JPH0674847B2 - Shift control device for continuously variable transmission - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a shift control device for a continuously variable transmission, and more particularly to a continuously variable shift for correcting a deviation of an engine speed or a gear ratio from a target value by a PI or PID operating means. The present invention relates to a shift control device for a machine.

[Prior art and its problems]

Conventionally, various continuously variable transmissions such as a V-belt type continuously variable transmission and a toroidal type continuously variable transmission have been proposed. As a shift control method therefor, an engine speed or a gear ratio is used as a control amount, and this is used as a PID operation means. It is known to perform feedback control using the.

As is well known, the PID operation means is a proportional operation (P operation).
In order to eliminate the problem that the steady deviation and the phase delay remain with respect to the target value at, the feedback signal is not only the signal proportional to the output of the system, but also the signal obtained by integrating and differentiating the output of the system. It is a thing.

However, in the PID control, the operation starts only when the state changes and a deviation occurs between the control amount and the target value, so there is a delay in the transient characteristics until the target value is reached, or the control amount is the target value. There is a problem that the value fluctuates oscillatingly and the convergence to the target value is delayed. In particular, in the shift control of the continuously variable transmission for vehicles, the PID control that simply detects the deviation because the engine speed and the gear ratio, which are the target values, change greatly, such as during kickdown, or change momentarily. I can't follow quickly.

Therefore, a means for determining the reference operation amount according to the driving signal such as the throttle opening is provided, and when the operating state changes, the steady-state reference operation amount that can be reached in advance is predicted, and this reference operation amount is used for the PID operation. By adding to the output of the means and inputting it to the shift control actuator, it is possible to exhibit excellent followability even with a large state change.

However, if the reference operation amount is uniformly determined in accordance with the operation signal such as the throttle opening, the integrated correction amount of the PID operating means may not be 0. The reason for this is that the reference manipulated variable deviates from the optimum value due to variations in individual vehicles and changes over time.If this deviation is attempted to be corrected by integral correction, transient overshoot or undershoot will occur. Cheap.

[Object of the Invention]

The present invention has been made in view of the above problems, and an object thereof is to exhibit excellent followability to a large state change, absorb variations in individual vehicles and changes over time, and improve transient characteristics. It is an object of the present invention to provide a shift control device for a multi-speed transmission.

[Structure of Invention]

In order to achieve the above object, the present invention provides a shift control device for a continuously variable transmission that corrects a deviation of an engine speed or a gear ratio from a target value by a PI or PID operating means, and a reference manipulated variable according to a driving signal. And a reference operation amount determining means for adding the reference operation amount to the output of the PI or PID operating means and inputting it to the shift control actuator, and a learning command signal and the PI for each set time in the steady operation state. Alternatively, the learning command means for outputting a reset signal for clearing the integrator of the PID operation means and the learning command signal are operated in conjunction with the output of the integrator of the PI or PID operation means to reduce this. And a learning means for correcting the reference operation amount in the direction.

That is, by using the output of the integrator of the PI or PID operating means and learning-correcting the reference operation amount in the direction in which it decreases, the deviation of the reference operation amount due to individual vehicle variations and changes over time is eliminated, and the shift characteristics are improved. It can be automatically adjusted to almost constant.

[Explanation of Examples]

FIG. 1 shows a schematic structure of a V-belt type continuously variable transmission which is an example of the continuously variable transmission according to the present invention, in which a crankshaft 2 of an engine 1 is connected to an input shaft 4 via a damper mechanism 3. . An external gear 5 is fixed to the end of the input shaft 4,
The external gear 5 meshes with the internal gear 6 fixed to the drive shaft 11 of the continuously variable transmission 10, decelerates the power of the input shaft 4 and transmits it to the drive shaft 11.

The continuously variable transmission 10 includes a drive pulley 12 provided on a drive shaft 11,
The driven shaft 13 is composed of a driven pulley 14 and a V belt 15 wound between the pulleys. Drive pulley 12
Has a fixed sheave 12a and a movable sheave 12b, and a torque cam device 16 and a compression spring 17 are provided behind the movable sheave 12b. The torque cam device 16 generates a thrust force proportional to the input torque, and the compression spring 17 generates V
The belt 15 generates an initial thrust that does not slacken, and these thrusts apply the belt tension required for torque transmission to the V-belt 15. On the other hand, the driven pulley 14 is also the drive pulley 12
Similarly, it has a fixed sheave 14a and a movable sheave 14b, and behind the movable sheave 14b is a hydraulic chamber 18 for gear ratio control.
Is provided. The hydraulic pressure to the hydraulic chamber 18 is controlled by a pulley control valve 43 described later.

