contribution would allow a prompter reaction, as the derivative term is an anticipative control component. Unfortunately, the derivative term typically introduces high frequency dynamics into the system, making the control unstable .

In a full PID controller, the outputs of the controllers P, I and D are added together and the result is applied to the input of the system to be controlled. Due to the above-mentioned instability, the derivative contribution is generally not used.

Summary of the invention

The object of the present invention is to provide a control method of an internal combustion engine that allows the implementation of a speed control which is rapid but, at the same time, stable and fluid.

The basic idea of the present invention is to provide a feedback speed control by means of a PID controller, comprising a proportional, integral and derivative contribution only when an engine speed error is caused by a variation in the speed reference value, otherwise the controller is PI, namely it comprises only proportional and integral contributions.

In other words, when the above-mentioned operating condition occurs, a full PID controller is used, in which the sum of the respective outputs, namely of the respective generated control signals, is applied to the input of the internal combustion engine to cause the engine output to converge with the speed reference value.

This means that when the speed error is caused by an external disturbance, for example a rapid variation in the road gradient, or the activation of a device connected to a power take-off of the engine, etc... the derivative contribution is not added to the proportional and integral contributions.

To discriminate the case in which the error depends on a variation of the reference signal or on a load variation, it is sufficient to compare a sign of said engine rotation speed error with a sign of a derivative of the engine rotation speed. In particular, if the signs are discordant, then the speed error is caused by a load variation. Below, by "speed" we mean "engine rotation speed".

When the speed derivative is detected as positive and the error is negative, it means that the engine is accelerating beyond the reference value. This generally occurs when a load is suddenly detached.

Vice versa, when the speed derivative is negative and the speed error is positive, it means that the torque delivered by the engine is not sufficient to cope with the load applied, suddenly, for example.

Therefore, according to the present invention, the derivative contribution is added to the proportional and integral control outputs when the error and derivative signs are concordant.

Vice versa, when the signs are discordant, the derivative contribution is not added to the integral and proportional contributions .

In relation to the specific implementations of the invention, use of the derivative contribution can be decided when the above-mentioned signs are concordant and positive, or concordant and negative.

The fact of using the derivative contribution according to the present invention, makes the engine extremely prompt, but at the same time the introduction of high frequency dynamics into the control system, that can make the control unstable, is avoided .

By "controller" we mean the revolution control carried out on the internal combustion engine overall. According to a preferred variation of the invention, not only the signs must be concordant, but the absolute value of the speed error must exceed a predetermined first threshold and the absolute value of the speed derivative must exceed a second predetermined threshold.

According to a preferred variation of the invention, which combines with the preceding ones, when only the proportional and integral contributions are used, the derivative of the engine rotation speed (which is normally multiplied by a predetermined parameter that defines the above-mentioned derivative contribution) , is used to modify an operating parameter of the integral controller so as to compensate for sudden variations due to loads.

In particular, the integral controller comprises a saturator, which saturates the control signal towards the internal combustion engine, based on a speed error and on the speed derivative of said internal combustion engine.

The derivative contribution responds much more rapidly than the other proportional and integral contributions, but its contribution is used to intervene on the integral contribution, minimizing the overshoots and undershoots of the internal combustion engine response, without affecting its stability .

From the above it is clear that a signal which is added to the proportional and integral contributions to minimize the over elongations is not generated; rather, the derivative contribution is used to cope with sudden load variations.

Preferably, the signal obtained from the sum of the integral proportional and derivative contributions are not saturated. The claims describe preferred variations of the invention forming an integral part of the present description.

Brief description of the figures

Further objects and advantages of the present invention will become clear from the following detailed description of an embodiment example of the same (and of variations thereof) and from the attached drawings provided for purely illustrative non-limiting purposes, in which: figure 1 shows a control diagram based on an implementation example of the present invention;

figures 2a and 2b show two equivalent control diagrams used in a mutually exclusive manner according to another implementation example of the present invention;

figure 3 shows in detail a block of the control diagram of figure 2b.

The same numbers and the same reference letters in the figures identify the same elements or components.

