JP2018123896A - Control method and control device for automatic transmission - Google Patents

Abstract

PROBLEM TO BE SOLVED: To properly suppress hysteresis generated at a solenoid valve.SOLUTION: While an automatic transmission is in operation, a dither is added to a solenoid current command value for a solenoid valve (B304, B305), and the solenoid valve is driven based upon the solenoid current command value to which the dither is added (B306). An amplitude of the dither is learnt (B303) with pressure applied to the solenoid valve made less than when the automatic transmission is in operation. In the learning, a current is supplied to the solenoid valve based upon a learning current command valve, and an amplitude of the dither is learnt based upon a solenoid actual current actually flowing to the solenoid valve when the learning current command value is varied.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a control method and a control apparatus for an automatic transmission that changes a gear ratio by hydraulic control.

  In automatic transmissions, solenoid valves are widely used in hydraulic control circuits that control the pressure of hydraulic oil (hereinafter sometimes simply referred to as “hydraulic pressure”). The solenoid valve can be controlled by moving the valve body by electromagnetic force generated by the solenoid and indirectly using the hydraulic pressure required for shifting directly or by using the hydraulic pressure adjusted by the solenoid valve as the pilot pressure. . Here, in the hydraulic control circuit, since the valve body exists in the oil, it is affected by various frictions including the viscous resistance of the hydraulic oil. Therefore, hysteresis exists in the operation of the valve body.

  Patent Document 1 discloses that hysteresis is suppressed by adding a dither to the solenoid drive current and moving the spool, which is a valve body, while slightly vibrating. Patent Document 1 further discloses that the dither amplitude is set by a decreasing function with respect to the temperature of the hydraulic oil (hereinafter sometimes simply referred to as “oil temperature”) (page 5, lower left column, 6th to 6th pages). 19).

Japanese Patent Laid-Open No. 03-213763

  However, in Patent Document 1, since the dither amplitude is only set according to the oil temperature, there are individual variations in hysteresis due to differences in assembly accuracy at the time of manufacture, or due to deterioration of hydraulic oil, etc. It is feared that the dither inherent effect cannot be obtained with respect to the suppression of hysteresis.

  Therefore, an object of the present invention is to enable appropriate suppression of hysteresis generated in a solenoid valve.

  In one embodiment of the present invention, hydraulic oil is supplied to an automatic transmission via a hydraulic control circuit having a solenoid valve, and when the automatic transmission is operated, a dither is added to the solenoid current command value for the solenoid valve. An automatic transmission control method for driving a solenoid valve based on a solenoid current command value to which is added. In the present embodiment, when a predetermined learning condition is established, the pressure applied to the solenoid valve is reduced more than when the automatic transmission is operated, current is supplied to the solenoid valve based on the learning current command value, and the learning current command value is changed. The dither amplitude is learned based on the actual solenoid current actually flowing through the solenoid valve.

  According to the present invention, the inherent effect of dither can be ensured by learning based on the actual solenoid current, and hysteresis generated in the solenoid valve can be appropriately suppressed. In learning, the accuracy of learning can be ensured by lowering the pressure applied to the solenoid valve than during operation of the automatic transmission.

FIG. 1 is a schematic diagram showing the overall configuration of a vehicle drive system according to an embodiment of the present invention. FIG. 2 is a schematic diagram showing a configuration of a hydraulic control circuit according to the embodiment. FIG. 3 is a schematic diagram showing a configuration of a control system of the solenoid valve according to the embodiment. FIG. 4 is an explanatory diagram showing the relationship between the solenoid current command value, the solenoid actual current, and the stroke of the solenoid valve. FIG. 5 is an explanatory diagram showing current-stroke characteristics of the solenoid valve. FIG. 6 is an explanatory diagram showing an effect (suppression of hysteresis) by adding dither. FIG. 7 is an explanatory diagram of the dither amplitude learning operation according to the first embodiment. FIG. 8 is an explanatory diagram of the dither amplitude learning operation according to the second embodiment. FIG. 9 is a flowchart showing a flow of learning execution determination. FIG. 10 is a flowchart showing a basic flow of gear ratio control. FIG. 11 is a flowchart showing the flow of dither amplitude learning according to the first embodiment. FIG. 12 is a flowchart showing the contents of processing in part A of the flowchart shown in FIG. FIG. 13 is a flowchart showing the flow of dither amplitude learning according to the second embodiment. FIG. 14 is a flowchart showing the contents of processing in part B of the flowchart shown in FIG. FIG. 15 is a target gear ratio map used for gear ratio control. FIG. 16 is an explanatory diagram showing the relationship between the hysteresis width and the dither amplitude learning value.

