JP3505771B2 - Control device for automatic transmission - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control device for an automatic transmission.

[0002]

2. Description of the Related Art Generally, an automatic transmission for an automobile is provided with a torque converter and a mechanical transmission mechanism. The torque converter decelerates the output torque of an engine and outputs it to a turbine shaft. The torque of is further decelerated or accelerated and output to the output shaft. Here, the speed change mechanism is usually
It is a planetary gear mechanism consisting of multiple gears such as sun gear, ring gear, pinion gear, etc., and by changing the power transmission path inside it, the shift stage or range can be switched (hereinafter referred to as shift). There is.

In order to switch the power transmission path of the speed change mechanism, the automatic transmission is provided with a clutch for turning on / off (interrupting) transmission of torque to a predetermined gear and braking for a predetermined gear. Various hydraulic friction elements such as brakes that are turned on (fixed) and turned off (released) are provided. Further, the automatic transmission is provided with a hydraulic mechanism that supplies and discharges working hydraulic pressure to and from each of these friction elements, and this hydraulic mechanism switches the on / off pattern of each friction element so that gear shifting is performed. Has become.

In such an automatic transmission, the fastening force of each friction element, that is, the force in the direction perpendicular to the friction engagement surface, is substantially proportional to the operating oil pressure or line pressure (original pressure) of the hydraulic mechanism.
At the time of gear shifting, the engaging force of each friction element, that is, the line pressure of the hydraulic mechanism must be appropriate according to the rotational force applied to the friction element. If the line pressure is higher than necessary, power loss will occur. And a problem such as a strong shift shock is generated. On the other hand, if the line pressure is too low, the time required for gear shifting becomes long and the running performance is degraded. In addition, it is a matter of course that the line pressure must be appropriate according to the transmission torque even in normal time (other than shifting).
If the line pressure is too high, power loss increases, and if the line pressure is too low, slippage occurs, causing abnormal wear or abnormal heat generation in the friction engagement portion.

Therefore, in the automatic transmission, it is preferable to control the line pressure according to the input torque to the transmission mechanism or the engine torque. However, it is difficult to directly detect the torque of the turbine shaft. Therefore, in conventional automatic transmissions, some physical quantity that reflects the shift state, such as the time required for shifting, is usually used.
The line pressure is controlled based on (shift time) and the like.

That is, if the line pressure is too low with respect to the input torque to the transmission mechanism, the shift time becomes longer than the normal value, and if the line pressure is too high, the shift time becomes shorter. Therefore, it is possible to determine whether or not the line pressure is appropriate based on the shift time. Therefore, many conventional automatic transmissions perform learning control (learning correction) of the line pressure, for example, setting the target line pressure based on the difference between the target value and the actual value of the shift time.

By the way, at the time of gear shifting of the automatic transmission, a predetermined number of friction elements or a plurality of friction elements are turned on / off, but if the on / off timings of these friction elements are not appropriate, the engine speed will change. A shift shock occurs due to a phenomenon such as an abnormal rise, that is, a so-called upstroke, or a phenomenon that the engine speed drops, a so-called pull-in.

In this case, there is a close relationship between the engagement timing of the friction element and the shift shock, and the engagement timing of the friction element is changed by changing the hydraulic fluid supply / discharge characteristics such as the hydraulic fluid throttle amount. You can For this reason, in the conventional automatic transmission, by adjusting the supply / discharge characteristics of the hydraulic oil, for example, the throttle amount of the hydraulic oil, the engagement timing of the friction element is held at an appropriate value to prevent the occurrence of a shift shock. Techniques are often used. Also in this case, in many cases, learning control of the aperture amount is performed such that the target aperture amount is set based on the difference between the target value and the actual value of the engagement timing.

Further, an automatic transmission has been widely used in which, when shifting, the ignition timing is retarded to reduce the torque to reduce the input torque to the shifting mechanism to prevent a shift shock. There is. In many automatic transmissions, learning control of the torque down amount is performed such that the target torque down amount is set based on the difference between the target value and the actual value such as the shift time.

[0010]

However, for example, in an automatic transmission in which the learning control of the torque down amount is performed, the torque down may be prohibited due to the circumstances on the engine side. For example, when the engine is cold, torque reduction may be prohibited in order to improve the ignitability and combustibility of the air-fuel mixture. When the learning control of the torque reduction amount is performed when the torque reduction is prohibited in this way, the target torque reduction amount is set on the premise that the torque reduction is performed, so that erroneous learning occurs. The accuracy of the shift control deteriorates.