In the context of the present description the term "second" component does not imply the presence of a "first" component. Said terms are used only for clarity and should not be understood in a limiting manner.

Detailed disclosure of embodiment examples

Figure 1 shows an example of a control example with a block diagram.

The control is carried out recursively in discrete time, according to the operating frequency of the ECU processing unit which controls the internal combustion engine.

The internal combustion engine is shown in figure 1 with the engine block. A revolution sensor, for example a phonic wheel applied to the drive shaft, provides a measurement of the internal combustion engine speed.

Furthermore, the accelerator pedal or any vehicular device suited to imparting a reference speed, for example a robotized transmission or a power take-off, generates the reference signal Ref in terms of target engine speed.

The speed measured, Speed, is subtracted from the target speed, Ref, generating a (speed) error Err by means of the first adder node Si on the left of figure 1.

The same (speed) error value is conveyed to the input of blocks P and I, while a speed signal is conveyed to the input of block D providing a PID control.

The outputs of the controllers P, I and D converge in the adder node S2 on the right of figure 1 to control the internal combustion engine. Generally, the controllers receive speed signals at input and generate control signals relative to the torque percentage that the engine has to deliver, with respect to the relative nominal torque.

The derivative contribution D is given by the derivative of the engine rotation speed ASpeed multiplied by a fixed or adjustable parameter KD .

The Sign block receives, at input, the above-mentioned error and also the derivative of the speed ASpeed from the derivative block D and compares the relative signs, allowing or not allowing, by means of the Switch, the output of the derivative D to flow into the adder node S2. Therefore, when the switch enables the derivative output, a full PID control is obtained, otherwise a PI type control is obtained .

Preferably, the Sign block enables the contribution of the derivative controller when both signs are positive and also when they are both negative.

According to a preferred variation of the invention, the Sign block not only compares the signs but also verifies that when both are positive, the speed error exceeds a first predetermined positive threshold and the derivative exceeds a second predetermined positive threshold. When both conditions, regarding concordance of the signs and exceeding of the respective thresholds, are met, then the derivative contribution is used.

Therefore, verification that also the thresholds have been exceeded (positively and/or negatively) represents a more restrictive condition than sole concordance of the signs.

Vice versa, when both the signs are negative, the Sign block verifies that the speed error is below a third predetermined negative threshold and the derivative contribution is below a fourth predetermined negative threshold. When both the conditions, on the concordance of the signs and exceeding of the respective thresholds, are met, then the derivative contribution is used.

When the module (absolute value) of the first and the third thresholds are equal to each other and the second and fourth thresholds are equal to each other, then the Sign block enables said derivative control when the above-mentioned signs are concordant, and the absolute value of the speed error exceeds a predetermined first threshold and the absolute value of the speed derivative exceeds a second predetermined threshold .

Consequently, a first interval is defined between the first and the third thresholds and a second interval ranging between the second and fourth thresholds, wherein the derivative controller is disconnected from the adder node S2.

In a revolution error/derivative 3 x 3 array, a so-called "dead band" is therefore defined, in which the derivative controller is disconnected from the adder node S2.

The following table shows an application example of the activation mechanism of the derivative contribution as a function of the signs and the thresholds.

The presence of a "1" indicates that the derivative contribution is added in the adder node S2, whereas the "0" indicates that the derivative contribution is not added and therefore is disconnected from the adder node S2.

According to another preferred variation of the invention which combines with any one of the preceding variations, when the Sign block disconnects the first output of the derivative control D from the second adder node S2, and connects a second output of the derivative control D to the integral controller, in particular to the saturator Sat_l with reference to figure 3, which limits the contribution of the integral controller I,

The first output of the derivative control provides KD*ASpeed, while the second output provides only ASpeed.

Therefore, figures 2a and 2b correspond to the equivalent arrangements existing when the Sign block respectively enables or disables the connection of the derivative control output with the second adder node S2.

Therefore, the integral contribution is saturated according to a function F (Err, ASpeed) of the speed error Err and of the speed derivative.