A hollow shaft 19 is rotatably supported on the outer periphery of the driven shaft 13, and the driven shaft 13 and the hollow shaft 19 are disengaged by a starting clutch 20 composed of a wet multi-plate clutch. The hydraulic pressure to the starting clutch 20 is controlled by a starting control valve 45 described later. A forward gear 21 and a reverse gear 22 are rotatably supported on the hollow shaft 19, and one of the forward gear 21 and the reverse gear 22 is connected to the hollow shaft 19 by a forward / reverse switching dog clutch 23. It is designed to be connected. Reverse idler shaft 24
Is a reverse idler gear 25 that meshes with the reverse gear 22,
Another reverse idler gear 26 is fixed. Also,
The counter shaft 27 includes a counter gear 28 that simultaneously meshes with the forward gear 21 and the reverse idler gear 26, and a final reduction gear.
29 is fixed, and the final reduction gear 29 meshes with the ring gear 31 of the differential device 30 to transfer power to the output shaft 32.
Have been communicated to.

The pressure regulating valve 40 regulates the hydraulic pressure discharged from the oil sump 41 by the oil pump 42 and outputs it as a line pressure to the pulley control valve 43 and the start control valve 45. The pulley control valve 43 and the start control valve 45 actuate the solenoids 44 and 46 by a control signal (for example, a duty control signal) output from the electronic control unit 60 to regulate the line pressure to respectively control the hydraulic chambers of the driven pulley 14.
The control hydraulic pressure is output to 18 and the starting clutch 20, respectively. Therefore, from the electronic control unit 60 to the solenoids 44,46
It is possible to freely control the gear ratio of the continuously variable transmission 10 and the torque transmission capacity of the starting clutch 20 only by the control signal to.

FIG. 2 shows a structural diagram of the electronic control unit 60, in which 61 is a sensor for detecting the engine speed Nin (the rotational speed of the input shaft 4), and 62 is a vehicle speed V (the rotational speed of the output shaft 32). A sensor for detecting the rotational speed Nout of the driven shaft 13 (the input rotational speed of the starting clutch 20 or the rotational speed of the driven pulley 14), 64 is a shift position of P, R, N, D, L. Sensor to detect, 6
5 is a sensor for detecting the throttle opening, the signals of the sensors 61-64 are input to the input interface 66,
The signal of the sensor 65 is converted into a digital signal by the A / D converter 67. 68 is a central processing unit (CPU), 69 is a read only memory (ROM) in which programs and data for controlling the pulley control solenoid 44 and the start control solenoid 46 are stored, and 70 is sent from each sensor. Random access memory (RA that temporarily stores signals and parameters
M), 71 is an output interface, and these CPU68,
The ROM 69, the RAM 70, the output interface 71, the input interface 66 and the A / D converter 67 are interconnected by a bus 72. The output of the output interface 71 is output as a control signal to the pulley control solenoid 44 and the start control solenoid 46 via the output driver 73.

FIG. 3 is a block diagram of the shift control system of the electronic control unit 60. The target engine speed determining means 80 determines the target engine speed N _R according to the throttle opening, vehicle speed and range mode signal. There is. Reference numeral 81 is a PID operating means, which includes an integrator 82 having a gain K _I , a proportional device 83 having a gain Kp, and a differentiator 84 having a gain Kd. In addition, s is a Laplace operator.
As is well known, the integrator 82 outputs a signal D ₁ that integrates the deviation between the target engine speed that is the input and the actual engine speed, and has the function of eliminating the steady deviation. Proportioner 83
Has a function of outputting a signal D ₂ proportional to the deviation and adjusting the gain, and the differentiator 84 has a function of outputting a signal D ₃ obtained by differentiating the deviation to improve the responsiveness.

Here, the output D ₁ of the integrator 82 is D ₁ = K _I · S (1) where S = ∫ (Nin−N _R ) dt The respective output signals (duty ratio) of the PID operating means 81 are added. Is input to the control valve 43, which is an actuator for gear shift control, and the control valve 43 provides the control hydraulic pressure according to the input signal to the continuously variable transmission (CVT) 10, and the engine after the shift control by the continuously variable transmission 10 is performed. The engine speed is negatively fed back to the target engine speed.

Reference numeral 85 is a reference manipulated variable determining means for predicting a steady-state standard manipulated variable D ₀ (duty ratio) that can be reached in advance, and adding the reference manipulated variable D ₀ to the output of the PID operating means 81 to control valve 43. Is output to. The reference manipulated variable D ₀ is determined by the following equation according to the throttle opening which is an input signal and the reference value A.

D ₀ = f (θ) + A (2) Note that f (θ) is a function of the throttle opening θ.

The learning instruction means 86 has a built-in timer, counts the elapsed time T, and compares it with a predetermined learning sampling time T _L. Further, the learning command means 86 has an actual engine speed Ni.
A deviation (| Nin−N _R |) between n and the target engine speed N _R is input, and this deviation is compared with a predetermined value δ. And | Nin−
When N _R | ≦ δ and T ≧ T _L , the learning command signal and the integrator 8
The reset signal for clearing 2 is output at the same time.

