JP2007232160A - Lock up controller - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a lock up controller suppressing fluctuation in the connecting timing of a lock up clutch due to fluctuation of individual units, and making the same control for the all vehicles. <P>SOLUTION: The lock up controller calculates an indicated current to a solenoid valve 6, based on the operation state of the vehicle, controls the differential pressure of the lock up clutch 4 by the indicated current to switch and control the lock up clutch 4 for a connection state and a disconnection state. The controller 7 initially sets IP characteristics indicating the relationship between the indicated current and the differential pressure, measures the switching time from the start to the end of the switching control of the lock up clutch 4, and adjusts the IP characteristics by learning according to the switching time. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

The present invention relates to a lockup control device, that is, a control device for a lockup clutch provided in a vehicle torque converter.

Conventionally, a vehicle in which a torque converter is mounted between an engine and an automatic transmission (including a continuously variable transmission) is known. The torque converter is provided with a lockup clutch that mechanically connects the input side and the output side of the torque converter. The lockup clutch is turned on (engaged) based on the driving state of the vehicle (for example, vehicle speed and accelerator opening). ) State and OFF (released) state.
In the case of a vehicle equipped with such a torque converter, control variations occur due to individual differences in the hydraulic control device for controlling the lockup clutch.

FIG. 6 shows the differential rotation (the rotational speed difference between the input side and the output side of the lockup clutch) when changing the lockup clutch from the released state to the engaged state, and the target differential pressure (the engagement side oil chamber of the lockup clutch) It shows the change over time with the hydraulic pressure difference in the release side oil chamber.
After the lock-up ON command is issued, the differential pressure is increased by a predetermined value at time t ₁ , the differential pressure is increased at a constant gradient at time t ₂ after a certain time from time t ₁ , and the differential rotation becomes almost zero. at the time t ₃ that became a maximum differential pressure.
In FIG. 6, N0 indicated by a solid line is a target differential rotation, and P0 is a target differential pressure.

However, even if the differential pressure is controlled to the target differential pressure P0, the actual differential rotation does not always coincide with the target differential rotation N0. As indicated by the broken line N1 in FIG. 6, the switching time T1 becomes longer, or the broken line N2 As shown, the switching time T2 may be shortened. Therefore, when the switching time is long as T1, the target differential pressure is increased as P1, and conversely when the switching time is as short as T2, the target differential pressure is decreased as P2. The learning control is performed so that the switching time becomes the reference value T0.

In the learning method, the target differential pressure is increased or decreased depending on whether the actual switching time is longer or shorter than the reference value T0, and learning control is performed so that the actual differential rotation approaches the target differential rotation. However, the actual differential pressure is not measured, and the variation between the target differential pressure and the actual differential pressure is not taken into consideration. One of the main factors of such variation is variation in characteristics of the solenoid valve. In the conventional case, it is assumed that the characteristics of the solenoid valve, that is, the relationship between the input command current and the output hydraulic pressure is constant, but the characteristics actually vary depending on the individual solenoid valves. Therefore, even though the target differential pressure is increased, the switching time cannot be shortened, or conversely, the switching time cannot be increased despite the target differential pressure being lowered, and individual variations cannot be absorbed, There was a risk of learning emanating.

Patent Document 1 calculates an actual differential pressure based on input torque (engine torque) and corrects a differential pressure command value in order to suppress variations in lock-up clutch engagement timing due to individual variations in torque converters. Is disclosed.
However, since the actual differential pressure is calculated based on the unstable input torque, it is necessary to prohibit learning depending on the situation, and there is a problem that learning becomes unstable. Furthermore, the characteristic variation of the solenoid valve, which is the main cause of the individual variation, is not properly considered, and there is a problem that the same control cannot be performed for all vehicles.
JP 2005-291345 A

An object of the present invention is to provide a lockup control device that can suppress the variation in the engagement timing of the lockup clutch due to individual variation even without means for detecting the actual differential pressure, and can perform the same control for all vehicles. There is.

In order to achieve the above object, the present invention calculates a command signal based on a torque converter having a lock-up clutch and a driving state of the vehicle, and controls the differential pressure of the lock-up clutch based on the command signal to lock the clutch. Lockup control means for switching control between a fastening state and a release state, wherein the lockup control means includes a time measuring means for measuring a switching time from the start to the end of the switching control. A storage means for storing an IP characteristic indicating a relationship between the command signal and the differential pressure, and a learning correction means for learning and correcting the IP characteristic in accordance with the switching time. The lockup control means includes the learning Provided is a lockup control device that controls a differential pressure of the lockup clutch based on a corrected IP characteristic.