In order to improve this, Japanese Patent Laid-Open No. 5775/1990
Japanese Patent Laid-Open No. 9 discloses an automatic transmission in which the learning control of the torque reduction amount is performed by retarding the ignition timing at the time of shifting, while the learning control is prohibited when the torque reduction is prohibited. Then, in this conventional automatic transmission, it is determined whether or not the learning control is prohibited only by whether or not the torque down is activated. Thus, in this case, even if torque down is performed,
When the amount of torque reduction is guarded by the occurrence of misfire or the generation of unburned gas, the learning control is performed because the torque reduction execution command is output. In this case, the learning control is executed assuming that the torque reduction is completely performed even though the amount of torque reduction is not normal due to the guard, and erroneous learning occurs, resulting in the accuracy of the shift control. There is a problem that it becomes worse.

In an automatic transmission, the target line pressure may be guarded by a drift-on ball guard or the like (for example, Japanese Patent Laid-Open No. 4-165158).
However, in the case where the learning control of the line pressure is performed, there is a problem that erroneous learning occurs even when the line pressure is guarded in this way and the accuracy of the shift control deteriorates.

The present invention has been made in order to solve the above-mentioned conventional problems. The learning control can be performed as widely as possible, and even when the control amount to be learning-controlled is guarded. An object of the present invention is to provide a control device for an automatic transmission that can effectively prevent erroneous learning.

[0014]

To reach the above object, according to the solution to ## to indicate the configuration in FIG. 1, a first invention, ene
Jin A and automatic transmission B are integratedly controlled.
To reduce the engine output torque during shifting.
Engine torque down control means C capable of controlling the engine down,
Work of learning controlling the operation pressure of the friction element D based on a comparison between the target value and the actual value of the physical quantity that reflects the shifting state during the shift
In the control device for an automatic transmission provided with the dynamic pressure learning control means E , the operating pressure learning control means E is
When torque down control of the torque down control means C is possible
The target value of the above physical quantity differs depending on whether it is impossible or not.
The operating pressure of the rubbing element D is learned and controlled, and the torque dough is controlled.
Engine torque reduction amount when engine control is possible
When the limit is applied, the learning system for the operating pressure of the friction element D
Provided is a control device for an automatic transmission, which is characterized in that the control is stopped .

A second aspect of the present invention is the automatic transmission control device according to the first aspect, wherein the operating pressure learning control means E when the operating pressure of the friction element D is limited.
A control device for an automatic transmission, characterized in that learning control of the operating pressure is stopped .

A third aspect of the present invention is the control device for an automatic transmission according to the second aspect , wherein the operating pressure of the friction element D is a line.
Provided is a control device for an automatic transmission characterized by being pressure .

A fourth aspect of the present invention is the control device for an automatic transmission according to the second aspect , wherein the operating pressure of the friction element D is a clutch.
Provided is a control device for an automatic transmission, which is characterized in that the pressure is Q pressure.

The fifth invention is any of the first to fourth inventions .
Or the control device for an automatic transmission according to one, the speed change shape
Provided is a control device for an automatic transmission, wherein the physical quantity that reflects the state is a shift time .

A sixth invention is any one of the first to fifth inventions .
Or the control device for an automatic transmission according to one, the speed change shape
Provided is a control device for an automatic transmission, wherein a physical quantity that reflects the state is a change in input rotation of a transmission mechanism .

[0020]

[0021]

[0022]

[0023]

EXAMPLES Examples of the present invention will be specifically described below.
As shown in FIG. 2, the vehicle 1 has left and right front wheels 2a and 2b as driving wheels, and the output torque of the engine 3 is from the automatic transmission 4 to the differential device 5 and the left and right drive shafts 6a and 6a.
It is adapted to be transmitted to the front wheels 2a, 2b via b.
The engine 3 is provided with spark plugs 7 ... 7 for each cylinder.

The automatic transmission 4 includes a torque converter 20 connected to an output shaft 8 of the engine 3, as shown in FIG.
A transmission mechanism 30 to which the output torque (turbine torque) is input, a plurality of friction elements 41 to 46 such as clutches and brakes for switching the power transmission path of the transmission mechanism 30, and one-way clutches 51, 52; 41-4
The transmission mechanism 3 by selectively supplying the line pressure to 6
Hydraulic control unit 60 for switching the gear ratio (shift stage) of 0
, And D, S, L, and
Each range of R, 1 to 4 speeds in the D range, 1 to 3 speeds in the S range, and 1 to 2 speeds in the L range can be obtained.