  Embodiments of the present invention will be described below with reference to the drawings.

(Overall configuration of vehicle drive system)
FIG. 1 schematically shows the overall configuration of a vehicle drive system PT according to an embodiment of the present invention.

  In the present embodiment, an internal combustion engine (hereinafter simply referred to as “engine”) 1 is provided as a drive source, and the rotational power of the engine 1 is the torque converter 2, the first gear train 3, the automatic transmission 4, and the second gear train. (Final gear) 5 and the differential 6 are transmitted to the left and right drive wheels 7. In the present embodiment, the second gear train 5 is provided with a park lock mechanism 8 that mechanically fixes the output shaft of the automatic transmission 4 so that it cannot rotate during parking.

  The torque converter 2 includes a lock-up clutch 2a, and by fastening the lock-up clutch 2a, the input shaft and the output shaft of the torque converter 2 are directly connected, and slip between input and output in the torque converter 2 is suppressed. Loss due to fluid connection can be reduced.

  The automatic transmission 4 includes a continuously variable transmission mechanism (hereinafter referred to as “variator”) 20 constituting a main transmission, and a sub-transmission 30 disposed on the output side of the variator 20. The rotational power input via the first gear train 3 is converted at a predetermined gear ratio and then output to the second gear train 5. In this embodiment, the variator 20 is a belt-type continuously variable transmission mechanism, and includes a primary pulley 21, a secondary pulley 22, and a belt 23 wound between the pair of pulleys 21 and 22.

  Each of the primary pulley 21 and the secondary pulley 22 includes a fixed conical plate and a movable conical plate arranged with the sheave surfaces facing each other with respect to the fixed conical plate. A letter-shaped groove is formed. The movable conical plate is provided with hydraulic cylinders 23a and 23b on the back, and by adjusting the pressure of the hydraulic oil supplied to the hydraulic cylinders 23a and 23b, the movable conical plate is moved coaxially with respect to the fixed conical plate, The width of the V groove can be changed. As a result, the contact diameters of the sheave surfaces of the pulleys 21 and 22 and the belt 23 are changed, and the gear ratio of the variator 20 is changed steplessly.

  Furthermore, in the present embodiment, the oil pump 10 driven by a part of the rotational power of the engine, the lockup clutch 2a of the torque converter 2 and the pulleys 21 of the variator 20, 22 (hydraulic cylinders 23a, 23b), and a hydraulic control circuit 11 that adjusts the pressure of hydraulic oil supplied to the hydraulic oil. The hydraulic control circuit 11 includes a plurality of oil passages and a plurality of pressure control valves, and the operation of the pressure control valve is controlled based on a shift control signal from the transmission controller 12 to switch the hydraulic pressure supply path.

  FIG. 2 shows a configuration of a portion related to the shift control of the variator 20 in the hydraulic control circuit 11.

  The hydraulic oil discharged from the oil pump 10 is regulated by the control valve 111, and the pressure control valves 116 and 117 for adjusting the hydraulic pressure applied to the low pressure control valve 112 and the hydraulic cylinders 23a and 23b of the pulleys 21 and 22, respectively. And supplied to. The low-pressure control valve 112 is connected to the solenoid valves 113, 114, 115 on the output side, and the outputs of the solenoid valves 114, 115 adjusted according to the degree of excitation are supplied to the pressure control valves 116, 117. Therefore, it is possible to adjust the hydraulic pressure of the hydraulic cylinders 23a and 23b by controlling the operation of the solenoid valves 113 to 115. In the present embodiment, linear solenoid valves are employed for the solenoid valves 113 to 115.

(Drive system control configuration and basic operation)
Operations of the engine 1 and the automatic transmission 4 are controlled by an engine controller and a transmission controller 12 (not shown), respectively. Each of these controllers is configured as an electronic control unit, and includes a central processing unit (CPU), various storage devices such as a RAM and a ROM, a microcomputer provided with an input / output interface and the like.

