FR3056661A1 - LEARNING METHOD FOR CONTROLLING A CLUTCH - Google Patents

Abstract

A training method for driving a clutch of a power train of a motor vehicle with a discrete ratio gearbox, the control being controlled, comprising, during a period during which the engine is allowed to rotate at idle and the clutch is open (E2, E3, E4), a step of detecting the stabilization (100) of the torque supplied by the engine running at idle, a step of controlling the progressive closing of the clutch (E3 ) after the detection of the stabilization of the torque, and a step of detecting a reaction (200) of the dynamics of the engine to identify the point of licking of the clutch during said progressive closing.

Description

Holder (s): PEUGEOT CITROEN AUTOMOBILES SA Société anonyme.

Extension request (s)

Agent (s): PEUGEOT CITROEN AUTOMOBILES SA Public limited company.

54) LEARNING PROCESS FOR DRIVING A CLUTCH.

FR 3 056 661 - A1 f5j) Learning method for piloting a clutch of a power chain of a motor vehicle with gearbox with discrete ratios, the control being piloted, comprising, during a period during which the engine is left idling and the clutch is open (E2, E3, E4), a step of detecting the stabilization (100) of the torque supplied by the engine idling, a step of controlling the progressive closing of the clutch (E3) after the detection of the stabilization of the torque, and a step of detection of a reaction (200) of the engine dynamics to locate the point of licking the clutch during said closing progressive.

LEARNING METHOD FOR DRIVING A CLUTCH The invention is in the field of motor vehicle control, in particular motor vehicles with thermal engine (or even cars with hybrid thermal and electric propulsion) having , on their traction power chain of a gearbox with several discrete gear ratios, the gearbox being manual, that is to say that the gear engaged is chosen by the driver.

The vehicles mentioned above have a clutch on their traction power chain.

Some manual gearboxes are said to be controlled. They include a computer that controls the opening and closing of the clutch and the transmission of gear ratios, based on instructions given by the driver.

These gearboxes have a function, essential, for regular learning of the licking point of the clutch. It is necessarily carried out when the gearbox is in neutral, and by observing the variation (increase) in the speed of the primary shaft of the gearbox.

The point of licking the clutch is the position of the clutch where the discs begin, during the closing of the clutch, to come into friction with each other, so that from l energy begins to be transmitted through the clutch. Knowledge of the licking point allows you to reset the position transfer function of the actuator / transmitted torque and thus control the clutch control. The objective of the licking point determination function is therefore to gain reliability and consistency in the torque transmission benefits by the clutch (at start-up and during a gear change).

Known manual gearboxes are thus known for which the learning of the licking point is carried out with the engine running, gearbox in neutral and by detection of the rotation of the primary shaft.

[0007] However, there is no known manual gearbox of which only the clutch is controlled, and which would similarly have a licking point learning function. Indeed, in such a box, the return to neutral is not controlled by the computer, since it remains the responsibility of the driver. Thus, there is no easy possibility of carrying out a regular learning of the licking point of the clutch.

The rare occasions when the gearbox is neutral are essentially the so-called Stop & Start situations, that is to say engine shutdown when the vehicle is near stop, waiting for a new start. But then the engine being off, it is impossible to carry out a learning of the licking point which requires in any event that the engine is running.

But the manual gearboxes of which only the clutch is controlled are attractive because they offer a possibility of saving energy by opening the automatic clutch, with a view to reducing the engine brake, in a favorable situation for fuel consumption (slight deceleration, without depressing the accelerator pedal, nor depressing the brake pedal). The driver keeps a gear engaged while the automatic clutch is released. The term "freewheeling" is used (for evolution in freewheeling).

So we had to look for another way to learn the point of licking for manual gearboxes with controlled clutch.

DE102014208781 presents a method for determining a licking point, when the vehicle is stationary. This is not suitable for regular learning on a vehicle during the journey.

The invention overcomes the disadvantages of existing methods.

