JPH0742828A - Clutch engagement control device - Google Patents

Abstract

PURPOSE:To perform judgement of the non-controllable zone and. controllable zone distinguishedly inevitably to make control of clutch engagement accurately under all possible conditions including a transient operation. CONSTITUTION:A clutching system is turned into model using two sorts of models, and by the use of the least square parameter identification logic, the slip amount presumative value N1 with influence f3 (D) of the control taken into consideration is determined from the formula N1=f1(N)+f2(TVO)+f3 (D)on one of the models, while the slip amount presumative value N2 with no influence f3 (D) of the control taken into account is determined on the other model, where N is real slip amount value of lockup clutch, TVO is the degree of opening of an engine throttle, and D is control command value. When the difference ¦N-N2¦ between the real slip amount value N and presumative value N2 is less than the set value, a judgement as noncontrollable zone is passed and the slip of a torque converter is put in feedforward control, and when over the set value, a judgement as controllable zone is passed to lead to transition to the feedback control.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to engagement control of a hydraulically actuated clutch such as a lockup clutch of a torque converter used between an engine and an automatic transmission and a shift control clutch of the automatic transmission itself. It relates to the device.

[0002]

2. Description of the Related Art As an engagement control device for this type of clutch, "RE4R" issued by Nissan Motor Co., Ltd. in March 1987.
As seen in the lockup clutch control technology for automatic transmissions described in "01A Type Automatic Transmission Maintenance Manual", after the moment when the slip amount represented by the input / output rotation difference of the lockup clutch falls to the set slip amount, There is a system in which the lockup clutch pressure is feedback-controlled so as to decrease with the set gradient, and thereby the lockup of the converter is advanced.

However, since the above-mentioned set slip amount has been constant in the past, in view of the demand for fuel efficiency improvement as in recent years,
When attempting to further reduce the speed of the lock-up region and increase the number of gears, the lock-up control must be performed in an extremely wide operating state, which causes the following problems. (1) Since the slip rotation speed before starting the control has a larger value than in the conventional case, if the set slip amount is fixed, the wasteful time until the control shifts to the feedback control becomes long, and the fuel efficiency effect due to the lockup is increased. Fades. (2) Since the slip amount before starting the control varies greatly depending on the operating state, the fixed set slip amount cannot be handled and must be variable. In addition, matching man-hours for that purpose occur.

To solve these problems, FIG.
It is possible to control the engagement of the lockup clutch as shown in FIG. In this control, there is no correlation between the change of the control command value which is the control input and the change of the slip amount which is the control output, that is, the slip amount (engagement degree) of the clutch is irrespective of the control command value. In the so-called control dead zone that changes, the feed-forward control such as rapidly filling (precharging) the hydraulic fluid in the lockup clutch for the initial fixed time and then increasing the operating pressure at a constant speed (ramp control) is performed. Advance the engagement of the clutch. Then, in a so-called controllable range where a change exists in the control command value, which is the control input, and a change in the slip amount, which is the control output, the lock-up clutch is engaged while monitoring the slip amount. Feedback control.

Here, FIG. 6 shows the torque converter T / C between the engine E and the automatic transmission A / T from the converter state in which the direct connection between the input / output elements is released, and the input / output element is engaged by engaging the lockup clutch. 1 shows a device for engaging and controlling a lockup clutch when a lockup state in which the spaces are directly connected is established. The determination of the control dead zone or the controllable zone performed in this device will be described below with reference to FIG.

Although the slip amount set value calculation unit S / T constantly receives the slip amount N of the torque converter T / C (lock-up clutch), the slip amount N _1S at the start of the precharge control is set as shown in FIG. 20% from memory
A low slip amount is obtained as the slip amount upper limit set value N _SU , and the slip amount lower limit set value N _SL = 20 rpm is set. The controller CNT compares the slip amount upper limit set value N _SU and the slip amount lower limit set value N _SL with the slip amount true value N, and as shown in FIG.
The instant t _{1 at} which the slip amount upper limit set value N _SU decreases,
It is considered that the precharge control region (control dead band for feedforward control) has shifted to the PID control region (controllable region for feedback control), and then the true value N of the slip amount reaches the lower limit set value N _{SL of the} slip amount. It is considered that the lowered instant t ₂ is transferred from the PID control region to the ramp control region (control dead band for feedforward control).