  The engine controller inputs a detection signal from an operation state sensor that detects the operation state of the engine 1 and executes a predetermined calculation based on the operation state, so that the fuel injection amount, fuel injection timing, ignition timing, etc. of the engine 1 Set.

  A transmission controller 12 includes an accelerator sensor 41 that detects an accelerator pedal operation amount (hereinafter referred to as “accelerator opening”) APO by a driver, and a rotational speed sensor (input) that detects a rotational speed Ni of an input shaft of the automatic transmission 4. Side rotation speed sensor) 42 and a detection signal of a rotation speed sensor (output side rotation speed sensor) 43 for detecting the rotation speed No of the output shaft of the automatic transmission 4 are input.

  Further, the transmission controller 12 includes a vehicle speed sensor 44 that detects a traveling speed (hereinafter referred to as “vehicle speed”) VSP of the vehicle, a brake sensor 45 that detects a brake pedal force BPF indicating a depression amount of a brake pedal by a driver, an automatic transmission. An oil temperature sensor 46 for detecting the temperature of the hydraulic oil, an inhibitor switch 47 for detecting the position of a shift select lever provided in the driver's seat, a rotational speed sensor 48 for detecting the rotational speed of the engine, and a shift range of the automatic transmission 1 OD (overdrive) switch 49 for expanding to a smaller gear ratio, hydraulic sensor 50 for detecting the hydraulic pressure (hereinafter referred to as “primary pressure”) Ppri of the hydraulic cylinder 23a of the primary pulley 21, and hydraulic cylinder 23b of the secondary pulley 22 Oil pressure (hereinafter “secondary pressure”) Psec is detected. In addition to inputting the detection signals Sa 51, inputs the detection signals of the various current sensor placed ignition switch, the hydraulic control circuit 11. In the present embodiment, as a current sensor installed in the hydraulic control circuit 11, a current sensor (represented by the current sensor 52) that detects a current flowing through the solenoids or coils of the solenoid valves 114 and 115 is provided.

  The transmission controller 12 sets a target speed ratio tRatio of the automatic transmission 4 (in this embodiment, the variator 20) based on various signals indicating the accelerator opening APO, the input side rotational speed Ni, the vehicle speed VSP, and the like. The primary pressure Ppri and the secondary pressure Psec are controlled so that the actual speed ratio Ratio becomes closer to the target speed ratio tRatio. Specifically, a control signal is output to the hydraulic control circuit 11 so that a predetermined shift hydraulic pressure acts on the primary pulley 21 and the secondary pulley 22 using the hydraulic pressure generated by the oil pump 10 as a base pressure.

(Solenoid valve control configuration)
FIG. 3 schematically shows the configuration of the control system of the solenoid valves 114 and 115.

  The transmission controller 12 subtracts the actual hydraulic pressure (actual hydraulic pressure) P from the target value (target hydraulic pressure) tP of the hydraulic pressure supplied to the hydraulic cylinders 23a and 23b by the subtraction unit B301, and divides the subtracted value from the actual hydraulic pressure. The amount ΔP is output to the feedback calculation unit B302. The feedback calculation unit B302 calculates a feedback control amount by PI calculation or the like from the actual hydraulic pressure deviation amount ΔP, and outputs it as a solenoid current command value tIsol. On the other hand, the transmission controller 12 learns the dither amplitude by the dither learning calculation unit B303 based on the solenoid actual current Isol detected by the current sensor 52, and dithers the learning value (hereinafter referred to as “amplitude learning value”). Output to the setting unit B304. Then, the dither setting unit B304 sets a dither having the amplitude learning value as an amplitude, and the addition unit B305 adds the dither to the solenoid current command value tIsol output from the feedback calculation unit B302. Based on a solenoid current command value to which dither is added (hereinafter, sometimes referred to as “solenoid current correction command value”) tIsol ′, a drive signal is generated by the PWM control unit B306, and the switch signal provided in the hydraulic control circuit 11 It outputs to SW and drives the solenoid valves 114 and 115 which are controlled objects. In the present embodiment, the dither cycle is set to a constant value.