To obtain this result, it consists of a learning process for controlling a clutch of a power chain of a motor vehicle with a discreet gearbox, the clutch control being controlled, comprising, during a period during which the engine is left idling and the clutch is open while a gear ratio remains engaged, a step of detecting the stabilization of the torque supplied by the engine idling, a step of controlling the progressive closing of the clutch after the detection of the stabilization of the torque, and a step of detecting a reaction of the engine dynamics to locate the point of licking the clutch during said closing progressive. In this process, the vehicle coasts and a gear ratio remains engaged, said reaction of the engine dynamics corresponding to a transfer of energy from the wheels to the engine idling.

Optional and advantageous characteristics can be introduced, among which:

The step of detecting a reaction of the engine dynamics includes the detection of the crossing, by the torque supplied by the engine, of a downward threshold;

- The step of detecting a reaction of the engine dynamics includes monitoring the derivative with respect to the time of the engine rotation speed, taking into account the engine rotation inertia;

- The method is implemented in a vehicle with manual gearbox during a period during which the clutch is open under the control of a computer implementing a function of removing the engine brake to search a reduction in fuel consumption, independently of the action of the driver on the choice of the gear ratio engaged and its maintenance;

- such a succession of steps to identify the point of licking is carried out periodically.

The invention also relates to a learning device for controlling a clutch of a power chain of a motor vehicle with discrete gearbox, the clutch control being controlled, comprising means for, during a period during which the engine is left idling and the clutch is open, detecting the stabilization of the torque supplied by the engine idling, controlling the gradual closing of the clutch after detection stabilization of the torque, and detect a reaction of the engine dynamics to locate the point of licking the clutch during said progressive closure. In this device, the means are adapted so that, with the vehicle coasting and a gear ratio remaining engaged, said reaction of the engine dynamics corresponds to a transfer of energy from the wheels to the engine idling.

The invention also relates to a gearbox with discrete ratios for the traction power chain of a motor vehicle, comprising a learning device as mentioned above. It can be a robotic box.

The invention also relates to a motor vehicle comprising, on its traction power chain, a gearbox with discrete ratios, the gearbox being as mentioned above.

The invention will be better understood, and other objects, characteristics, details and advantages thereof will appear more clearly in the explanatory description which follows, made with reference to the appended drawings given solely by way of example illustrating a embodiment of the invention:

- Figure 1 being a representation of an implementation of an embodiment of the invention;

- Figure 2 is a flow diagram showing the implementation of the embodiment mentioned.

Figure 1 relates to a power chain of a motor vehicle with discreet gearbox, the clutch control being controlled.

It presents in the upper part the evolution, on the ordinate, of the regime values (number of rotations of a tree per unit of time) as a function of time during a scenario, represented on the abscissa. The lower part shows the evolution, on the ordinate, of the torque values, always as a function of time, the course of which is identical in the two parts of the figure.

Curve 10 shows the speed of the primary shaft (or clutch output shaft) and curve 20 that of the engine.

Curve 30 shows the evolution of the torque available on the clutch output shaft, and curve 40 shows the evolution of the torque available at the output of the engine.

The figure shows a scenario in which initially the engine was in operation to provide traction power, the driver pressing the accelerator and having a gear ratio engaged, and in which one enters a period during which the driver stops pressing the accelerator but keeps the gear engaged, the available torques on the clutch output shaft and the engine output suddenly decrease, then the engine is left idling and l clutch is open. The clutch is for example opened under the control of a computer implementing a function for removing the engine brake in order to seek a reduction in fuel consumption, independently of the action of the driver on the choice of the gear ratio engaged and of its maintenance.

At an instant tO, a computer controls the opening of the clutch which is controlled, this opening being done over a period of time lasting from tO to approximately t1. During this period of time the available torques on the two shafts continue to fall, and the revs of the two shafts decrease while remaining equal to each other, due to their connection still in progress by the clutch. Maintaining a rotation is linked to the rotation of the wheels, imposed by the inertia of the vehicle, and the reduction in speeds which is presented here linear, is linked to the deceleration of the vehicle caused by friction of the ground and the air. , and possibly a slight upward slope.

