JPH09204201A - Control device having correction function - Google Patents

Abstract

(57) [Summary] [Problem] Simultaneously correcting the variation of control command value-system output characteristic due to measurable parameter and the variation of characteristic due to unmeasurable parameter such as individual difference,
Even in the case of strict correction, the amount of data is not increased and correction is performed easily and with high accuracy. For example, in a L / U control, a controller 9 first divides a region into sub-regions with measurable parameters.
The L / U control command value corresponding to the predetermined system output is mapped on the map, the control command value is read by the measurement parameter, the change of the control command value-system output characteristic is corrected, and the region is similarly measured by the measurable parameter. The control command value in the vicinity of the predetermined system output-gradient of change in the system output characteristic is mapped on the subdivided second map, and the change gradient is read by the measurement parameter, whereby the control command value by the parameter that cannot be measured-system. The deviation between the control command value and the system output characteristic obtained by the means for estimating the change in the output characteristic is corrected.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control device having a correction function, and more particularly to a control device useful when applied to a control system for a vehicle or the like.

[0002]

2. Description of the Related Art Various control systems are incorporated in a vehicle. For example, in an automatic transmission, a torque converter inserted in its transmission system may have a torque converter input / output element between the torque converter input / output elements under a vehicle operating condition in a lockup region where a torque increasing function and a torque fluctuation absorbing function are not required. There is a tendency to switch to a lock-up type that can be brought into a directly connected lock-up state. Further, in such a lock-up type during coasting of the vehicle, the torque converter also employs a coast L / U that puts it in a lock-up (L / U) state.

[0003]

In the control of this type of lock-up type automatic transmission, the coast L / U control is used. However, by further improving this control, the coast L / U is controlled.
The coasting L / U capacity reduction control for controlling the capacity to reduce the capacity is an effective means for avoiding engine stall (stalling) due to rapid deceleration during coasting L / U. This is to reduce deformation of the lockup clutch, the torque converter cover, etc. by reducing the L / U capacity, and improve the L / U engagement release response.

In this case, however, if the L / U capacity is reduced too much, slip rotation of the torque converter will occur and the fuel efficiency effect of the coast L / U will be reduced. Therefore, in the coast L / U capacity reduction control, the coast L / U capacity is set within a predetermined range divided by the slip occurrence area on the low capacity side and the engine stall avoidable area on the high capacity side. It is important to However, the L / U control command value-capacity characteristic is
In general, individual variations are large, and it is difficult to keep the L / U capacity during coasting within a predetermined range without any adjusting means. Further, in the case of a control system in which the characteristics when a control command value is given to control the system output may be affected by environmental conditions, variations in the environmental conditions can be absorbed and adjusted in order to improve control accuracy. It is better to do so. In the case of the L / U capacity reduction control during coasting of the automatic transmission as described above, if the temperature and other conditions during vehicle running are further taken into consideration, the influence due to this may occur. Due to the variation and the environmental condition variation, the L / U capacity generated with the same control command value may differ for each individual and for each environmental condition. Therefore, in consideration of the environmental conditions in terms of accuracy, it is difficult to keep the L / U capacity during coasting within the predetermined range as described above without adjusting means for suppressing the variations, including variations in the environmental conditions. Is.

The detection of the asymptotic approach to the low-capacity side, which is the slip occurrence region, can be easily realized by monitoring the occurrence of slip. But on the other hand,
Regarding the occurrence of engine stall, the fact that it actually occurs (actual engine stop) is inherently unacceptable even once from the aim of such control, so we will try to check whether or not actual slip occurs on the low capacity side. The situation is different from doing
It is difficult to detect the asymptotic approach to the high-capacity side, which is the engine unavoidable area. Therefore, it is difficult to realize a control method that deals with real-time F / B control or learning control on the high capacity side.

Therefore, in view of these points, it is considered that these variations are absorbed by the learning control and the environmental condition correction means on the low capacity side, and when the required accuracy is high, it is necessary to respond to this. Is more desirable to do so. In the case of the above-mentioned example of L / U capacity control, the environmental condition variation is the parameter of the change of the capacity characteristic (control command value-output characteristic) with respect to the control command value with a predetermined accuracy. The correction can be made by the measurement of the control device. That is, it is a method of subdividing an area by a map or the like, and interpolating and correcting by regarding the area between them as linear, and such means can be used for correction of environmental condition variations. On the other hand, the individual variation is basically incapable of measuring its size and parameters. Therefore, correction is performed by estimating from the measurable value using the logic for estimating the size. Since the parameters cannot be measured,
Learning control is more effective than the logical model reference estimation method.

However, when both of them are combined, the area is subdivided by a map or the like in order to perform the strict correction, and the data amount increases. For example, it is conceivable to use a subdivided map also on the side of correcting individual variations. In this case, the control command value-capacity characteristic map for each environmental condition for correcting the environmental condition variations and the individual variations are used. A map is required to subdivide the learning value for correction into environmental conditions. In this case, how to correlate each subdivided area also becomes a big problem. In addition, when the fragmented map is used for learning, there is a problem that learning opportunities for each environmental condition decrease. And to solve this problem, the logic can become more complex.