The torque converter 20 includes a pump 22 fixed in a case 21 connected to the engine output shaft 8.
A turbine 23 disposed opposite to the pump 22 and driven by the pump 22 via hydraulic oil; and a one-way clutch 24 provided between the pump 22 and the turbine 23 in the transmission case 9. A stator 25 that is supported via the case 21 to increase the torque, and a lockup clutch 26 that is provided between the case 21 and the turbine 23 and that directly connects the engine output shaft 8 and the turbine 23 via the case 21. It is composed of. The rotation of the turbine 23 is output to the transmission mechanism 30 side via the turbine shaft 27. Here, a pump shaft 10 penetrating the inside of the turbine shaft 27 is connected to the engine output shaft 8, and the oil pump 11 provided at the rear end portion of the transmission is driven by the shaft 10.

The speed change mechanism 30 is composed of a Ravigneaux type planetary gear device, and has a small diameter small sun gear 31 loosely fitted on the turbine shaft 27 and the sun gear 31.
Similarly, a large sun gear 32 having a large diameter that is loosely fitted on the turbine shaft 27 and a plurality of short pinion gears 3 meshed with the small sun gear 31.
3, a long pinion gear 34 whose front half is meshed with the short pinion gear 33 and whose rear half is meshed with the large sun gear 32, a carrier 35 which rotatably supports the long pinion gear 34 and the short pinion gear 33, and a long The ring gear 36 meshes with the pinion gear 34.

A forward clutch 41 and a first one-way clutch 51 are provided in series between the turbine shaft 27 and the small sun gear 31, and a coast clutch 42 is provided in parallel with these clutches 41, 51. Installed, turbine shaft 27
A 3-4 clutch 43 is interposed between the carrier 35 and the carrier 35, and further, the turbine shaft 27 and the large sun gear 3 are provided.
A reverse clutch 44 is interposed between the two. A 2-4 brake 45, which is a band brake that fixes the large sun gear 32, is provided between the large sun gear 32 and the reverse clutch 44.
The second one-way clutch 5 that receives the reaction force of the carrier 35 is provided between the carrier 35 and the transmission case 9.
2 and a low reverse brake 4 for fixing the carrier 35
6 and 6 are provided in parallel. The ring gear 36 is connected to the output gear 12, and the rotation is transmitted from the output gear 12 to the left and right front wheels 2a, 2b via the differential device 5.

Table 1 shows the relationship between the operating states of the friction elements 41 to 46 such as the clutches and brakes and the one-way clutches 51 and 52 and the shift speeds.

[Table 1]

Further, a control unit (hereinafter referred to as ECU) for integrally controlling the engine 3 and the automatic transmission 4
The ECU 70 includes a signal from a vehicle speed sensor 71 that detects the vehicle speed of the automobile 1, a signal from a throttle sensor 72 that detects the opening degree of a throttle valve of the engine 3, and an air flow that detects the intake flow rate of the engine 3. Detects a signal from the sensor 73, a signal from an engine rotation sensor 74 that detects the engine speed, a signal from a water temperature sensor 75 that detects the cooling water temperature of the engine 3, and an output speed (turbine speed) of the torque converter 20. Signal from the turbine rotation sensor 76, a signal from the output rotation sensor 77 that detects the output rotation speed of the transmission mechanism 30, a signal from the shift position sensor 78 that detects the shift position (range) by the select lever 13, an automatic transmission The signal from the oil temperature sensor 79 for detecting the hydraulic oil temperature of No. 4 is input to the automatic transmission 4. The shift control is performed by the shift solenoid valves 61 ... 61 provided in the hydraulic control unit 60 and the line pressure control is performed by the duty solenoid valve 62 also provided in the hydraulic control unit 60. Controls the ignition plugs 7 ... further,
In the present embodiment, control is performed to reduce the output torque of the engine 3 by ignition control during gear shifting. In addition, E
The CU70 is the engine engine described in the claims .
It is a comprehensive control means including a "look-down control means" and an "operating pressure learning control means" .

Here, the structure of the line pressure control portion of the hydraulic control unit 60 will be described. As shown in FIG. 4, a regulator valve 63 for adjusting the pressure of the hydraulic oil discharged from the oil pump 11 to a predetermined line pressure,
A throttle modulator valve 64 for supplying a control pressure to the regulator valve 63 is provided. The throttle pump valve 64 has an oil pump 11
A constant pressure line 67 is connected from a main line 65 through which the hydraulic oil is discharged from the main line 65 through a reducing valve 66 that reduces the hydraulic oil to a constant pressure, and
From the modulator valve 64 to the regulator valve 63
To the pressure boosting port 63a provided at one end of the pressure boosting line 68
Has been led. The constant pressure line 6 is connected to the control port 64a at one end of the throttle modulator valve 64.
A pilot line 69 branched from 7 is connected.