From time t1, the vehicle evolves in "freewheeling", that is to say "freewheeling", the wheels being no longer connected to the engine. The clutch output shaft continues to have a positive speed which decreases linearly, again due to the inertia of the vehicle and its deceleration, two phenomena not influenced by the opening of the clutch.

On the other hand, the torques available on the shafts both reach a minimum (-20 Nm for example, non-limiting value) and around time t1, a computer triggers the engine idling speed and regulates the torque of the motor output shaft, causing the available torque on its output shaft to increase after reaching its minimum, towards a low value (which can be 0 Nm or in any case typically lower, in absolute value, to 10 Nm, nonlimiting value), intermediate between the torque which was available during acceleration and the minimum, and closer to the minimum value than to the value corresponding to acceleration. The engine output shaft keeps idling at a low speed value, for example 750 rpm (non-limiting value). The torque available on the clutch output shaft remains close to the low value.

The period E1, extending from t0 to t2 is a period of entry into freewheeling evolution.

A duration is observed from t2 (for example at least 5 seconds, that is to say a prolonged phase of coasting) so that the engine idling behavior is reached in a stabilized manner. A detector searches for stabilization of the engine speed value, which in the scenario shown is observed at time t3, corresponding to a stabilization point 100. The value of the available torque is also recorded on the output shaft of the engine.

This is a first function, implemented in a period E2 extending between t2 and t3, of stabilization at idle. Once completed, a second function is implemented.

From this instant t3, it is carried out, by the computer, a search for the licking point, by a slow and cautious action of the clutch closing command. The computer monitors the evolution of available torque on the engine output shaft, and notes that it is decreasing. This is a reaction of the engine dynamics, corresponding to a transfer of energy from the wheels to the idling engine. In the embodiment described, the computer monitors the exceeding, by this torque, of a threshold S typically 5 to 10 Nm (non-limiting value), downward, beyond which it is considered that the licking point 200 was reached. During this time, available torque appears on the clutch output shaft. The computer controlling the idle speed, due to the torque drawdown, slightly increases the rotation speed of the engine output shaft.

In the scenario shown, the torque drop threshold is reached at time t5. The corresponding value of the clutch closing command is memorized. The clutch closing command is interrupted and the clutch reopened from this time t5. The period E3, extending from t3 to t5 is a period of searching for the licking point.

From this instant t5, the torque available on the engine output shaft returns to the idle value, and the speed of this shaft, too. The torque available on the clutch output shaft returns to a low value, lower than the value available on the engine output shaft. At time t6, the torque on the clutch output shaft becomes less than the torque on the engine output shaft. The phase E4 following t5 is the exit phase from learning.

The modification of the clutch closing and opening control is carried out, in two stages, very gradually so as not to cause feelings for the driver and his passengers.

This process of finding the value of the clutch closing command corresponding to the licking point is carried out regularly, so as to achieve permanent control of the piloted clutch. Preferably, this search process is carried out every 10 minutes at least.

In Figure 2, there is shown the following steps of the invention.

In a step E10, the driver releases the accelerator, then, in a step E11, the clutch opens, under the control of a computer controlling the clutch. In a step E12, the idling is triggered, under the control of an idling regulator. A few moments later, the idle stabilizes during a step E13.

A learning phase then begins, under the control of a learning logic controller.

During a step E14, the clutch is partially closed, with a clutch increment applied to its control. At the end of this step, a test step E15 is carried out, during which the computer checks whether the condition mentioned above is reached. This is the condition for locating the point of licking.

If the condition is not reached, the learning logic controller returns to step E14 and introduces an additional increment in the closing of the clutch.

If the condition is reached, there is, during a step E16, recording of the licking point, that is to say of the value of the clutch closing command, by the computer. learning logic.

Then, a declutching step E17 is implemented.

The steps E14 and E17 are carried out slowly, so as not to cause a sudden blow by the driver or the passengers.

In an alternative embodiment, we take all the principles and all the steps above, but we seek, during the careful closing of the clutch while the engine is idling, the evolution of the derivative ( positive) with respect to the time of the engine output shaft rotation speed, taking into account the engine rotation inertia. This detection mode allows a more reactive detection, in particular if the idle governor reacts slowly, and does not influence the derivative of the engine speed.