  In the present embodiment, the “current command value setting unit” is configured by the subtraction unit B301 and the feedback calculation unit B302, the “current command value correction unit” is configured by the dither setting unit B304 and the addition unit B305, and the PWM control unit B306 The “solenoid driving unit” is configured, and the “dither learning unit” is configured by the dither learning calculation unit B303.

(Explanation of dither learning control)
A basic procedure for learning dither to be added to the solenoid current command value tIsol will be described.

  In the present embodiment, the dither amplitude is a learning target, and the period is a constant value.

  FIG. 4 shows changes in the solenoid current command value tIsol and the solenoid actual current Isol with respect to time in the upper stage, and changes in the stroke of the valve bodies provided in the solenoid valves 114 and 115 in the lower stage. FIG. 5 shows current-stroke characteristics when the solenoid valves 114 and 115 are driven by the same solenoid current command value tIsol as in FIG.

  4 and 5 show a state in which no hydraulic pressure is applied to the solenoid valves 114 and 115 (in other words, a state in which the hydraulic pressure is not applied to the valve body of the solenoid valve, including a case where the valve body is not in the oil). It is the result of the simulation which simulated. “Hydraulic pressure applied to the solenoid valve” refers to a force acting on the valve body in the direction of its stroke based on the pressure of the hydraulic oil. Specifically, in the case of a solenoid valve that achieves pressure reduction at the port on the inlet side. The primary pressure corresponds to the secondary pressure, and the primary pressure corresponds to the solenoid valve that achieves pressure reduction at the port on the outlet side. In this embodiment, a solenoid valve that achieves pressure reduction at the port on the inlet side is employed, and the “hydraulic pressure applied to the solenoid valve” refers to the secondary pressure of the solenoid valve, in other words, the pressure control valves 116 and 117. Applicable to pilot pressure.

  In the present embodiment, during learning, a solenoid current command value (hereinafter, a solenoid current command value set at the time of learning is referred to as “learning current command value”) tIsol is changed to a ramp shape, and the solenoid valves 114 and 115 are used for learning. Supply current. With respect to the learning current command value tIsol, the solenoid actual current Isol changes with a response delay according to the inductance of the solenoid and the like, and also changes according to the position or stroke of the valve body. In the present embodiment, the learning current command value tIsol is changed so that the valve bodies of the solenoid valves 114 and 115 move over the entire stroke region. The learning current command value tIsol is changed at a speed slower than the current response of the solenoid valves 114 and 115. However, the speed is close to the current response so that the influence of the stroke on the actual solenoid current Isol does not disappear.

  Here, as shown in FIG. 5, the solenoid valves 114 and 115 have a stroke start point or a start point depending on the friction that the valve body receives, depending on whether the solenoid current command value tIsol is increased or decreased. There is a difference (hysteresis) between the current at the end point or stop point. The current difference between point A and point D indicates hysteresis at the start point (hereinafter, the current difference at a specific position or stroke is referred to as “hysteresis width”), and the current difference between point B and point C indicates hysteresis at the end point. As shown in FIG. 4, since a characteristic change appears in the solenoid actual current Isol at each of the points A to D, the points A to D are identified based on the solenoid actual current Isol, and the points A to D are identified. It is possible to detect the solenoid actual current Isol at D.

  In general, a solenoid valve used for pressure control generally stays at a pressure equilibrium point under normal operating conditions. 5 and 6, the stroke at the pressure equilibrium point is indicated by a one-dot chain line. The hysteresis width at this pressure equilibrium point is estimated from the solenoid actual current Isol at each point A to D, and the dither amplitude is learned based on the hysteresis width at the pressure equilibrium point. By imagining a trapezoid (shown by a two-dot chain line in FIG. 5) having points A to D as four corners, the current difference (hysteresis width) at the pressure equilibrium point can be calculated geometrically. The stroke at the pressure equilibrium point can be grasped in advance based on the operating conditions of the solenoid valve. Then, the amplitude after learning is set as the amplitude learning value to the amplitude of the dither used in the next gear ratio control.

  FIG. 6 shows the effect (suppression of hysteresis) by adding dither. The stroke when the solenoid current command value tIsol before the dither is added is supplied to the solenoid valves 114 and 115 is indicated by a dotted line, and the stroke when the solenoid current command value tIsol ′ to which the dither is added is indicated by a solid line. Yes. Thus, in this embodiment, the addition of dither suppresses hysteresis in the entire stroke region, and the hysteresis width is reduced to almost half at the pressure equilibrium point.