Based on these determination results, the controller CNT commands the control command value D as shown in FIG. 7 to the lockup clutch of the torque converter T / C by switching the logic and performing the corresponding feedforward control or feedback control. To do.

[0008]

However, in such a configuration, P from the precharge region (control dead zone)
When discriminating the shift to the ID control region (controllable region), the slip amount N is compared with the set value N _SL, and the discrimination is made according to the result. Therefore, as shown in FIG. 7, the accelerator pedal is substantially constant. During steady operation with the pedal depressed,
Although the shift from the control dead zone to the controllable zone can be accurately determined, as shown in FIG. 8, when the lockup clutch is engaged during transient operation such as acceleration while depressing the accelerator pedal, the following description will be given. As described above, the transition from the control dead zone to the controllable zone cannot be accurately determined.

That is, during such a transient operation, the torque input largely changes, and the slip amount changes drastically in addition to the control command value, so that the control dead band widens as shown in FIG. Therefore, although the slip amount true value N is still in the uncontrollable range at the instant t _{1 when} it falls to the slip amount upper limit set value N _SU , according to the conventional determination described above, the feedback control is already started at this time. It will end up. Therefore, the decrease of the slip amount N is delayed, and this is sharply reduced from the instant t ₃ which is outside the control dead zone and enters the controllable zone,
In addition to the engagement shock of the lock-up clutch, the slip amount N greatly overshoots the control target value N _SM , resulting in control hunting.

An object of the present invention is to solve the above-mentioned problem by making it possible to accurately determine the transition from the control dead zone to the controllable zone even in a transient operation state.

[0011]

To this end, according to the present invention, as set forth in claim 1, in the control dead zone in which the degree of engagement of the clutch changes regardless of the control command value, the degree of engagement of the clutch is feedforward. In the clutch engagement control device, the engagement degree of the clutch is controlled in a controllable range other than the control dead zone by feedback control according to the slip amount represented by the input / output rotation difference of the clutch. Using a clutch system estimation system preset to estimate the amount, from the torque input to the clutch, the slip amount and the control command value,
Control command value contribution degree estimating means for estimating the contribution degree of the control command value to the slip amount, and when the contribution degree estimated by the means becomes equal to or more than a setting, as a transition from the control dead zone to the controllable range, Area determining means for switching from the feedforward control to the feedback control is provided.

According to a second aspect of the present invention, as the clutch system estimation system, at least two types of clutch system estimation systems are set, one with and without the control command value. The control command value contribution estimating means can estimate the contribution by comparing the slip amount estimated value based on these systems with the true value of the slip amount.

Further, according to a third aspect of the present invention, as a clutch system estimation system, a slip amount and its past value,
A clutch system estimation system that includes a model composed of a control command value and its past value, a clutch torque input and its past value, and an estimated value that is the output of the model, and a part based on the control command value from this model is deleted. Depending on the difference obtained by comparing the estimated value that is the output and the actual slip amount,
The control command value contribution estimating means can estimate the contribution.

Further, in the present invention, as described in claim 4, the control command value contribution estimating means identifies the clutch system estimation system by the model, and the control command value is obtained from the model obtained as a result. It is possible to estimate the contribution degree according to a difference obtained by comparing an estimated value which is an output of the system in which the portion due to is deleted and the true value of the slip amount.

Further, in the present invention, as described in claim 5, the clutch system system is identified by the model with respect to the clutch system estimation system, and the part by the control command value obtained as a result is defined as a predetermined standard system. The control command value contribution degree estimating means can estimate the contribution degree from the difference obtained by comparison.