In this case, the derivative of the engine speed leads to a reduction in the torque available on the engine shaft, proportional to the inertial torque of the engine. Thus we follow the progressive subtraction by the engine inertial torque (defined as the product of the engine rotation inertia and the derivative of the engine speed) of part of the engine torque. When a threshold of this subtraction is reached, it is considered to constitute a reaction of the engine dynamics, corresponding to a transfer of energy from the wheels to the idling engine and indicating that the point of licking the clutch is reached. . The corresponding command value is noted. The clutch is also closed.

These principles, in the two embodiments, make it possible to carry out a licking point learning strategy, for a manual gearbox equipped with a controlled clutch, which offers a good approval of the “evolution in freewheel >>. It therefore makes it possible to offer this "freewheeling evolution" function on a wide range of vehicles fitted with manual gearboxes. The function allows the reduction of fuel consumption.

The control is efficient, and offers good reliability and employability of the "freewheeling" function, with a manual gearbox. The information from the learning process is used to control the clutch during the following routes.

We can implement the principles mentioned in a learning device for driving a clutch. Such a device comprises means for, during a period during which the engine is left idling and the clutch is open - detecting the stabilization of the torque supplied by the engine idling, [0054] - controlling the progressive closing of the clutch after the detection of the stabilization of the torque, and detecting a reaction of the engine dynamics to locate the point of licking of the clutch during the progressive closing.

This device can be inserted in a discreet gearbox, itself installed in a motor vehicle comprising, on its traction power chain, such a box. The box can be robotized.

Claims (9)

1. A learning method for controlling a clutch of a power chain of a motor vehicle with discrete gearbox, the clutch control being controlled, comprising, during a period during which the engine is left idling and the clutch is open (E2, E3, E4), a step of detecting the stabilization (100) of the torque supplied by the engine idling, a step of controlling the progressive closing of the clutch (E3) after detecting the stabilization of the torque, a step of detecting a reaction (200) of the engine dynamics to locate the point of licking the clutch during said progressive closing, characterized in that, with the vehicle coasting and a gear ratio remaining engaged, said reaction of the engine dynamics corresponds to a transfer of energy from the wheels to the engine idling. 2. A learning method according to claim 1, characterized in that the step of detecting a reaction (200) of the engine dynamics comprises the detection of the crossing, by the torque supplied by the engine (40), d '' a torque variation threshold (S). 3. A learning method according to claim 1, characterized in that the step of detecting a reaction (200) of the engine dynamics comprises monitoring the derivative with respect to the time of the engine rotation speed, account given the engine rotation inertia. 4. Learning method according to one of claims 1 to 3, characterized in that it is implemented in a vehicle with manual gearbox during a period during which the clutch is open (E1, E2, E3, E4) under the control of a computer implementing a function for removing the engine brake to seek a reduction in fuel consumption. 5. Learning method according to one of claims 1 to 4, characterized in that such a succession of steps (100, E3, 200) for locating the point of licking is carried out periodically. 6. Learning device for controlling a clutch of a power chain of a motor vehicle with discrete gearbox, the clutch control being controlled, comprising means for, during a period during which the engine is left idling and the clutch is open (E2, E3, E4), detecting the stabilization (100) of the torque supplied by the engine idling, controlling the gradual closing of the clutch ( E3) after detection of the 5 stabilization of the torque, and detecting a reaction (200) of the engine dynamics to locate the point of licking the clutch during said progressive closing, characterized in that, the vehicle advancing in freewheel and a gear ratio remaining engaged, said reaction of the engine dynamics corresponds to a transfer of energy from the wheels to the engine idling. 10 7. Discrete gearbox for motor vehicle traction power chain, characterized in that the gearbox includes a learning device according to claim 6. 8. Gearbox according to claim 7, characterized in that it is a robotic gearbox. 15 9. Motor vehicle comprising, on its traction power chain, a gearbox with discrete ratios, characterized in that the gearbox is according to claim 7 or claim 8. 1/2 -r ω ω _α> Q. = 5 Ο Ο Diet 2/2