The duty solenoid valve 62 for controlling the line pressure shown in FIG. 2 is installed in the pilot line 69, and the duty solenoid valve 6 is installed.
By introducing the pilot pressure according to the duty ratio of 2 into the control port 64a of the throttle modulator valve 64, the constant pressure supplied from the line is
After being adjusted to the pilot pressure or a pressure according to the duty ratio, it is supplied to the pressure increasing port 63a of the regulator valve 63 through the pressure increasing line 68. Therefore, the line pressure adjusted by the regulator valve 63 becomes a pressure according to the duty ratio.

Next, the line pressure control at the time of the schedule up shift which is performed with the increase of the vehicle speed will be described with reference to the flowchart shown in FIG. That is, the ECU
After reading various signals in step # 1, the step 70 determines a shift in the next step # 2 and outputs a shift-up shift determination flag SFTUP. The shift-up shift determination flag SFTUP is set to 1 when a shift command is output.
This is a flag that is reset and reset to 0 when the upshift is completed.

Next, at step # 3, it is determined whether or not the present shift-up shift determination flag SFTUP [i] exceeds 0. If SFTUP [i]> 0 (YES), that is, the current shift is made. If it is medium, step # 3 to step # 2
The shift up routine of No. 5 is executed. On the other hand, SFTU
If P [i] ≦ 0 (NO), that is, if the upshift is not being performed, the non-upshift routine of steps # 26 to # 29 is executed.

When executing the shift-up routine,
First, in step # 4, the rotation change amount ΔNt of the turbine rotational speed Nt before and after the shift is calculated according to the following equation 1, and in step # 5, the turbine torque Tt is calculated according to equation 2.

[Equation 1]       ΔNt = Nts-Nos.Go ……………………………………………… Equation 1

[Equation 2] Tt = (Nts / Nes) · Te · t …………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………… Nts The output rotation speed of the speed change mechanism 30, Go is the gear ratio after completion of the speed change, Te is the engine torque, and t is the torque converter 20.
The torque increase coefficient of is shown. The engine torque Te
Is calculated based on, for example, the engine speed, the intake flow rate, the ignition timing, and the like.

Next, the ECU 70 proceeds to step # 6 to determine whether or not the torque-down enable flag Ftd is set to 1. In addition, torque reduction possible flag Ftd
Is set to 1 when the cooling water temperature indicated by the signal from the water temperature sensor 75 indicates the warm-up state of the engine 3, for example. When the ECU 70 determines that the torque-down enable flag Ftd is set to 1, the process proceeds to step # 7, and the turbine torque Tt, the rotation change amount ΔNt, and the target shift stage Lm are set as parameters in advance. After calculating the target shift time Ts according to the target shift time map for use, the target angular acceleration Am is calculated according to the following equation 3 in step # 8.

[Expression 3] Am = | ΔNt / Ts | ………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………… The absolute value is the target angular acceleration Am.

Next, the ECU 70 executes step # 9 to execute the actual turbine torque T according to the map in which the turbine torque and the angular acceleration are set in advance as parameters.
After the gear shift target input torque Tm corresponding to t and the target angular acceleration Am is obtained, step # 13 is executed. On the other hand,
When the ECU 70 determines in step # 6 that the torque reduction possible flag Ftd is not set to 1, that is, when it is determined that torque reduction is impossible, the ECU 70 proceeds to step # 10 and sets the turbine torque Tt in advance. The target shift time Ts is calculated in accordance with the non-torque down target shift time map set by using the rotation change amount ΔNt and the target shift stage Lm as parameters, and then the rotation change amount ΔNt and the target shift time Ts are calculated in step # 11. The target angular acceleration Am is calculated by substituting into the above equation (3). In that case, the non-torque down target shift time map is set so that the target shift time is longer than the torque down target shift time map. Then, ECU 70 executes step # 12, Tar
The bin torque Tt is set as the target torque Tm during shifting. Therefore, in this case, the torque reduction of the engine 3 is not performed. Then, step # 13 is executed.

At step # 13, the target input torque Tm
Exceeds the input torque guard value TNX, and if Tm> TNX (YES), the target input torque Tm is set as the input torque output value TN in step # 15. On the other hand, if Tm ≦ TNX (NO), in step # 14, the input torque guard value TNX is set to the input torque output value TN, and the learning permission flag XPLRN is reset to 0. The learning flag XPLRN is a flag that allows learning control when set to 1 and prohibits learning control when reset to 0. Therefore, when the input torque output value TN is guarded (TNX), learning control is prohibited, erroneous learning does not occur, and learning control accuracy is improved.