[0016]

In the structure of claim 1, in the control dead zone,
When performing feed-forward control of the degree of engagement of the clutch and performing feedback control of the degree of engagement of the clutch in the controllable range according to the slip amount of the clutch, the control command value contribution estimating means preliminarily calculates the slip amount. Using the set clutch system estimation system, the contribution of the control command value to the slip amount is estimated from the torque input to the clutch, the slip amount of the clutch, and the control command value to the clutch. When the value becomes equal to or more than the set value, the area determination means determines that the control dead zone has shifted to the controllable area and switches from the feedforward control to the feedback control.

Thus, the transition from the control dead zone to the controllable zone can be accurately discriminated even in the transient operation state, and the feedback control is executed irrespective of the control dead zone so that the control becomes random and the clutch is large. It is possible to solve the problems that the engagement shock occurs, the slip overshoots the control target value, and the control becomes unstable.

Further, in the above configuration, for example, as described in claim 2, as the clutch system estimation system, at least two kinds of clutch system estimation systems are set, one with and without the control command value. The control command value contribution estimating means can be embodied by estimating the contribution by comparing the slip amount estimated value based on these systems with the true value of the slip amount.

As described in claim 3, this embodiment is configured as a clutch system estimation system by a slip amount and its past value, a control command value and its past value, a clutch torque input and its past value. The model is provided with an estimated value which is an output of the model, an estimated value which is an output of the clutch system estimation system in which a part based on the control command value is deleted from the model, and an actual slip amount. Accordingly, the control command value contribution degree estimating means estimates the contribution degree, which makes it more realistic.

Further, at this time, as described in claim 4, the control command value contribution estimating means identifies the clutch system estimating system by the model, and from the model obtained as a result, the part by the control command value is obtained. According to the difference obtained by comparing the estimated value that is the output of the system with the true value of the slip amount, if the contribution is estimated,
The contribution can be estimated more effectively.

Instead of this, as described in claim 5, the clutch system is identified by the model thereof with respect to the clutch system estimation system, and the part by the control command value obtained as a result is compared with a predetermined standard system. Even if the control command value contribution estimating means estimates the contribution from the difference obtained as described above, the contribution can be effectively estimated.

[0022]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 is a functional block diagram showing a clutch engagement control device according to an embodiment of the present invention. In the present example device, a torque converter T / C between an engine E and an automatic transmission A / T is connected to the input device. Lock-up clutch (not shown) from converter state where direct connection between output elements is released
When the lock-up state is established in which the input and output elements are directly connected by the engagement of, the lock-up clutch is configured as a device for engaging control.

For this reason, in this example, the clutch system of the lockup clutch is modeled by two simple ARMA models, and both models are sequentially identified so that they produce outputs similar to those of the actual clutch system. In this identification, a well-known least square method parameter identification logic is used, and this is performed as follows.

That is, first, the shadow of the control is
Hibiki, that is, controller CNT to torque converter T
/ C shadow of control command value D toward lock-up clutch
Hibiki f ₃ Considering (D) (however, f₃ Is a parameter)
The estimated slip amount N1 of the backup clutch is N1 = f
₁ (Slip amount true value N of lock-up clutch) + f ₂ 
(Engine throttle opening TVO) + f₃ (Control command value
D) is obtained (however, f₁ , F₂ Each is a parameter
Data).

Next, according to the other model, the estimated slip amount N2 of the lockup clutch which does not consider the influence of the above control f ₃ (D) is set to N2 = f ₁ (true slip amount N of the lockup clutch N) + f ₂ (Engine throttle opening TVO)

Here, in the least square method parameter identification logic, the parameters f ₁ and f are set so as to minimize the square of the difference between the estimated output value of the clutch system and the actual output value with respect to both models. ₂ and f ₃ are determined. Therefore, in this example, the estimated slip value N1 and the true slip value N are compared, and the parameters f ₁ , f ₂ , f ₃ are determined so that the square of the difference between them is minimized. Therefore, it does not matter whether the two models really match the actual clutch system, and it is sufficient if the future value can be predicted within a certain error range at least in the near future. .