The output D ₁ of the integrator 82 is input to the learning means 87, and the A value of the reference manipulated variable is updated as follows according to the input of the learning command signal as follows, and does not disappear even when the power is turned off. Write to memory.

In the above equation, the initial value of A value is 0, and α is 0
It is a learning weighting coefficient that takes any value up to 100.

Next, the operation of the shift control system having the above configuration will be described with reference to FIG.

First, when the control starts, first the count value T is reset to 0 and the timer is cleared (step 90), and then signals such as the engine speed Nin, the driven shaft speed Nout, the vehicle speed V, and the throttle opening θ are sent. Enter (91). Next, after determining the target engine speed N _R according to the throttle opening, vehicle speed and mode / range signal (at 92), the duty value (D ₁ + D ₂ + D ₃ ) by the PID control term is determined (at the same). 9
3). Further, the reference manipulated variable D ₀ corresponding to the throttle opening and the reference value A (initial value is 0) is determined by the equation (2) (94), and the duty value and the reference manipulated variable D according to the PID control term are determined.
The sum D with ₀ is calculated (at 95).

Next, the absolute value of the difference between the actual engine speed Nin and the target engine speed N _R is compared with a constant value δ (96), and | N
When in−N _R |> δ, the actual engine speed Nin greatly deviates from the target engine speed N _R , that is, because it is in a transient state, the value A is not updated and The obtained duty ratio D is directly output to the pulley control solenoid 44 (97) and the process returns to the beginning.

On the other hand, when | Nin−N _R | ≦ δ, the elapsed time T after the timer clear is compared with the learning sampling time T _L (see 9
8) When T < _TL, the duty ratio D is output to the pulley control solenoid 44 (99) and the process returns to.

Further, when the condition | Nin−N _R | ≦ δ is continued for the time T _L or more, the A value of the reference manipulated variable D ₀ is updated by the equation (3) (100), and the integrator 92 is set. Clear (101), return to the beginning.

As described above, the reference operation amount D ₀ is corrected by learning control so that this value D ₁ is gradually reduced using the integral correction term D ₁ , so that the reference operation amount due to variations in individual vehicles and changes with time The gap can be easily eliminated. Therefore, it is possible to realize optimal shift characteristics with little variation for each vehicle, and it is possible to prevent transient overshoot and undershoot because the deviation of the reference operation amount is not corrected by integral correction.

In particular, only the steady state is detected under the conditions of | Nin−N _R | ≦ δ and T ≧ T _L , and learning control is not performed in the transient state.
Frequent correction of the standard operation amount is not performed, and running is not adversely affected. Also, since the integrator 92 is cleared each time correction is made,
The past data has no influence after the correction, and there is no fear that the state will change significantly during the correction.

Although the shift control device using the PID operating means has been described in the above embodiment, it can be similarly applied to the case where the PI operating means is used.

〔The invention's effect〕

As is clear from the above description, according to the present invention, PI or
By using the main force of the integrator of the PID operation means to correct the reference operation amount by learning control in the direction of decreasing it, the deviation of the reference operation amount due to individual vehicle variations and changes over time does not depend on integral correction. It can be resolved. Therefore, it is possible to easily absorb variations in each vehicle and changes with time without impairing the excellent responsiveness due to the addition of the reference operation amount, and it is also possible to eliminate transient overshoot and undershoot.

[Brief description of drawings]

FIG. 1 is a schematic diagram of a V-belt type continuously variable transmission to which the present invention is applied, FIG. 2 is a configuration diagram of an electronic control unit, FIG. 3 is a block diagram of a shift control system of the electronic control unit, and FIG. FIG. 4 is a flow chart showing the operation of the control device of the present invention. 1 ... Engine, 10 ... Continuously variable transmission, 43 ... Pulley control valve, 44 ... Pulley control solenoid, 60 ... Electronic control unit, 80 ... Target engine speed determining means, 81 ... PID
Operating means, 82 ... integrator, 83 ... proportionalizer, 84 ... differentiator, 85 ... reference manipulated variable determining means, 86 ... learning command means,
87 …… Learning means.

Claims (1)

[Claims] 1. A shift control device for a continuously variable transmission in which a deviation of an engine speed or a gear ratio from a target value is corrected by PI or PID operating means, and a reference operation amount is determined according to a driving signal, and the reference operation is performed. A reference manipulated variable determining means for adding an amount to the output of the PI or PID operating means and inputting it to the shift control actuator, and a learning command signal and an integrator for the PI or PID operating means at every set time in the steady operation state. Learning instruction means for outputting a reset signal for clearing, and the PI or PID
A shift control device for a continuously variable transmission, comprising: learning means for correcting a reference operation amount in a direction in which the output of an integrator of the operating means is reduced.