In this embodiment, the torque reduction is basically performed by retarding the ignition timing at the time of gear shifting. If the intake air amount and the engine speed are constant, the change characteristic of the output torque of the engine with respect to the ignition timing (advance amount) is as shown by the curve G ₁ in FIG. 9. Therefore, as is clear from FIG. 9, as the ignition timing is retarded, the output torque becomes smaller and the torque can be effectively reduced. However, if the ignition timing is retarded too much, misfire occurs, or The temperature of the catalytic converter rises abnormally. Therefore, in this embodiment, the output torque of the engine is not reduced below the straight line G ₂ to prevent the ignition timing from being retarded beyond the allowable limit by the learning control. That is, the torque or the amount of torque reduction is guarded corresponding to the straight line G ₂ . And this straight line G
_The torque corresponding to ₂ is the input torque guard value TNX.

Next, at step # 16, although not shown, the input torque output value TN and the gear Lm are set based on the input torque hydraulic pressure setting map which is set in advance for each gear using the input torque as a parameter. The input torque hydraulic pressure Pt corresponding to is set. Here, the input torque hydraulic pressure setting map is set so that the input torque hydraulic pressure Pt increases as the input torque output value TN increases. Also,
The ECU 70 executes step # 17, and based on an inertia torque hydraulic pressure setting map (not shown) set for each shift speed using the angular acceleration as a parameter, the inertia torque corresponding to the target angular acceleration Am and the shift speed Lm is set. Set the hydraulic pressure Pi. Also in this case, the inertia torque hydraulic pressure setting map is set such that the inertia torque hydraulic pressure Pi increases as the target angular acceleration Am increases.

Then, the ECU 70 executes step # 18, and from the learning hydraulic pressure table for shift-up, which is set for each shift stage (not shown), the target shift stage Lm of this time is set.
After reading the learning hydraulic pressure Pg corresponding to,
19 is executed and the input torque hydraulic pressure P is calculated according to the following equation 4.
A target line pressure Pcl (target clutch pressure) is calculated from t, the inertia torque Pi and the learned oil pressure Pg.

[Equation 4]       Pc1 = Pt + Pi + Pg …………………………………………………… Equation 4

Next, at step # 20, the target line pressure Pcl
Is greater than the line pressure guard value PLX. If Pcl> PLX (YES), the target line pressure Pcl is set to the line pressure output value Pl in step # 22. On the other hand, Pc
If l ≦ PLX (NO), the line pressure guard value PLX is set to the line pressure output value Pl in step # 21, and the learning permission flag XPLRN is reset to 0. Learning flag X
As described above, when the PLRN is set to 1, learning control is permitted, and when reset to 0, learning control is prohibited.

In this embodiment, two kinds of guards are applied to the line pressure, which is a controlled variable. One is a guard for securing the static capacity limit of hydraulic oil. The guard due to this static capacity limit is, for example, the curve G _{3 in} FIG.
It is a guard with the characteristics shown in. In addition, this guard is a guard that does not reduce the line pressure below the curve G ₃ . The other is a guard for ensuring the minimum required lubrication flow rate. The guard based on this lubricating flow rate
For example, it is a guard having the characteristics shown by the curve G ₄ in FIG. The guard is a guard that does not lower the line pressure below the curve G ₄ .

In this case, since the guard by the static capacity limit and the guard by the lubricating flow rate are overlapped, the curve G ₃ in FIG. 10 and the curve G _{4 in} FIG. When one of the guards is applied, the learning permission flag XPLRN is reset to 0. Thus, when the line pressure is actually guarded, the learning control of the line pressure is prohibited, and erroneous learning does not occur, so that the accuracy of the learning control of the line pressure is improved. Although not shown, a drift-on-ball guard may be applied to the line pressure.

Next, the ECU 70 proceeds to step # 23 to calculate the final target line pressure Ple by correcting the oil temperature to the line pressure output value Pl. That is, in general, the friction element is fastened by frictional contact between the friction members, but the friction coefficient μ of the contact surface or the sliding surface of the friction members facing each other is determined by the hydraulic oil temperature To between the friction members. Depends on. Specifically, the friction coefficient μ increases as the hydraulic oil temperature To decreases. Therefore, although not shown, the oil temperature correction coefficient Kμ corresponding to the current hydraulic oil temperature To is read from the table of the oil temperature correction coefficient in which the hydraulic oil temperature is set as a parameter, and the correction coefficient K is then read.
The final target line pressure Ple is obtained by substituting μ and the target engagement pressure Pc1 into the following equation 5.