According to the present invention, in a device for giving a control command value to a system to be controlled and controlling the system output, the control command value-system output characteristic variation caused by measurable factors such as environmental condition variation. as well as,
It is intended to correct variations in control command value-system output characteristics due to factors such as individual variations that cannot be measured or are substantially unmeasurable, and to realize this easily and highly accurately. Another object is that it can be suitably used by being applied to lock-up control of an automatic transmission in a vehicle. An object of the present invention is to provide a lock-up control device capable of correction and realizing the above, and having a more improved correction function.

[0009]

According to the present invention, there is provided a control device having the following correction function. That is, in a control device that gives a control command value to a system to be controlled and controls the system output, a unit that measures a measurable first parameter other than the control command value and the system output, Means for estimating a deviation of a control command value-system output characteristic from a reference value by a second parameter which is substantially unmeasurable; and a change of the control command value-system output characteristic by the first parameter, and And a means for correcting the deviation between the control command value and the system output characteristic obtained by the estimating means, the correcting means corresponding to a predetermined system output on a map obtained by subdividing the area by the first parameter. On the first map in which the control command values are mapped and the map in which the region is subdivided by the first parameter, the vicinity of the predetermined system output is displayed. Control value-a second map in which a change gradient of the system output characteristic is mapped, and the control command value is read from the first map according to the first parameter measured by the measuring means. , While correcting the change in the control command value-system output characteristic by the first parameter, and reading the change gradient from the second map in accordance with the first parameter measured by the measuring means. Based on the correction, the deviation of the control command value-system output characteristic obtained by the estimating means is corrected to change the control command value-system output characteristic by the first parameter and the control command value by the second parameter. A control device having a correction function characterized in that it corrects both the deviation of the system output characteristic from the basic value.

Further, in the above, the product of the change gradient read from the second map and the deviation obtained by the value estimation means is added to the control command value read from the first map to obtain the final value. By obtaining a positive correction result, the control command value by the first parameter −
Both the change of the system output characteristic and the deviation of the control command value by the second parameter from the basic value of the system output characteristic are corrected. Further, the invention is characterized by being applied to a lockup control system of an automatic transmission in a vehicle. In addition, the first
Is used as a parameter of the hydraulic control valve drive voltage or an element corresponding thereto and a transmission operating oil temperature or an element corresponding thereto. Further, a learning control of a lockup capacity of an automatic transmission in a vehicle is applied to the estimating means. Further, the lock-up control of the automatic transmission during coasting of the vehicle is included, and the lock-up control can improve the lock-up engagement release response by controlling so as to reduce the lock-up capacity during coasting. The lockup capacity reduction control during coasting is performed.

[0011]

According to the present invention, when a control command value is given to the system to be controlled to control the system output, the control command value and the control command based on the measurable parameter other than the system output are used. Value-Control command value obtained by means of correcting a change in system output characteristics and simultaneously estimating a deviation of a system output characteristic from a reference value by a parameter that cannot be measured or is substantially unmeasurable-System output When correcting the deviation of the characteristic, a control command value corresponding to a predetermined system output is mapped on the first map in which the region is subdivided by the measurable parameter, and the parameter measured by the measuring means is set. By reading this control command value, the control command value by this parameter-system output The change in characteristics is corrected, and similarly, the control command value near the predetermined system output and the change gradient of the system output characteristics are mapped on the second map in which the region is subdivided by the measurable parameters. , The control command value obtained by means for estimating the change in the system output characteristic by reading the gradient of the change with the measured parameter, and thereby reading the control command value due to the non-measurable or substantially non-measurable parameter -The deviation of the system output characteristic can be corrected, and the control command value is based on the control command value and the measurable parameter other than the system output. It becomes possible to simultaneously correct the control command value and the deviation of the system output characteristic from the reference value. Therefore, control command values based on measurable parameters such as environmental conditions when using the device
It is possible to simultaneously correct both system output characteristic variations and control command values due to unmeasurable or substantially unmeasurable parameters such as individual differences among the devices-variations in system output characteristics, and at the same time Even when trying to make a strict correction using a map that has been subdivided, it is possible to suppress the increase in the amount of data required for correction and absorb the effects of these variations, and to perform simple and highly accurate corrections to control command values. If possible.

Further, in this case, it is preferable that the final correction result is obtained as in the second aspect, and the predetermined correction is performed on the first map in which the region is subdivided by the measurable parameters. Control command value corresponding to the system output is previously mapped, and the control command value read by the measured parameter is used to display the change gradient of the control command value near the predetermined system output-system output characteristic on the second map. Is preliminarily mapped to, and the change gradient read by the measured parameter and the deviation from the reference value of the control command value-system output characteristic due to the unmeasurable or substantially unmeasurable parameter are estimated. The control command value obtained by the means-the deviation of the system output characteristic It can be preferred to be carried out, similarly to realize that described above. As a result, even in this case, the control command value and the change in the system output characteristic due to the measurable parameter other than the output of the system,
It is possible to simultaneously correct the control command value due to a parameter that cannot be measured or that is substantially unmeasurable-the deviation of the system output characteristic from the reference value, and it is possible to perform a simpler and more accurate correction for the control command value.