First, the switching time from the start to the end of the switching control is measured. The switching time can be measured by a change in differential rotation. Next, the IP characteristic indicating the relationship between the command signal to the lockup control means and the differential pressure is learned and corrected according to the switching time. For example, a target value for the switching time is set, a deviation between the measured switching time and the target value is obtained, and the IP characteristic is learned and corrected using this deviation. When the switching time is longer than the target value, it means that the differential pressure output with respect to the command signal is low due to individual variation, so the differential pressure with respect to the command signal is increased. On the contrary, when the switching time is shorter than the target value, it means that the differential pressure output with respect to the command signal is high due to individual variation, so the differential pressure with respect to the command signal may be lowered.

The learning control of the present invention corrects the IP characteristic, not the differential pressure according to the switching time. That is, since the IP characteristic indicating the relationship between the command signal and the differential pressure is learned and corrected according to individual variation, it can be converged to the same control characteristic in all vehicles, and individual variation can be eliminated.

The command signal in the present invention means an instruction current when a linear solenoid valve is used as a lockup control means, and a duty ratio when a duty solenoid valve is used. Both can output hydraulic pressure proportional to the command signal.

The switching control of the present invention includes not only switching from the released state of the lockup clutch to the engaged state but also switching from the engaged state to the released state. The start and end of switching control can be measured by changes in the differential rotation, but the differential rotation changes from moment to moment, and this is measured in a certain cycle, so the start and end must be clearly determined. Is difficult. Therefore, as a method for determining the start of the switching control when the lockup is ON, for example, the start time when detecting that the differential rotation has decreased a predetermined value (for example, 25 rpm) or more from the released state of the lockup clutch a plurality of times (for example, three times). Can be determined. Further, as a method for determining the end, for example, the end may be determined when it is detected a plurality of times that the differential rotation has become a predetermined value near 0 (for example, 25 rpm) or less. The end of the switching control at the time of lockup OFF may be determined, for example, as the end time when the differential rotation exceeds a predetermined value (for example, 50 rpm).

As described above, in the present invention, the switching time from the start to the end of the switching control of the lockup clutch is measured, the IP characteristic is learned and corrected according to the switching time, and the lockup is performed based on the learned and corrected IP characteristic. Since the clutch differential pressure is controlled, the individual variation is eliminated while the switching control is experienced, and the engagement timing variation of the lockup clutch is eliminated. In particular, since learning can be performed in consideration of the characteristic variation of the solenoid valve, which is one of the main causes of individual variation, the same lock-up control can be performed for all vehicles.
In addition, learning is performed based on the switching time obtained each time, and it is not necessary to estimate the actual differential pressure with unstable input torque. Therefore, it is not necessary to prohibit learning, and learning can be performed stably.

Hereinafter, preferred embodiments of the present invention will be described with reference to examples.

FIG. 1 is a schematic view of a drive system of a vehicle provided with a torque converter according to the present invention.
The vehicle includes an engine 1, a torque converter 2 with a lock-up clutch, and an automatic transmission 3 having a planetary gear device and the like. Although the automatic transmission 3 is connected to the downstream side of the torque converter 2 here, a continuously variable transmission may be connected instead of the automatic transmission 3. The torque converter 2 includes an input-side pump impeller 2a, an output-side turbine runner 2b, and a stator 2c, and a lock-up clutch 4 that mechanically engages and disengages the pump impeller 2a and the turbine runner 2b. Is provided. The pump impeller 2 a is connected to the output shaft 1 a of the engine 1, and the turbine runner 2 b is connected to the input shaft 3 a of the automatic transmission 3. The lock-up clutch 4 is provided with a fastening-side oil chamber 4a on one side and a release-side oil chamber 4b on the other side, and a differential pressure between these oil chambers 4a and 4b is a lock-up control valve 5 serving as a lock-up control means. Controlled by the solenoid valve 6 and the controller 7.

The lockup control valve 5 of this embodiment is a conventionally known one. That is, the spring 5a is biased from one direction, and the output pressure Ps of the solenoid valve 6 is input to the signal port 5c facing the spring 5a. A source pressure Po regulated by a pressure regulating valve (not shown) is inputted to the input ports 5d and 5e. The first output port 5f is connected to the release side oil chamber 4b of the lockup clutch 4, and the release hydraulic pressure Pr is fed back to the port 5g in a direction opposite to the spring 5a. The second output port 5h is connected to the engagement side oil chamber 4a of the lockup clutch 4, and the engagement hydraulic pressure Pa is fed back to the port 5i in the same direction as the spring 5a.