[Equation 5] Ple = Pcl · kμ …………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………… The input torque output value TN obtained in step # 14 or step # 15 is output.
Then, the line pressure learning value is calculated in step # 25, and then the process returns to step # 1. The line pressure learning value calculation is as described later according to the flowchart shown in FIG.

By the way, in the step # 3, the SFTU
When it is determined that P [i] ≦ 0 (NO), the non-shift up routine is executed. First, in step # 26, it is determined whether or not the previous SFTUP [i-1] exceeds 0, and if it exceeds 0, it means that the shift-up state has started from this time. Therefore, in step # 27, XPLRN> Only when it is determined that the learning permission flag X is updated in step # 28, the line pressure learning value is updated.
Set PLRN to 1 and return to step # 1. That is, the line pressure learning value is updated every time the upshift is completed. When NO is determined in step # 26 or step # 27, the line pressure learning value is not updated and step # 29 is executed and then the process returns to step # 1.

Therefore, in the automatic transmission 4, the duty solenoid valve 62 is duty-controlled so that the final target line pressure P1e is obtained, and in the engine 3, a predetermined input torque Tm is realized. The engine output is torque-down controlled according to the program. As described above, when the torque can be reduced, the target torque Tm during shifting of the engine 3 is set based on the target angular acceleration Am and the turbine torque Tt, and the input torque hydraulic pressure Pt corresponding to the input torque of the transmission mechanism 30 is set. Since it is set based on the target input torque Tm, the input torque hydraulic pressure Pt accurately corresponds to the actual input torque to the speed change mechanism 30 at the time of shifting.

The ECU 70 executes the learning process of the inertia torque hydraulic pressure Pi in parallel with the line pressure control at the time of the above-mentioned schedule up shift according to the flow chart shown in FIG. That is, E
The CU 70 reads various signals in step # 31 and then determines in step # 32 whether or not the learning flag Fg is set to 1. Here, the learning flag Fg is set to 1 at the time of shift-up shifting in the normal drive state in which the automobile 1 is driven by the engine output.

Then, the ECU 70 sets the learning flag Fg to 1
If it is determined that it is not set to step #
33 is executed to set the timer value of the shift time timer to 0, and when the value of the learning flag Fg is set to 1, the routine proceeds to step # 34, where it is determined from the turbine speed Nt and the output speed No. It is determined whether the current gear ratio Gr of the transmission mechanism 30 is smaller than a predetermined shift start determination value Gs. That is, it is determined whether or not the shift operation has actually started.

When it is determined in step # 34 that the gear ratio Gr is smaller than the gear shift start determination value Gs, the ECU 70 proceeds to step # 35 and increments the timer value of the gear shift time timer by 1 and at step # 36. Then, it is determined whether or not the gear ratio Gr is smaller than a predetermined shift end determination value Ge, and the loop processing of steps # 35 and # 36 is executed until YES is determined. When it is determined that the gear ratio Gr is smaller than the shift end determination value Ge, the process exits from the loop process and proceeds to step # 37, where the actual shift time T and the rotation change amount ΔNt indicated by the timer value of the shift time timer are indicated. From the above, the actual average angular acceleration Ar is calculated according to the following equation 6.

## EQU6 ## Ar = | ΔNt / T | …………………………………………………… Step 6 and step # 38 is executed to calculate the corrected hydraulic pressure ΔPi. In other words, the ECU 70 applies the average angular acceleration Ar and the target angular acceleration Am to a shift-up inertia torque hydraulic table for line pressure control (not shown) to actually measure the inertia torque hydraulic Pim corresponding to the target angular acceleration Am. The difference amount of the inertia torque oil pressure Pir corresponding to the average angular acceleration Ar of is the correction oil pressure ΔPi. In this case, when the actual average angular acceleration Ar is larger than the target angular acceleration Am, the value of the correction hydraulic pressure ΔPi becomes negative.

Next, the ECU 70 executes step # 39 and substitutes the correction hydraulic pressure ΔPi obtained in step # 38 into the following equation 7 to calculate the learning correction hydraulic pressure ΔPg.

[Formula 7] ΔPg = ΔPi · Ko / Kμ …………………………………………………………………………………………………………………………………………………………………………………………………………………… Shows the learning reflection coefficient. And the ECU 7
In step 0, step # 40 is executed to update the learning hydraulic pressure Pg with the current value obtained by adding the learning correction hydraulic pressure ΔPg to the previous value of the learning hydraulic pressure Pg.