According to a third aspect of the present invention, the above control is applied to a lockup control system of an automatic transmission,
It can be suitably implemented, and the above can be similarly realized. With this product, it is possible to correct the lockup capacity considering not only the individual variations of the applied lockup control system but also the variations in the environmental conditions when the vehicle is running, and the lockup of the automatic transmission with the improved correction function. A control device can be provided. In this case, preferably, as the measurable parameters other than the control command value and the output of the system, as described in claim 4, the hydraulic control valve drive voltage or an element corresponding thereto and the transmission operating oil temperature or As a configuration using a corresponding element, and as described in claim 5, lock-up is provided to the means for estimating the deviation from the reference value of the control command value-system output characteristic due to the parameter that cannot be measured or is substantially not measurable. The present invention can be suitably implemented as a configuration in which capacity learning control is applied, and the above can be similarly realized. In the case of the mode in which the learning control is applied, for example, a control command value-capacity characteristic map for each environmental condition for correcting the environmental condition variation such as the transmission hydraulic oil temperature, and for correcting the individual variation. There is no need for a map to subdivide the learning value for each environmental condition, and it is possible to avoid major problems such as how to correlate each subdivided area. The present invention is more effective without the disadvantage that learning opportunities for each environmental condition are reduced, which may occur when a fragmented map is used for learning.

According to the sixth aspect of the present invention, the lockup control of the automatic transmission in the vehicle is controlled so as to reduce the lockup capacity during coasting, thereby improving the lockup engagement release response. The present invention can be suitably implemented as the lockup capacity reduction control during coasting, and the above can be similarly realized. In this case, in the lockup capacity reduction control during coasting, it is possible to more effectively utilize the state that the lockup capacity width used as the control target can be limited.
The two maps are, for each lock-up capacity that is a control target in the control, a control command value map corresponding to the capacity, which is mapped by a parameter of environmental condition variation such as transmission operating oil temperature, and the vicinity of the capacity. In the mode in which a map in which the gradient change of the control command value-capacity characteristic is mapped may be used, and the lock-up capacity as the control target is limited, the correlation problem of the map disappears,
There is no reduction in learning opportunities. Further, in this case, not only the individual variation but also the environmental condition variation such as the transmission hydraulic oil temperature is suppressed so that it does not depend on the individual to be applied, and does not depend on the environmental condition, and the lockup capacity reduction control is performed. Even if it is, the lockup control system of the automatic transmission can be easily introduced into the lockup control system of the automatic transmission, and the simple and highly accurate correction of the lockup control command value during coasting can be performed. As a result, lockup capacity reduction control during coasting can be realized. Therefore, it is effective for such lock-up capacity reduction control during coasting, aiming to improve the lock-up engagement release response by reducing the lock-up capacity, and unnecessary lock due to excessive reduction of the lock-up capacity. While suppressing the occurrence of differential up and avoiding the fuel consumption effect due to lockup during coasting, the lockup capacity reduction control during coasting is effective for avoiding engine stall due to sudden deceleration during lockup during coasting. The effectiveness of can be secured, and its ability can be more fully brought out and exerted, and this can be enhanced.

[0015]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows an embodiment of the present invention, in which a control command value is given to a system to be controlled,
As a control device for controlling the system output, an example in the case of being applied to a lockup control device for an automatic transmission in a vehicle will be shown. In the figure, reference numeral 1 denotes an engine as a prime mover, and 2 denotes an automatic transmission (A / T). The automatic transmission 2 receives the power of the engine 1 through a torque converter (T / C) 3, changes the input rotation at a gear ratio according to the selected shift speed, and transmits the input rotation to the output shaft 4.

Here, in the automatic transmission 2, the shift solenoids 6 and 7 in the control valve 5 are turned on and off.
The selected shift speed is determined by the combination of F, and the torque converter 3 also performs lockup in which the input / output elements are directly connected by a lockup clutch (not shown) by the duty control of the lockup solenoid 8 in the control valve 5. (L / U) state or a converter state (T / C state) in which input / output elements are not directly connected can be set.

Incidentally, for example, the lockup solenoid 8
When the drive duty (D) is 0%, the torque converter 3 is brought into the converter state by opening the lockup clutch. For example, if the drive duty is 100%, the lockup clutch is locked under the maximum engagement force. Shall be in a state. The lockup control valve in the control valve 5 used for this control is a hydraulic control valve (control valve) that switches to the open side at a drive duty of 0% and to the engagement side at a drive duty of 100%, and has a range between them. In the intermediate capacity control based on the duty value, the engagement capacity of the lockup clutch can be arbitrarily set according to the duty. Here, the lockup control system can be configured to include the lockup solenoid 8, a lockup control valve, and a hydraulic control system for the lockup clutch.