By the way, in the control dead zone where the control does not influence the change of the slip amount, the slip amount estimated values N1 and N2 have the same difference as shown by the instant t _{1 in} FIG. In the controllable range in which the control influences the change of the slip amount, the slip amount estimated value N1 as shown after the instant t _{1 in} FIG.
Only the slip amount true value N follows, and the slip amount estimated value N
2 cannot follow the true value N of the slip amount, and the estimated value N2
Based on the magnitude of the difference between the estimated value N1 or the true value N (| N−N2 | in FIG. 1), that is, the degree of control contribution, it is possible to discriminate between the control dead zone and the controllable zone.

When the above identification logic is of a sequential type, it is possible to perform precise measurements and examinations in advance.
Further, it is possible to accurately determine the control dead zone or the controllable zone even in a transient operating state in which the slip amount further changes, without being affected by product variations.

In the judgment, the controller CNT judges that | N-N2 | is less than the set value ΔN _S and the contribution of control is small, it is judged as the control dead zone, and | N-N2 | is set value ΔN _S. As described above, when the degree of control contribution is large, the controllable range is determined. Then, if the control dead zone is present, the engagement of the lockup clutch in the torque converter T / C is feedforward controlled, and if it is in the controllable area, the engagement of the lockup clutch is feedback controlled, and so on. Then, the corresponding control command value D is supplied to the lockup clutch.

As shown in FIG. 3, when the lock-up clutch is engaged during the same transient operation as described above with reference to FIG.
From the state where N-N2 | is less than the set value ΔN _S,
Precharge control (feedforward control) is accurately determined at the instant t ₁ when _S is equal to or higher than _S, from the control dead zone where the contribution of control is small to the controllable range where the contribution of control is large. From the PID control to the feedback control can be appropriately executed. Therefore, the feedback control is not performed in the control dead zone, and as is apparent from the change over time of the true slip value N, a lockup clutch engagement shock is generated or a slip is larger than the control target value N _SL. It is possible to eliminate the adverse effects described above with reference to FIG. 8 such as overshooting.

A specific example of the case where the difference equation is used as the clutch system estimation system model will be described below. The current slip amount is expressed as follows.

[Equation 1] Current slip amount = a ₁ × first delay of slip amount + ・ ・ ・ ・ + a ₁ × first delay of slip amount + b ₁₁ × first delay of control command value ・ ・ ・ ・ + b _1m × M-th delay of control command value + b ₂₁ × first-order delay of throttle opening + ... + b _2n × n-th delay of throttle opening + constant term This is because the estimation error of each term order 1, m, n It is determined in advance so as to be small.

On the other hand, the recursive least squares method parameter identification logic uses a ₁ ... so that the square of the error between the actual slip amount true value and the slip amount estimated value is minimized.
_{·· a 1, b 11 ···· b} 1m, to first determine the b ₂₁ ····· b _2n. Then, the current slip amount estimated value N1 is obtained from the difference equation model obtained by parameter identification, and the slip amount estimated value is further calculated based on the difference equation model corresponding to the difference equation from which the term of the control command value is omitted. N2 is obtained, and these are compared with the actual slip amount true value N. When the difference between the two estimated slip amount values and the true slip amount value is approximately the same, it is determined that the contribution of the control command value to the system is small and the control dead zone is reached, and the feedforward control is executed, and two slip amount values are executed. When the difference between the estimated value and the true value of the slip amount is significantly different, the feedback control is executed assuming that the control command value has a large contribution to the system and is within the controllable range.