According to this structure, the following effects can be obtained. As shown in FIG. 8, assuming that the 1-2 shift is performed at the time t ₁ in the state where the torque can be reduced, as shown by an arrow a, the learning hydraulic pressure Pg is calculated based on the shift-up learning hydraulic pressure table. Is set, and the learning hydraulic pressure Pg, the input torque hydraulic pressure Pt set based on the input torque output value TN, and the inertia torque hydraulic pressure Pi obtained from the inertia torque hydraulic pressure setting map are indicated by an arrow b. Is set to the line pressure output value Pl. Then, the learning hydraulic pressure Pg is updated when the gear ratio Gr converges to the gear change end determination gear ratio Ge. In that case,
As shown in the figure, when the actual shift time T indicated by the shift time timer is shorter than the target shift time Ts, the actual average angular acceleration Ar becomes larger than the target angular acceleration Am. The value becomes negative, and accordingly, the learning correction hydraulic pressure ΔPg also becomes negative. Therefore, the current value Pg 'of the learning hydraulic pressure is a value obtained by adding the learning correction hydraulic pressure ΔPg (negative value) to the previous value ΔPg as shown by the arrow c. Note that, as described above, when the learning correction hydraulic pressure ΔPg is negative, the value is actually a subtracted value. In this case,
Learning hydraulic Pg corresponding to the second speed is to be updated in the sheet Futoappu learning Hydraulic table.

Then, the line pressure output value Pl in the next 1-2 shift is reduced by the learning correction hydraulic pressure ΔPg as shown by the two-dot chain line in the figure. In that case, the target shift time is set based on the amount of change in the turbine speed before and after the shift and the input torque to the transmission mechanism 30, and based on the angular acceleration and the input torque for realizing the target shift time. Since the target line pressure at the time of shifting is set in accordance with the above, it is possible to obtain the line pressure at shifting corresponding to the operating condition. Moreover, since the inertia torque of the inertia torque corresponding to the inertia torque caused by the rotation change of the speed change mechanism 30 during the gear shift is learned and corrected,
It is possible to obtain the line pressure during shifting that appropriately corresponds to the change in the driving environment.

In particular, since the average angular acceleration at the time of shifting is adopted as the learning parameter, good learning accuracy can be obtained, and the line pressure at shifting can be set with higher accuracy. Moreover, by adopting the angular acceleration of the turbine speed as the learning parameter, it is only necessary to store the learned hydraulic pressure for each gear stage.
The memory capacity of the ECU 70 is saved. In particular, in the case where the torque down control is performed as in the embodiment, the learning process can always be performed regardless of whether the torque down control is executed, so that the line pressure during shifting that more appropriately corresponds to the operating state can be obtained. Will be obtained.

The prior art disclosed in Japanese Patent Laid-Open No. 2-57759 (hereinafter, simply referred to as prior art)
The difference between the present invention and the present invention is as follows. In other words, the conventional technique and the present invention are completely different in the method of setting the hydraulic pressure and the method of controlling the torque reduction during shifting.
In the prior art, the setting of the hydraulic pressure and the torque reduction are based on a completely general idea. The hydraulic pressure is set according to the throttle opening, and the torque reduction is delayed by a predetermined amount. In this conventional technique, when the shift point is uniquely determined with respect to the throttle opening, the hydraulic pressure and torque are reduced according to the input torque and the inertia in that state, but the shift point or the environment changes. When the engine torque changes, it is difficult to perform control corresponding to this.

On the other hand, in the present invention, the target time is set by the input torque (Q / N) and the turbine rotation change amount, and the target input torque is obtained from the angular velocity of the input shaft and the input torque within this time, and the actual So that the torque of becomes the target torque,
Controls engine torque (usually torque down), and adjusts hydraulic pressure according to input torque and angular acceleration of the input shaft.
Set corresponding to the hydraulic pressure corresponding to (inertia). For this reason, it is possible to perform the gear shift at a desired target time regardless of the shift point and the fluctuation of the engine torque. Note that the learning control is adapted to correct changes over time, unit variations, and the like. Although the above embodiment exemplifies the learning control of the line pressure, the clutch pressure learning control in which the clutch operating pressure is controlled by a combination of a duty solenoid or a regulating valve such as a duty solenoid, and the orifice of the clutch operating circuit is bypassed. It can also be applied to learning control of a timing valve.

[0056]

[Operation and effect of the invention] Generally, a controlled variable for learning control
When the setting of the target value (for example, line pressure) is subject to some limitation, the learning control means performs learning assuming that there is no such limitation, so that erroneous learning occurs and learning control accuracy deteriorates. However, according to the first aspect, ene
Since learning control is stopped at the time of gear shift in which the amount of torque reduction is limited, erroneous learning is prevented and learning control accuracy is improved. Also, learning control when the amount of engine torque reduction is limited to reality is stopped, ene in other words
Since the learning control is not stopped only by the possibility that the amount of torque reduction is limited, the regulation of the learning control is suppressed to the necessary minimum, and the chances of performing the learning control increase.