ON / OFF of shift solenoids 6 and 7,
And the drive duty of the lockup solenoid 8 (L /
U control command value) is controlled by the A / T controller 9, and a signal from the throttle opening sensor 10 for detecting the throttle opening TVO of the engine 1 is input to the controller 9 and the transmission output is A signal from the transmission output rotation sensor 13 that detects the rotation speed No of the shaft 4 is input. Further, here, the controller 9 includes an engine rotation sensor 11 for detecting the rotation speed Ne of the engine 1.
From the turbine rotation sensor 12 that detects the input rotation speed (output rotation speed of the torque converter 3) Nt of the automatic transmission 2 and the like. Here, the input information from the throttle opening sensor 10 is applied to the shift control and can also be used to determine whether or not the vehicle is in the inertia running state.

The controller 9 has an oil temperature sensor 20 for detecting the transmission operating oil temperature (ATF oil temperature).
Is input, and further, information about the auxiliary machine load, for example, auxiliary machine load information of an air conditioner or the like is input.
The auxiliary machine load is determined by the operating state of the auxiliary machine.
It may be OFF information or current consumption.

The controller 9 includes an input detection circuit, an arithmetic processing circuit, a control program such as a shift control program and a lockup control program executed by the arithmetic processing circuit, and a storage circuit for storing the arithmetic result and the like.
Shift solenoids 6, 7 and lockup solenoids 8
And an output circuit for transmitting a control signal for driving to, and executes shift control and lockup control based on input information and internal information.

Regarding the gear shift control, the following gear shift control can be performed by well-known calculation, which is not shown here, based on the input information of the throttle opening and the transmission output shaft rotation. That is, in the shift control, the controller 9
Is the optimum gear position for the current driving state, for example, from the table data by a look-up method from the throttle opening TVO and the vehicle speed V calculated from the transmission output speed No. The shift solenoids 6 and 7 are turned on and off so that the desired shift is selected.

In the lockup control, it is determined whether the torque converter 3 is operating in the lockup region where the torque increasing function and the torque fluctuation absorbing function are not required, or in the converter region where these functions are required. In response to the control request, the drive control of the lockup solenoid 8 is performed so that the torque converter 3 is brought into the lockup state in the lockup region and is released in the converter region to bring the torque converter 3 into the converter state. To control.

For such control, the controller 9 uses, for example, lockup vehicle speed line (L / U line) data (table data) based on a lockup ON line and a lockup OFF line set in advance by the throttle opening TVO and the vehicle speed V. ) Is used to determine whether the vehicle is operating in the lockup region or the converter region defined in this way, based on the current throttle opening TVO and the vehicle speed V of the vehicle in operation. According to the torque converter 3
In the lockup region, the lockup clutch is engaged to bring the input / output elements into a directly connected lockup state, and in the converter region, the lockup clutch is released to bring the direct connection into a released state. be able to.

Furthermore, in this example, the driver (driver)
Is a so-called coasting state (coast running state) in which the accelerator pedal is released, the controller 9
Lock-up capacity reduction control for coasting that appropriately controls the lock-up state in order to avoid engine stall and improve fuel efficiency by sudden deceleration during lock-up (coast L / U) in such a state (coast L / U) In addition to performing the capacity reduction control), the control for learning the lockup clutch engagement capacity (L / U capacity) is also performed. Further, the learning is aimed mainly at correcting the solid-state variation of the applied L / U hydraulic control system, but the controller 9 is the L / U of the L / U at the coast here. In determining the control command value, in addition to the above variations, the variations in environmental conditions are also taken into consideration, and a process for effectively absorbing and compensating for these variations so that corrections for these are also made simultaneously ( 2) is also applied. In this case, the correction process is advantageous in terms of reducing the amount of data and securing learning opportunities. In order to realize a simple and highly accurate correction method for the coast L / U control command value, the following data It can be basically done by introducing a map.

One is a control command value map formed by mapping a coast L / U control command value corresponding to a predetermined capacity to be set in the coast L / U in advance in accordance with a parameter of environmental condition variations as a basic value. There is another one, which also uses the same measurable environmental condition variation parameter as the above as the search parameter for data reading, and also uses the parameter of the environmental condition variation in the vicinity of the capacity. It is a gain map in which the gain of the capacity-control command value characteristic is mapped (Fig. 3
(A), (b)). In this case, these maps are stored in advance in the memory circuit of the controller 9, and during the corresponding coast L / U capacity reduction control, the controller 9
In the calculation unit, these maps are used to simultaneously correct the control command value-capacity characteristic change caused by the environmental condition variation and the control command value-capacity characteristic deviation caused by the individual variation at the same time. At the final coast of L
The / U control command value is obtained and obtained, and an output process for giving a duty drive control signal according to the corrected command value to the L / U control system is executed.