The flow chart of the block diagram of FIG. 1 is as shown in FIGS. In FIG. 4, first, at step 11, it is checked whether the normal lockup region determination result based on the operating state is the same as or different from the last time.
If they are the same, the torque converter T /
A slip control region where the slip amount of C should be kept at a constant value, or a lockup region where this slip amount should be eliminated,
Check if the converter area does not limit the slip amount at all. In the lockup area, in step 14,
The lockup pressure, which is the engagement pressure of the lockup clutch, is set to the highest value to meet the demand, the lockup pressure is set to the lowest value in step 15 in the converter region, and the slip control region is set to slip in step 16 in the slip control region. The lockup pressure is controlled so that the amount is maintained at a suitable constant value.

If it is determined in step 11 that the lockup area determination result is different from the previous result, then in step 17, whether the lockup area should be engaged or the lockup area is changed.
Check if the lockup area has changed to release the lockup clutch. In the case of a change in the lockup region in which the lockup clutch is to be released, in step 18, the engagement pressure is reduced so that the lockup clutch is released. If there is a change in the lockup area in which the lockup clutch should be engaged, in the step 19,
Whether the controllable range or the control dead band is determined according to whether the control contribution represented by N-N2 | is greater than or less than the set value ΔN _S. If it is in the control dead zone, in step 20, the engagement pressure of the lockup clutch is feedforward-controlled,
If it is within the controllable range, the engagement pressure of the lockup clutch is feedback-controlled in step 21.

In addition to the control of the lockup clutch as described above, in steps 22 and 23, the slip amount of the torque converter is estimated by the parameter identification logic,
The result is stored as a slip amount estimated value N2 and is used as the data in step 19.

The above loop is continued until it is determined in step 24 that the torque converter slip amount has reached the target value, whereby the control dead zone is shifted to the controllable range and the torque converter is locked up as predetermined. Carry out control.

By the way, the processing in steps 22 and 23 is specifically performed as shown in FIG. Assuming that the present process is the Kth process, in step 31, the slip amount estimated value N1 (K) in consideration of the control command value is set to the past slip amount N (K-1), slip amount N (K-2). ...
.. Past throttle opening TVO (K-1), TVO
(K-2) ... And the past control command value D (K-
1), the slip amount estimation logic based on the control command value D (K-2) ...

In the next step 32, the slip amount estimation error (K) is calculated as the slip amount estimated value N1 obtained as described above.
It is calculated as the difference of (K) from the true value N (K) of the slip amount. Then, in step 33, the parameter (K) used in estimating the slip amount is corrected and set by the least square method logic so that the square of the slip amount estimation error (K) is minimized.

Further, in step 34, the slip amount estimated value N2 (K) not considering the control command value is used as the past slip amount N (K-1), slip amount N (K-2) ...
.. Past throttle opening TVO (K-1), TVO
(K-2) ... And the past control command value D (K-
1) = 0, control command value D (K-2) = 0 ...

In order to strictly define the vehicle load, the vehicle speed is required in addition to the throttle opening, and the vehicle speed or the clutch rotation output must be added to the above as an input of the clutch system estimation system. Of course.

[0041]

As described above, the clutch engagement control apparatus according to the present invention feed-forward-controls the engagement degree of the clutch in the control dead zone, and controls the engagement degree of the clutch in the controllable range. When performing feedback control according to the slip amount of the clutch, the clutch system estimation system set in advance to estimate the slip amount is used to determine the slip from the torque input to the clutch, the clutch slip amount, and the control command value to the clutch. A configuration in which the contribution of the control command value to the quantity is estimated, and when the contribution thus estimated exceeds the setting, it is determined that the control dead band has shifted to the controllable range and the feedforward control is switched to the feedback control. Therefore, the transition from the control dead zone to the controllable zone can be accurately determined even in the transient operation state. In this case, the feedback control is executed despite the control dead band, and the control becomes random, which causes a large engagement shock in the clutch, and slips overshoot the control target value to cause unstable control. Can be resolved.

In the above configuration, for example, as described in claim 2, as the clutch system estimation system, at least two kinds of clutch system estimation systems are set, one with and without the control command value. The control command value contribution estimating means can be embodied by estimating the contribution by comparing the slip amount estimated value based on these systems with the true value of the slip amount.