When the output pressure Ps of the solenoid valve 6 input to the signal port 5c is equal to or lower than a predetermined value, the original pressure Po is supplied to the release side oil chamber 4b of the lockup clutch 4 through the input port 5e and the first output port 5f. The lockup clutch 4 is released.
When the output pressure Ps of the solenoid valve 6 input to the signal port 5b rises above a predetermined value, the original pressure Po is supplied to the engagement side oil chamber 4a of the lockup clutch 4 via the input port 5d and the second output port 5h. Then, the lockup clutch 4 is engaged.
As described above, the control valve 5 is operated by the balance of the load of the spring 5a, the solenoid pressure Ps input to the signal port 5c, the release hydraulic pressure Pr fed back to the port 5g, and the fastening hydraulic pressure Pa fed back to the port 5i. The differential pressure (Pa-Pr) between the engagement hydraulic pressure Pa and the release hydraulic pressure Pr can be proportionally controlled by adjusting the solenoid pressure Ps. Therefore, switching control from fastening to releasing and switching control from releasing to fastening can be performed gently with a time gradient.

The solenoid valve 6 of this embodiment is a linear solenoid valve. By controlling the indicated current to the solenoid valve 6, the solenoid pressure Ps input to the signal port 5c of the control valve 5 can be proportionally controlled. As described above, the lockup control valve 5 can proportionally control the differential pressure (Pa-Pr) by the solenoid pressure Ps input to the signal port 5c. The differential pressure (Pa-Pr) can be controlled proportionally.

The controller 7 receives vehicle operation signals such as engine speed, turbine speed, vehicle speed, throttle opening (accelerator opening), and the like, and based on these input signals and preset data and programs. The solenoid valve 6 is controlled. That is, the controller 7 calculates an instruction current to the solenoid valve 6 based on the driving state of the vehicle (for example, engine torque, engine speed, etc.) so that the differential pressure of the lockup clutch 4 becomes the target differential pressure. The command current is output to the solenoid valve 6.

FIG. 2 is an example of an ON / OFF region determination map of the lockup clutch 4 set in the controller 7.
Based on the vehicle speed and the throttle opening (accelerator opening), a lockup ON (fastening) region and an OFF (release) region are set. In this example, considering the power performance, the ON region is expanded to the low vehicle speed side at a low throttle opening, but the lockup ON / OFF region determination map is not limited to FIG.

FIG. 3 is an example of a target engagement characteristic of the lockup clutch 4 set in the controller 7, that is, an example of a time characteristic of the target differential pressure of the lockup clutch 4 at the time of engagement.
After the lock-up ON command is issued, the differential pressure is increased by a predetermined value α at time t ₁ , the differential pressure is increased at a constant gradient β at time t ₂ after a certain time from time t ₁ , and the differential rotation is almost at the time t _3, which becomes zero and a maximum differential pressure. This characteristic is the same as that of the conventional target differential pressure P0 (see FIG. 6). However, in the present invention, the target differential pressure is fixed and does not need to be corrected by individual variation. That is, the differential pressure α and the gradient β are fixed.

FIG. 4 is an IP characteristic showing the relationship between the differential pressure of the lockup clutch 4 set in the controller 7 and the command current to the solenoid valve 6. The initial value of the IP characteristic is set as a straight line having a constant slope as shown by a solid line.
As described above, since the differential pressure of the lockup clutch 4 can be proportionally controlled by the command current to the solenoid valve 6, the command current and the differential pressure can be set with IP characteristics as shown in FIG. it can. However, since this characteristic varies depending on individual vehicles, the IP characteristic is learned and corrected and updated according to the switching time of the lockup clutch 4 as will be described later.

Here, an example of learning control according to the present invention will be described with reference to FIG. This learning control is an example of control when lockup is ON.
When the learning control is started, it is determined whether or not the driving state is locked up (step S1). If the lockup is not turned on, the process ends without performing learning correction. If the lockup is turned on, the switching time (actual time) from the start to the end of the switching control from the released state to the engaged state is measured ( Step S2). The start and end can be measured by changes in differential rotation. The start of switching control when the lockup is ON is determined, for example, when the differential rotation is detected a plurality of times (for example, 3 times) that the differential rotation has decreased by a predetermined value (for example, 25 rpm) from the differential rotation in the released state of the lockup clutch. What is necessary is just to judge. Further, the end determination may be determined to be ended when, for example, it is detected a plurality of times that the differential rotation is equal to or less than a predetermined value in the vicinity of 0 (for example, 25 rpm).
Next, a deviation between the measured real time and the target time is obtained (step S3). The target time is preset by the input torque and the engine speed.
Next, the deviation obtained as described above is compared with the dead zone (step S4). If the deviation is within the dead zone, the process ends without performing learning correction. The dead zone is to prevent the control from becoming unstable by suppressing frequent learning. If the deviation exceeds the dead zone, the difference between the indicated current values is calculated by the following formula (step S5).
Current value difference = K × (real time−target time) (1)
Here, K is a proportionality constant for obtaining the current value from the time deviation.
The difference between the current values obtained by the equation (1) is added to the current value of the reference IP characteristic to update the IP characteristic (step S6).