FIG. 4 is a block diagram showing an example of functions of the embodiment system shown in FIG. 1 for such correction processing. Reference symbol a represents an L / U control system of the automatic transmission 2, and the lockup solenoid 8, the lockup control valve, and the like described above that control and set the output L / U capacity so as to achieve the control target given. Is a hydraulic control system including, and here, a correction target system in which the output characteristics with respect to the control command value may be affected by individual differences, fluctuations in the usage environment, and the like. b
Each of f to f is applied for comprehensive correction of individual variation and environmental condition variation. Learning of L / U capacity, measurement of measurable parameters, correction calculation for control command value, control command used for the calculation Each means of a value map and a learning value correction gain map is shown.

In principle, the integrated correction according to the present invention is based on the following analysis and consideration, and the above-mentioned in the case of the coast L / U capacity control in the L / U control of the automatic transmission 2 is performed. The correction function embodies this.
In this control, first, with regard to individual variation, attention was paid to the factors of individual variation. Generally, the following three factors can be cited as the factors of individual variation in the control command value of the hydraulic valve system and the output hydraulic pressure characteristic. Variation of input / output characteristics of solenoid, variation of spring constant of valve spring, variation of valve pressure receiving area

Here, the above is mainly a factor of so-called offset variation in which the control command value-output hydraulic pressure characteristic moves in parallel. In addition, the above causes a so-called gain variation in which the control command value-output hydraulic pressure characteristic is rotationally moved. At this time, the variation range of is smaller than that of, and can be almost ignored. Therefore,
Regarding individual variation, only offset variation should be targeted.

On the other hand, the variation due to the environmental conditions varies depending on the parameter, but since it is generally a combination of complicated factors, unlike the factor of individual variation, it is possible to ignore the above-mentioned gain variation. Can not. In this embodiment, the L / U capacity control during coasting is targeted. Therefore, focusing only on the correction of the L / U capacity during coasting, the L / U capacity during coasting falls within a predetermined range.
Even when trying to perform the L / U capacity lowering control so as to contain the U capacity, the control target is within the predetermined range, and therefore,
It can be said that the L / U capacity width used is extremely limited. Therefore, from this point of view, if only considering the variation in the environmental condition of the control command value that outputs a certain capacity, the control command value corresponding to the capacity is controlled from the map mapped under each environmental condition. It only needs to read the command value basic value.

If the individual variation correction is performed for this, it is simple. Further, here, when the L / U capacity learning control is adopted to correct the individual variation, the individual variation is corrected by this, but at this time, if the learning value is simply added. (By adding the learning correction value to the control command value basic value), the correction is deviated by the gain variation of the environmental condition variation. It is more preferable to correct such an error.

Therefore, in this embodiment, the following is performed. That is, the gain of the capacity-control command value characteristic in the vicinity of the specific capacity is mapped for each environmental condition, and after the environmental condition variation correction for the individual variation offset is performed by the gain read from this, the control command value is obtained. By adding it to the basic value, the gain variation of the individual variation offset environmental condition variation is corrected.

According to such a correction method, the data required in this embodiment is the control command value map corresponding to the capacity, which is mapped by the parameter of the environmental condition variation, for each L / U capacity as the control target. Only the two maps of the gain map of the capacity-control command value characteristic in the vicinity and the offset value storage memory for absorbing individual variations are used. The L / U that is the control target
If the capacity is limited, there is no map correlation problem as described above, and there is also no reduction in learning opportunities because the learning value is a single offset.

The two maps e and f in FIG. 4 are provided from the above viewpoint. Here, as the environmental condition variation parameters to be measured by the measuring means c, for example, other than the control command value and the system output, the ATF oil temperature or its equivalent, and the hydraulic pressure of the control valve 5 are used. A control valve drive voltage or its equivalent can be used. Further, the L / U capacity learning means b basically has an L / U control command value-based on a parameter that cannot be measured (or is substantially not measurable).
It is used as a means for estimating the deviation (change) of the capacity characteristic (control command value-output characteristic) from the reference value. Thus, one map e is prepared as a control command value map in which the data of the control command value corresponding to the target capacity is mapped on the map obtained by subdividing the region with such measurable parameters, and the other map is prepared. f is prepared as a learning value correction gain map in which data of L / U control command value-capacity characteristic change gradient in the vicinity of the target capacity is mapped on a map obtained by subdividing the area with similarly measurable parameters. It

The correcting means d corrects the change in the L / U control command value-capacity characteristic due to this parameter with a measurable parameter, and at the same time, the L / U control command value obtained by the L / U capacity learning means b. -When correcting the deviation of the capacity characteristic, the change of the L / U control command value-capacity characteristic is corrected by reading the above-mentioned control command value data from the control command value map e by the above measured parameter. The change gradient data is read from the learning value correction gain map f by the measurement parameter, and the deviation of the L / U control command value-capacity characteristic obtained by the L / U capacity learning means b is corrected by this.