According to the second invention, basically, the same operation and effect as those of the first invention can be obtained. In addition, the friction element
When the operating pressure is limited, the operating pressure of the friction element
Since the learning control is stopped , erroneous learning at the time of learning control of the operating pressure is prevented.

According to the third invention, basically, the same operation and effect as those of the second invention can be obtained. In addition, the friction element
Since the operating pressure is the line pressure, the line pressure is learned and controlled.
It prevents false learning.

According to the fourth invention, basically, the same operation and effect as those of the second invention can be obtained. In addition, the friction element
Since the operating pressure is used as the clutch pressure, the clutch pressure can be learned.
False learning at the time of control is prevented.

[0060] According to a fifth aspect of the present invention, basically the first to
Any one the same effects of the invention of 4 is obtained. Furthermore, the physical quantity that reflects the gear shift state is the gear shift time.
Therefore, the shift time can be maintained at an appropriate value, and
The occurrence of quick shock can be prevented .

[0061] According to the sixth invention, basically the first to
Any one the same effects of the invention of 5 is obtained. In addition, the physical quantity that reflects the gear shift state is
Since the input rotation is changed, the friction element on / off timing
Can hold the gearing properly and generate a shift shock.
Life is prevented .

[0062]

[0063]

[0064]

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of first to sixth inventions corresponding to claims 1 to 6 .

FIG. 2 is a system configuration diagram of an automatic transmission including a control device according to the present invention.

FIG. 3 is a skeleton diagram showing a power transmission mechanism of an automatic transmission.

FIG. 4 is a circuit diagram of a hydraulic mechanism of an automatic transmission.

FIG. 5 is a part of a flowchart showing a control method of line pressure control according to the present invention.

FIG. 6 is a part of a flowchart showing a control method of line pressure control according to the present invention.

FIG. 7 is a flowchart showing a control method of hydraulic pressure learning control.

FIG. 8 is a time chart when hydraulic pressure learning control is performed.

FIG. 9 is a diagram showing characteristics of a guard with respect to a torque reduction amount.

FIG. 10 is a diagram showing characteristics of a guard with respect to line pressure.

FIG. 11 is a diagram showing characteristics of a guard with respect to line pressure.

[Explanation of symbols]

3 ... engine 4 ... Automatic transmission 30 ... Transmission mechanism 41-46 ... Friction elements (clutch, brake) 60 ... Hydraulic control unit 63 ... Regulator valve 70 ... ECU

─────────────────────────────────────────────────── ─── Continuation of front page (56) Reference JP-A-2-42248 (JP, A) JP-A-62-67354 (JP, A) JP-A-61-119434 (JP, A) JP-A-63- 312558 (JP, A) JP 3-79857 (JP, A) JP 5-203026 (JP, A) JP 2-57760 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) F16H 59/00-61/12 F16H 61/16-61/24 F16H 63/40-63/48

Claims (6)

(57) [Claims] 1. An integrated control of an engine and an automatic transmission.
The engine output torque is reduced during gear shifting.
Engine torque down controllable
And control means, for the automatic transmission and the operating pressure learning control means for learning control of the working pressure of friction elements based on a comparison between the target value and the actual value of the physical quantity that reflects the shifting state in the gear shifting is provided In the control device, the operating pressure learning control means is an engine torque down control means.
Depending on when torque down control is possible and when it is not possible
Learn the operating pressure of the friction element by changing the target value of the physical quantity
Control, and when torque down control is possible.
When the amount of engine torque reduction is limited
Has come to stop learning control of the operating pressure of the friction element.
Control apparatus for an automatic transmission, characterized in that there. 2. The control device for an automatic transmission according to claim 1, wherein the operating pressure learning control means limits the operating pressure of the friction element.
Stop the learning control of the operating pressure of the friction element
Control device for an automatic transmission characterized in that it is so. 3. The control device for an automatic transmission according to claim 2 , wherein the operating pressure of the friction element is a line pressure . 4. The control device for an automatic transmission according to claim 2 , wherein the operating pressure of the friction element is a clutch pressure. 5. The control device for an automatic transmission according to any one of claims 1 to 4, wherein the physical quantity that reflects the shift state is a shift time . 6. The control device for an automatic transmission according to any one of claims 1 to 5, characterized in that the physical quantity that reflects the shift state is a change in input rotation of the transmission mechanism. Control device for automatic transmission.