As shown in FIG. 6, even if the reference switching time (target time) T0 at the time of engagement of the lockup clutch 4 is set, it may be as long as the time T1 due to individual variation, or the time T2 As short as possible.
When the actual time is longer than the target time T0, the difference between the current values in the equation (1) is a positive value. When the difference is added to the current value of the reference IP characteristic, the IP characteristic (A) indicated by the broken line in FIG. become that way. On the contrary, when the actual time is shorter than the target time T0, the difference between the current values in the equation (1) is a negative value, and when the difference is added to the current value of the reference IP characteristic, the IP characteristic (B) in FIG. become that way.
If the differential pressure of the lockup clutch 4 is controlled based on the IP characteristics (A) or (B) corrected as described above, switching control without individual variation can be performed.

In the deviation between the actual time and the target time of the switching time, all individual variation factors such as the variation in characteristics of the solenoid valve 6, the variation in the control valve 5, and the variation in the lockup clutch 4 itself are collected.
In the present invention, by learning and correcting the IP characteristic indicating the relationship between the command current and the differential pressure, which is the basis of switching control, using the switching time, all individual variation factors can be absorbed, and all vehicles can be absorbed. Thus, almost the same switching control can be performed.

In the above description, the reference IP characteristic is set as shown by the solid line in FIG. 4 and the gradient of this straight line is changed by learning control. However, the present invention is not limited to this.

In the above description, the IP characteristic is defined as the relationship between the command current and the differential pressure (Pa−Pr), but it can also be defined as the relationship between the command current of the solenoid valve 6 and the output pressure Ps. That is, as described above, the indicated current of the solenoid valve 6 and the output pressure Ps are in a proportional relationship, and the output pressure Ps of the solenoid valve 6 and the differential pressure (Pa-Pr) that is the output pressure of the control valve 5 are in a proportional relationship. If the relationship between the command current of the solenoid valve 6 and the output pressure Ps is known, the relationship between the command current and the differential pressure (Pa-Pr) can be uniquely determined.
However, when the output pressure Ps of the solenoid valve and the differential pressure (Pa-Pr) are not in a proportional relationship, the IP characteristic may be defined as the relationship between the command current and the differential pressure.

In the above embodiment, the switching control at the time of engaging the lockup clutch has been described, but the same applies to the switching control at the time of releasing. That is, the switching time from the start to the end at the time of lockup OFF is measured, the deviation between this switching time (actual time) and the target time is obtained, and the IP characteristic is learned and corrected using this deviation. The end of the switching control when the lockup is OFF may be determined as the end time, for example, when the differential rotation exceeds a predetermined value (for example, 50 rpm).
Of course, the IP characteristic when the lockup is ON and the IP characteristic when the lockup is OFF are set separately.
In the above embodiment, a linear solenoid valve is used as the solenoid valve, but a duty solenoid valve may be used. The duty solenoid valve can also control the output pressure in proportion to the duty ratio.

It is the schematic of the drive system of the vehicle containing the lockup clutch concerning this invention. It is a figure which shows each control area | region of a lockup clutch. It is a characteristic figure of the target differential pressure at the time of lockup ON. It is an IP characteristic figure which shows the relationship between the instruction | indication current of a solenoid valve, and differential pressure | voltage. It is a flowchart figure which shows the learning control method concerning this invention. It is a figure which shows the time change of the differential rotation at the time of lockup ON, and a target differential pressure | voltage.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Engine 2 Torque converter 3 Automatic transmission 4 Lockup clutch 4a Engagement side oil chamber 4b Release side oil chamber 5 Lockup control valve 6 Solenoid valve 7 Controller

Claims (2)

A command signal is calculated based on the torque converter having the lock-up clutch and the driving state of the vehicle, and the differential pressure of the lock-up clutch is controlled by this command signal to switch the lock-up clutch between the engaged state and the released state. In a lockup control device comprising a lockup control means,
The lockup control means includes
Time measuring means for measuring the switching time from the start to the end of the switching control;
Storage means for storing IP characteristics indicating the relationship between the command signal and the differential pressure;
Learning correction means for learning correction of the IP characteristics according to the switching time,
The lockup control device controls the differential pressure of the lockup clutch based on the learned and corrected IP characteristics. The learning correction unit sets a target value of the switching time, obtains a deviation between the switching time measured by the time measuring unit and the target value, and learns and corrects the IP characteristic using the deviation. The lockup control device according to claim 1.

Images