In this case, preferably, the final correction result is obtained by the control command value read from the control command value map e, the change gradient read from the learning value correction gain map f, and the L / U capacity learning means b. By adding the product of the L / U control command value and the deviation of the capacity characteristic,
This is obtained, and L /
The change in the U control command value-capacity characteristic and the deviation from the basic value of the L / U control command value-capacity characteristic due to the unmeasurable parameter are simultaneously corrected. The L / U capacity learning means b,
The correction means d is configured by the controller 9 of FIG. 1, and the measurement means c uses the oil temperature information of the automatic transmission 2 and the power supply voltage information of the output circuit portion of the controller 9 as the environmental condition variation parameters of the measurement target. If included,
The oil temperature sensor 20 of FIG. 1 and a part of the controller 9 are included.

FIG. 2 shows the controller 9 in this embodiment.
Is executed during the coast L including the correction processing as described above.
7 is an example of a control program flowchart for setting a / U control command value. Further, FIGS. 3A and 3B exemplify a control command value map and a learning value correction gain map applicable in the correction process. In FIG.
First, in step S101, it is determined whether or not control should be performed in the complete L / U region, and in step S102, it is determined whether or not the coast state is set.

As a result of the above judgment, here, only in the case of the coast complete L / U area, and therefore, when all the answers of the judgment steps are YES, the processing is branched to the control of this example after step S104, and other When the control state is requested, other L / U in step S103
It should be branched to the process for control.

Whether or not the vehicle is in the coast state can be determined by, for example, whether or not the throttle opening TVO is less than or equal to the minute set value. Alternatively, it goes without saying that it is also possible to make a determination based on a signal from an idle switch that is turned on when the accelerator pedal is released.

Regarding the processing for other control states in step S103, if the processing is to maximize the L / U clutch engagement capacity by L / U control during normal driving, for example. In the present embodiment, the controller 9 sets the drive duty of the lockup solenoid 8 to 100% (L / U capacity maximum) as a control command value and outputs it to the lockup solenoid 8 to output the torque converter 3 to L / U. By engaging the clutch, the L / U control state as required can be achieved. Also,
In the case of the L / U release control, the drive duty is set to 0% as the control command value (L / U capacity minimum), whereby the torque converter 3 can be brought into the converter state as required by opening the L / U clutch.

On the other hand, when branching to the side of step S104, in this program example, steps S104 to S104
At 107, coast L / U capacity reduction control is performed, and processing for setting a coast L / U control command value is executed to reduce the L / U capacity to a set value. In addition,
Regarding the L / U capacity reduction control during coasting, basically, for example, target L / U clutch engagement for coasting is performed based on map data set for coasting so as to reduce the L / U capacity during coasting. The capacity is calculated, and based on this, the drive duty (D
%) Is set as a control command value and output to the lockup solenoid 8 so that the L / U clutch has a small engagement capacity. Thus, L
L / U clutch, T
The deformation of the / C cover and the like can be suppressed, and the L / U engagement release response is improved.

This is because during L / U capacity reduction control during such coasting, even when L / U release control is performed when sudden deceleration occurs due to sudden braking, such L / U capacity reduction is performed. From the state, torque converter 3 becomes L
It is effective to prevent the engine 1 from being stopped by the braked drive wheels at the time of the sudden deceleration of the vehicle when / U is released and the vehicle enters the converter state. for example,
Even if the vehicle is suddenly decelerated at the time of coast L / U, the above L / U cancellation can be quickly completed (with little response delay), and the occurrence of engine stall can be appropriately avoided. L / U at coast
The capacity reduction control can be basically performed as described above, but in this case, in this embodiment, L / U
The final coasting L / U control command value CstDty is determined in consideration of the individual variation and the environmental condition variation that also incorporate the learning result while performing the learning control of the capacity.

First, in step S104, in this program example, the target engagement capacity of the L / U clutch in the coast L / U is calculated from the information on the vehicle load and the auxiliary equipment load. Although the vehicle load is determined from the vehicle speed V and the auxiliary load is determined from the operating state of the auxiliary, a map using these as parameters may be used. Here, it is assumed that the target engagement capacity is obtained based on the target engagement capacity map, the vehicle speed V calculated as described above, and the auxiliary machine operating state status based on the auxiliary machine load information input to the controller 9. Further, when the implementation condition of the L / U capacity reduction control during coast is narrow, it may be a single value.

Next, in step S105, a control corresponding to the target engagement capacity is performed based on the target engagement capacity obtained in step S104 and environmental condition variation parameters such as ATF oil temperature, hydraulic valve drive voltage or power supply voltage. Read the basic command value by map search. Here, the ATF oil temperature and the power supply voltage are used as the environmental condition variation parameters to adjust the coast L / U corresponding to the target engagement capacity.
The control command value basic value CstDty0 is mapped in FIG.
Using the control command value basic value map of (a), the coast L / U control command value basic value CstDty0 corresponding to the target engagement capacity is calculated based on the ATF oil temperature and the power supply voltage detected at that time. read.

Further, in the next step S106, the target engagement capacity obtained in step S104 and the same environmental condition variation parameters as those used for the map search of the above-mentioned basic control command value, for example, ATF oil. The learning value correction gain is read by map search from the temperature, hydraulic valve drive voltage or power supply voltage. Here, FIG.
As shown in (b), AT is used as an environmental condition variation parameter.
Using the learning value correction gain map in which the change gradient near the target engagement capacity is mapped using the F oil temperature and the power supply voltage,
From the map, the change gradient in the vicinity of the target engagement capacity is also read based on the detected ATF oil temperature and the power supply voltage, and the learning value correction gain CstDtyG is set.