  When learning is performed in a state where hydraulic pressure is applied to the solenoid valves 114 and 115, the solenoid valves 114 and 115 are already at the pressure equilibrium point when the supply of the learning current command value tIsol is started. Therefore, the hysteresis width at the substantial start point of the stroke, in other words, the current difference between the points A and D in the characteristic diagram shown in FIG. 5 is detected, and the dither amplitude is learned based on this.

  FIG. 7 is an explanatory diagram of the dither amplitude learning operation according to the first embodiment. The relationship between the actual current change rate, which is the change amount per unit of the solenoid actual current Isol, and the stroke of the solenoid valves 114 and 115 is shown. In addition, each point A to D of the characteristic diagram shown in FIG. 5 is shown.

In the present embodiment, the points A to D are identified from the actual current change rate, and the dither amplitude is learned based on the solenoid actual currents Isol _{A to} Isol _D at the points A to _D. 4 and 7 together, each of the points A to D can be specified as a point showing the following characteristics in the change in the actual current change rate.

Point A: The actual current change rate decreases while the solenoid current command value tIsol is increasing, and reaches the first predetermined value α _A.
Point B: The point at which the actual current change rate further decreases from the first predetermined value α _A and reaches the second predetermined value α _B while the solenoid current command value tIsol increases after the point A has elapsed. .
Point C: This is a point at which the actual current change rate increases while the solenoid current command value tIsol turned from an increase, and reaches a third predetermined value α _C.
Point D: While the solenoid current command value tIsol has decreased after the point C has elapsed, the actual current change rate has changed from decreasing to increasing, and has reached a fourth predetermined value α _D smaller than the third predetermined value α _C. Is a point.

  FIG. 8 is an operation explanatory diagram of dither amplitude learning according to the second embodiment. The actual current divergence amount, which is the divergence amount of the solenoid actual current Isol with respect to the learning current command value tIsol, the strokes of the solenoid valves 114 and 115, Shows the relationship. In the present embodiment, the actual current deviation amount is a subtraction value obtained by subtracting the solenoid actual current Isol from the learning current command value tIsol.

  Each point A to D is not limited to the actual current change rate, but can be specified from the actual current deviation amount. 4 and 8 together, each of the points A to D can be specified as a point that exhibits the following characteristics in the change in the actual current deviation amount.

Point A: The point at which the actual current divergence amount increases after reaching the predetermined value after the current supply to the solenoid valve is started.
Point B: The point at which the actual current deviation amount further increased and reached the positive peak.
Point C: The point at which the actual current divergence amount once stabilized after passing the positive peak, and then the actual current divergence amount began to decrease again.
Point D: The point at which the actual current deviation amount increases and reaches a predetermined value after the actual current deviation amount has passed the negative peak.

(Explanation based on flowchart)
Hereinafter, speed ratio control and dither amplitude learning according to the present embodiment will be described with reference to flowcharts.

  FIG. 9 is a flowchart showing a flow of learning execution determination. The transmission controller 12 executes a learning execution determination routine every predetermined time after the ignition switch is turned on.

  In S101, it is determined whether a learning condition is satisfied. If the learning condition is satisfied, it is determined that the learning condition is satisfied, and the process proceeds to S102. If the learning condition is not satisfied, the process proceeds to S105. The transmission controller 12 has, as a learning condition, whether or not a predetermined time has elapsed after the ignition switch is switched from on to off (condition A), when the vehicle speed VSP is stopped at a predetermined vehicle speed VSP1 or less, and automatically It is determined whether the transmission range of the transmission 4 is a P (parking) range (condition B1), the vehicle speed VSP is stopped at a predetermined vehicle speed VSP1 or less, and the transmission range of the automatic transmission 4 is a range other than the P range (for example, It is determined whether or not the vehicle is in a brake-on state where the brake pedal is depressed (condition B2). Then, when any of the condition A and the conditions B1 and B2 is satisfied, it is determined that the learning condition is satisfied.