The presence of a "-1" indicates that the derivative contribution is not used for control of the engine, according to the present variation, and a second output of the derivative block, which generates the derivative of the engine rotation speed ASpeed, interacts with the integral controller saturator as described below.

Also to allow interaction of the derivative block (ASpeed) with the saturator in the integral controller, thresholds can be established which the speed error and the speed derivative must exceed (both positively and negatively) , the above- mentioned dead band being defined symmetrical with respect to the two principal and secondary diagonals of the square array.

An integral controller I (discrete time) , according to a preferred variation of the present invention, can be schematized, with reference to figure 3, as a memory which contains a value Int-1 generated by the same integral controller I as in the preceding step (therefore "Int-1" indicates that it is generated at "step-1"), to which the current value of the speed error Err is iteratively added, by means of the adder node S3, expediently multiplied by the integral coefficient KI, by means of the multiplier node Ml. The result of the sum, carried out by the adder node S3, represents the output of the integral controller I, i.e. the above-mentioned control signal, and simultaneously the input Int-0 of the Memory (i.e. at "step-0" i.e. current step) which stores it for the subsequent integration step.

The saturator Sat_l, according to the present invention, is arranged between the adder node S3 and the input of the Memory block, so that the limitation is performed not only on the integral controller output, but also on the Memory block contained in it .

According to a preferred variation of the present invention, the saturator Sat_l carries out a symmetrical saturation relative to zero and the saturation module is given by the following formula:

| Saturation | = |K * RadQ { [ exp ( - | Err | / A) ] * [ exp ( -

| ASpeed | / B ) ] } |

Where :

- RadQ represents the operator square root

- Exp represents the exponential - I Error I represents the module of the error Err described above

- | ASpeed | represents the module of the speed derivative of the engine ASpeed.

K, A and B represent constant values.

The speed derivative of the engine can be expressed by means of the Newton equation:

ASpeed = Cost * [ (Torque delivered by the engine) - (Resistant torque due to the external loads) ]

Where Cost is generally the reciprocal of the moment of inertia J of the internal combustion engine.

The external loads include, for example, the gradient of the road travelled by the vehicle guided by the internal combustion engine or the resistant torque offered by an electric generator, a compressor or a pump connected to a relative PTO (power take-off) .

The CALC block based on the inputs:

- Speed error Err

- Speed derivative ASpeed

applies the above-mentioned formula calculating the saturation module | Saturation | .

The CALC block has two outputs, each addressed to the two inputs: high and low of the saturator Sat_l .

A positive or negative saturation is only applied when the Err and ASpeed signs are discordant, otherwise the signal is saturated to +/- 100% of the actuator authority.

A second saturation Sat_2 is preferably arranged between the multiplier node Ml and the adder node S3 in order to limit the control signal to +100% and -100% of the so-called "actuator authority". The reasons for implementation of the saturations to the actuator authority are known to a person skilled in the art and are substantially intended to avoid unnecessarily requesting a performance of the actuator (Engine) in excess of the relative nominal characteristics. In this case the actuator is the internal combustion engine Engine which cannot deliver more than the relative nominal torque in the relative speed/nominal torque map.

According to a preferred variation of the invention, the above-mentioned formula is implemented in the CALC block by means of a Look up Table, with the advantage of greater flexibility as it allows modification of the coefficients of the table itself, introducing deviations from the output of the above-mentioned formula, which are better suited to the specific internal combustion engine. Furthermore, the Look up Table allows reduction of the computational burden.

An example of Look up Table is shown below having | ASpeed | and | Err | as inputs .

The values shown in the array, for example 5 x 5, in the preceding table are those considered optimal for an implementation of the present invention. Nevertheless, they can be appropriately varied. Said values are positive when the error Err is negative and ASpeed is positive, whereas each of the values shown in the table are multiplied by "-1" when Err is positive and ASpeed is negative.

In other words, the saturator Sat_l is symmetrical relative to zero and when the error is negative and ASpeed is positive, it indicates manipulation of the upper (positive) portion of the saturator (upper limitation) ; vice versa when Err is positive and ASpeed is negative, it indicates manipulation of the lower (negative) portion of the saturator (lower limitation) .