As described in claim 3, this embodiment is configured as a clutch system estimation system by a slip amount and its past value, a control command value and its past value, a clutch torque input and its past value. The model is provided with an estimated value which is an output of the model, an estimated value which is an output of the clutch system estimation system in which a part based on the control command value is deleted from the model, and an actual slip amount. Accordingly, the control command value contribution degree estimating means estimates the contribution degree, which makes it more realistic.

Further, at this time, as described in claim 4, the control command value contribution estimating means identifies the clutch system estimating system by the model, and from the model obtained as a result, the part by the control command value is obtained. According to the difference obtained by comparing the estimated value that is the output of the system with the true value of the slip amount, if the contribution is estimated,
The contribution can be estimated more effectively.

Instead of this, as described in claim 5, the clutch system is identified by the model with respect to the clutch system estimation system, and the part by the control command value obtained as a result is compared with a predetermined standard system. Even if the control command value contribution estimating means estimates the contribution from the difference obtained as described above, the contribution can be effectively estimated.

[Brief description of drawings]

FIG. 1 is a functional block diagram showing a lockup clutch engagement control device for an automatic transmission according to an embodiment of the present invention.

FIG. 2 is a slip amount change time chart showing a control dead zone-controllable area discrimination operation in the same example.

FIG. 3 is a time chart similar to FIG. 2, showing the operation in the case of determining the shift from the control dead zone to the controllable zone by the same discrimination operation.

FIG. 4 is a flowchart showing the example of FIG. 1 as a lockup clutch engagement control program.

FIG. 5 is a flowchart showing a slip amount estimation program in the control program.

FIG. 6 is a functional block diagram showing the best possible lockup clutch engagement control device to date.

FIG. 7 is a time chart showing an area discrimination operation of the same apparatus during steady operation.

FIG. 8 is a time chart showing an area discrimination operation of the same apparatus during a transient operation.

[Explanation of symbols]

 E engine T / C torque converter A / T automatic transmission CNT controller

Claims (5)

[Claims] 1. An engagement degree of the clutch is feed-forward controlled in a control dead zone in which the engagement degree of the clutch changes regardless of a control command value, and the engagement degree of the clutch is changed in a controllable area other than the control dead zone. In a clutch engagement control device for performing feedback control according to a slip amount represented by a clutch input / output rotation difference, a clutch system estimation system preset for estimating the slip amount is used to detect the clutch. Control command value contribution estimating means for estimating the contribution of the control command value to the slip amount from the torque input to the slip, the slip amount, and the control command value, and the contribution estimated by the means is greater than or equal to a setting. When
A clutch engagement control device comprising: a region determination unit that switches from the feedforward control to the feedback control when the control dead band is shifted to the controllable region. 2. The clutch system estimation system according to claim 1, wherein at least two clutch system estimation systems, one with and without the control command value, are set, and the control command value contribution degree is set. The clutch engagement control device, wherein the estimating means estimates the contribution by comparing a slip amount estimated value based on these systems with a true value of the slip amount. 3. The clutch system estimation system according to claim 2, comprising a model constituted by a slip amount and its past value, a control command value and its past value, a clutch torque input and its past value, and the model of the model. Depending on the difference between the estimated value, which is the output, the estimated value, which is the output of the clutch system estimation system in which the control command value is removed from this model, and the actual slip amount, the control command value contribution An engagement control device for a clutch, wherein the estimating means estimates the contribution. 4. The system according to claim 3, wherein the control command value contribution estimating means identifies the clutch system estimation system by the model, and deletes the part by the control command value from the model obtained as a result. A clutch engagement control device characterized in that the contribution is estimated according to a difference obtained by comparing an estimated value which is an output and a true value of a slip amount. 5. The clutch system estimation system according to claim 3, wherein the clutch system system is identified by the model thereof, and a difference obtained by comparing a portion based on a control command value obtained as a result with a predetermined standard system. Therefore, the clutch engagement control device is characterized in that the control command value contribution estimating means estimates the contribution.