Then, in the next step S107, the learning value correction gain Cst obtained in the above step S106.
The product of DtyG and the learning value that is the result of the individual variation learning control that exists independently of the control of this example is calculated in step S1 above.
In order to add the control command value basic value CstDty0 obtained in 05,

## EQU1 ## CstDty = CstDty0 + (CstDtyG × CstPrsLrn) 1 where CstPrsLrn is the learning correction value and the result is the final coast L / U.
The control command value is CstDty. Therefore, simply L / U
The learning value on the side of the capacity learning control is directly set as the learning correction value CstPrsLrn for absorbing individual variations, and the control command value basic value C
Learning value correction gain C is not added to stDty0
The value CstPrsLrn corresponding to the learning value can be corrected by stDtyG, and the corrected value CstDtyG × CstPrsLrn can be applied to the control command value basic value CstDty0.

Thus, using the two maps of FIGS. 3A and 3B, the coast L / U control command value basic value Cst read from the map of FIG. 3A with the ATF oil temperature and the power supply voltage.
Similarly, the product of the learning value correction gain CstDtyG and the learning correction value CstPrsLrn read from the map in the same figure (b) is added to Dty0 to determine the coast L / U control command value CstDty. It is possible to set the U capacity and execute the L / U capacity lowering control during the coast.

According to the present correction method, in addition to individual variations in the L / U capacity in the L / U control system of the automatic transmission 2 having the L / U type torque converter 3, the ATF oil temperature and the environmental conditions of the power supply voltage are set. When attempting to correct the effects of variations, the learning control on the low-capacity side and the environmental condition correction processing that uses a map based on the environmental condition variation parameters should be combined in a precise manner in order to absorb those variations. If the correction method uses a map that is subdivided even on the learning control side of individual variation correction when attempting correction, if only the control command value-capacity characteristic map for each environmental condition for correcting the environmental condition variation is used. However, there is a disadvantage such as the need for a map for subdividing the learning value for correcting individual variation for each environmental condition, Not even mow problem. Therefore, it is advantageous in terms of the amount of data required for correction, and in that case there is no major problem of how to correlate each of the subdivided regions, and individual variations of the L / U control system and the ATF oil described above. The L / U capacity can be corrected in consideration of variations in environmental conditions such as temperature.

In particular, this is the L / U used as the control target.
The capacity width is limited (as described above, in the target engagement capacity calculation processing in step S104, a single value may be used if the execution conditions for the coast L / U capacity reduction control are narrow). This is a suitable and useful correction method in the case of L / U capacity reduction control during coasting. Basically, regarding the variation in the environmental conditions of the ATF oil temperature and the power supply voltage, the basic value C of the control command value corresponding to the target engagement capacity is set.
It suffices to read stDty0 from the map of FIG. 3 (a), and for individual variation correction, from the map of FIG. 3 (b), which maps under the same environmental conditions of ATF oil temperature and power supply voltage according to the above equation 1. After correcting the environmental condition variation for the learning value which is the individual variation offset by the read correction value CstDtyG, the basic value C
A correction process such as adding to stDty0 is sufficient, so that a correction deviation corresponding to the gain variation of the environmental condition variation of the ATF oil temperature and the power supply voltage, which would occur when simply adding the learning correction value, does not occur. In addition, individual variations and variations in environmental conditions such as the above ATF oil temperature can be absorbed, and therefore, a simple and highly accurate coast L / U control command value power supply voltage and oil for realizing L / U capacity reduction control during coast. It is also a temperature correction method.

Furthermore, if a subdivided map is used for learning, in addition to an increase in the amount of data, there is a problem that the learning opportunity for each environmental condition decreases, so that the learning effect is increased accordingly. If it is difficult to reflect the problem and the problem is to be solved, the logic becomes more complicated to solve the problem, but the present correction method also avoids such a problem. As a result, without inviting the situation that the learning opportunity of the L / U capacity is reduced,
Therefore, the learning accuracy is improved accordingly, and the L
Even in the case of the control to reduce the / U capacity to the required optimum value, while appropriately securing the learning opportunity for the L / U capacity in the control, the learning is possible on the low capacity side near the slip occurrence side. Can be realized by learning.

In the case where the learning control of the L / U capacity is performed by monitoring the occurrence of slip, the detection of the occurrence of slip is performed by, for example, the difference between the engine rotation and the input rotation of the automatic transmission or the ratio thereof. , Or an amount corresponding thereto can be used. Here, for example, the controller 9 monitors the occurrence of slip rotation of the torque converter 3 by using the difference between the engine rotation Ne by the engine rotation sensor 11 and the automatic transmission input rotation Nt by the turbine rotation sensor 12, which are input to the controller. And, at the time of coast L / U capacity reduction control,
When the occurrence of slip is observed, the L / U capacity can be learned by performing learning value increase control so as to correct the L / U capacity in the direction of increasing it by a predetermined amount. In this case, in the learning control, it is possible to perform the learning irrespective of the ATF oil temperature and the power supply voltage, depending on whether or not the slip occurs.
In this way, the learning result can be appropriately reflected in the L / U capacity lowering control, and the coasting L / U capacity lowering control accompanied with the L / U capacity learning control can be favorably performed.