The present method can be advantageously implemented by means of an ECU (Engine Control Unit) processing unit for controlling the internal combustion engine, which processes the information on engine speed and pressure on the accelerator pedal or requests for revolutions or torque from other devices and consequently controls the internal combustion engine as described above.

The present invention can be advantageously carried out via a computer program that comprises coding means for the realization of one or more steps of the method, when this program is run on a computer. It is therefore understood that the protective scope extends to said computer program and also to computer readable means that comprise a recorded message, said computer readable means comprising program coding means for the realization of one or more steps of the method, when said program is run on a computer.

Embodiment variations of the non-limiting example described are possible, without departing from the scope of the present invention, comprising all the embodiments equivalent for a person skilled in the art . From the above description a person skilled in the art is able to produce the subject of the invention without introducing further construction details. The elements and the characteristics illustrated in the various preferred embodiments, including the drawings, can be combined with one another without departing from the scope of the present application. What is described in the chapter relative to the state of the art is provided only for a better understanding of the invention and does not represent a declaration of existence of what is described. Furthermore, if not specifically excluded in the detailed disclosure, what is described in the chapter on the state of the art should be considered an integral part of the detailed disclosure.

Claims

1. A control method for an internal combustion engine carried out by means of a feedback control of the speed of the engine itself, based on an error (Err) of said speed of the engine, calculated between a reference value (Ref) and a measured value (Speed) of the speed, wherein said control comprises a sum (S2) of a proportional contribution (P) and an integral contribution (I), the method comprising a step of also adding a derivative contribution (D) only when it occurs at least that a first sign of said error and a second sign of a derivative of said engine speed are concordant to one another.

2. The method according to claim 1, wherein said derivative contribution (D) is added when said speed error exceeds a first predetermined positive threshold and said derivative of the engine speed exceeds a second predetermined positive threshold and/or when said speed error is lower than a third predetermined negative threshold and said derivative contribution is lower than a fourth predetermined negative threshold .

3. The method according to claim 2, wherein a module of said first threshold and a module of said third threshold coincide, and wherein a module of said second threshold and a module of said fourth threshold coincide.

4. The method according to any one of the preceding claims 1- 3, wherein the method comprises a step of saturating said integral contribution as a function of (F (Err, ASpeed) ) :

- a derivative (ASpeed) of said measured value of the speed (Speed) of the engine, and

- said speed error (Err) ,

when said first sign of said error and said second sign of said derivative contribution are at least discordant to one another .

5. The method according to claim 4, wherein said integral contribution (I) consists of the sum (S3) of

- an error value (Err) calculated at the current step, and multiplied by an integral coefficient (KI) and

- an integral contribution value (Int-1) generated at the immediately preceding step,

and wherein said saturator (Sat_l) is applied to said sum.

6. The method according to one of the claims 4 or 5, wherein an absolute value of said saturation is given by the following formula :

| Saturation | = |K*RadQ { [ exp ( - |Err| / A) ] * [ exp ( -

|hSpeed| / B ) ] } |

Where :

- RadQ represents the operator square root

- Exp represents the operator exponential

- | Err | represents the module of the speed error Err described above,

- |hSpeed| represents the module of the derivative of the engine speed ASpeed,

- K, A and B represent constant values.

7. The method according to any one of the preceding claims, wherein said absolute value of said saturation is given by a Look up table, wherein relative values are included, in absolute values, between 0 and 100% of a nominal torque of the internal combustion engine.

8. A computer program that comprises program coding means adapted to carry out all the steps of any one of the claims from 1 to 7, when said program is made to run on a computer.

9. Computer readable means comprising a registered program, said computer readable means comprising program coding means adapted to carry out all the steps of any one of the claims from 1 to 7, when said program is made to run on a computer.

10. An internal combustion engine comprising

- a device for measuring a relative speed,

- a processing unit (ECU) configured to control said internal combustion engine based on a speed reference signal (Ref) , said processing unit being configured to carry out the control method according to any one of the preceding claims from 1 to

7.

11. A vehicle or fixed plant comprising the internal combustion engine according to claim 10.