Therefore, not only the individual variation but also the environmental condition variation such as the ATF oil temperature can be suppressed, and the L / U capacity reduction control during coast can be easily introduced into the L / U control system of the automatic transmission. Regardless of the individual to which it applies, A
Regardless of environmental conditions such as TF oil temperature and power supply voltage, and even in the case of intermediate capacity control, the effects of those are absorbed and the L / U capacity during coasting is adjusted appropriately so that the desired range is achieved. Can fit. Therefore, it is effective for the coast L / U capacity reduction control aiming at improving the L / U engagement release response by reducing the L / U capacity, and is unnecessary when the L / U capacity is excessively decreased. While suppressing the occurrence of slip rotation and avoiding the fuel consumption effect due to coast L / U, the effectiveness of such coast L / U capacity reduction control, which is effective for avoiding engine stall due to sudden deceleration during coast L / U, is also effective. It is possible to secure it, bring out its ability more fully, and bring out it to the full, thereby enhancing it.

The present invention is not limited to the above embodiment. For example, in the above, the automatic transmission L
However, the present invention is not limited to this, and in the case of giving a control command value to the system to be controlled and controlling the system output, the control command value and the measurable parameter other than the system output The present invention can be applied to the case where it is attempted to correct the change in the control command value-system output characteristic due to the parameter and at the same time to correct the deviation from the reference value of the control command value-system output characteristic due to the parameter that cannot be measured.

[Brief description of drawings]

FIG. 1 is a system diagram showing an example of a lockup control device for an automatic transmission according to an embodiment of a control device of the present invention.

FIG. 2 is a diagram showing an example of a program flow chart for explaining a correction process for a lockup control command value performed by a controller in the same example.

FIG. 3 is a diagram showing each example of a control command value map and a learning value correction gain map applicable to the same example.

FIG. 4 is a functional block diagram showing an example of correction of a control command value.

[Explanation of symbols]

 1 Engine 2 Automatic Transmission (A / T) 3 Torque Converter 5 Control Valve 6 Shift Solenoid 7 Shift Solenoid 8 Lockup (L / U) Solenoid 9 Controller 10 Throttle Opening Sensor 11 Engine Rotation Sensor 12 Turbine Rotation Sensor 13 Transmission Output rotation sensor 20 Oil temperature sensor

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location G05B 19/02 G05B 19/02 C

Claims (6)

[Claims] 1. A control device for giving a control command value to a system to be controlled and controlling the system output thereof, means for measuring a measurable first parameter other than the control command value and the system output; A means for estimating a deviation of a control command value-system output characteristic from a reference value by a second parameter that is impossible or substantially unmeasurable, and a change of the control command value-system output characteristic by the first parameter is corrected. And a means for correcting the deviation between the control command value and the system output characteristic obtained by the estimating means, the correcting means comprising a predetermined system on a map obtained by subdividing the region by the first parameter. The predetermined system is provided on a first map in which a control command value corresponding to the output is mapped and a map in which an area is subdivided by the first parameter. A control command value in the vicinity of the force-a second map in which a change gradient of the system output characteristic is mapped, and the control command value is calculated from the first map in accordance with the first parameter measured by the measuring means. By reading, the change of the control command value-system output characteristic by the first parameter is corrected, and the change gradient is read from the second map according to the first parameter measured by the measuring means. Based on this, by correcting the deviation of the control command value-system output characteristic obtained by the estimating means, the change of the control command value-system output characteristic by the first parameter and the change of the second parameter by the second parameter. A control device having a correction function characterized in that the control command value and the deviation of the system output characteristic from the basic value are both corrected. 2. A final correction is made by adding the product of the change gradient read from the second map and the deviation obtained by the value estimating means to the control command value read from the first map. By obtaining the result, both the change of the control command value-system output characteristic by the first parameter and the deviation of the control command value-system output characteristic by the second parameter from the basic value are corrected. A control device having a correction function according to claim 1. 3. A control device having a correction function according to claim 1, which is applied to a lockup control system of an automatic transmission in a vehicle. 4. The correction function according to claim 3, wherein as the first parameter, a hydraulic control valve drive voltage or an element corresponding thereto and a transmission operating oil temperature or an element corresponding thereto are used. Control device having. 5. The control device having a correction function according to claim 3, wherein learning control of a lockup capacity of an automatic transmission in a vehicle is applied to the estimating means. 6. A lock-up control of an automatic transmission during coasting of a vehicle, the lock-up control controlling the lock-up capacity during coasting to reduce a lock-up engagement release response. The control device having a correction function according to claim 5, wherein the control is improved lockup capacity reduction control during coasting.