  In S102, it is determined whether or not learning of the dither amplitude is incomplete. If the learning has not been completed, the process proceeds to S103. If the learning has already been completed, the current routine is terminated. Whether or not dither amplitude learning is incomplete can be determined based on the value of a dither learning completion flag Flrn described later.

  In S103, dither amplitude learning is executed according to the flowcharts shown in FIGS.

  In S104, it is determined that learning of the dither amplitude has been completed. Specifically, the value of the dither learning completion flag Flrn is switched from 0 to 1.

  In S105, as a normal control, the gear ratio of the automatic transmission 4 is controlled according to the flowchart shown in FIG.

  FIG. 10 is a flowchart showing a basic flow of gear ratio control. The transmission controller 12 executes a gear ratio control routine at predetermined intervals after the ignition switch is turned on.

  In S201, various input signals are read. Specifically, the accelerator opening APO, the actual current Isol (= Ipri) of the primary pressure control solenoid valve 114, the actual current Isol (= Isec) of the secondary pressure control solenoid valve 115, the engine rotational speed Ne, and the input side rotation The speed Ni, the output side rotational speed No, the vehicle speed VSP, and the like are read.

  In S202, the target speed ratio tRatio of the variator is set. The target gear ratio tRratio is set with reference to the map data of the trend shown in FIG. 15 based on the accelerator opening APO, the input side rotational speed Ni, and the vehicle speed VSP.

In S203, a norm response command value rRratio having a predetermined delay with respect to the target gear ratio tRratio is calculated. The calculation of the normative response command value rRratio is, for example, according to the following equation (1). s is a Laplace operator, and T _ref is a normative response time constant.
rRratio = {1 / (T _ref s + 1) × tRratio (1)

  In S204, a target secondary pressure (secondary pulley thrust) tPsec necessary for transmitting torque (engine torque) input to the primary pulley 21 to the secondary pulley 22 via the belt 23 is calculated. The transmission controller 12 has map data in which the target secondary pressure tPsec is assigned to the input torque, and calculates the target secondary pressure tPsec by referring to this map data based on the engine torque.

In S205, the target primary pressure tPpri is calculated based on the normative response command value rRratio. Specifically, the difference between the norm response command value rRratio and the actual speed ratio Ratio is calculated so that the actual gear ratio (hereinafter referred to as “actual gear ratio”) Ratio of the variator 20 approaches the norm response command value rRratio, Based on this difference, feedback control such as PI calculation is performed by the following equation (2). The actual speed ratio Rratio is calculated by dividing the rotational speed of the primary pulley 21 by the rotational speed of the secondary pulley 22. The rotational speed of the secondary pulley can be calculated based on the output side rotational speed No and the speed ratio of the auxiliary transmission 30. Kp and Ki are a proportional gain and an integral gain, respectively, and are set in advance in consideration of the stability of the feedback loop.
tPpri = [Kp + (Ki / s)] × (rRatio−Ratio) (2)

  In S206, a solenoid current command value tIsol (= tIpri) for the primary pressure control solenoid valve 114 is calculated based on the target primary pressure tPpri. Specifically, the difference between the target primary pressure tPpri and the actual hydraulic pressure Ppri is calculated so that the actual primary pressure Ppri approaches the target primary pressure tPpri, and the solenoid current command value tIpri is calculated by feedback control such as PI calculation. . Further, the solenoid current command value tIsol (= tIsec) for the secondary pressure control solenoid valve 115 is similarly calculated based on the target secondary pressure tPsec. The solenoid current command value tIsol can be calculated not only by feedback control but also by multiplying the target hydraulic pressure tP by a pressure-current conversion gain.

  In S207, a dither having a learned amplitude (a constant value when the vehicle is shipped) is added to the solenoid current command value tIsol (tIpri, tIsec) to calculate solenoid current correction command values tIpri 'and tIsec'. In the present embodiment, the dither cycle is set to a constant value.

In S208, voltage command values tVpri and tVsec for the solenoid valves 114 and 115 are calculated so that the solenoid actual currents Ipri and Isec approach the solenoid current correction command values tIpri ′ and tIsec ′. The calculation of the voltage command values tVpri and tVsec is performed, for example, by feedback control such as PI calculation represented by the following equations (3) and (4). Kpc and Kic are a proportional gain and an integral gain, respectively, and are set in advance in consideration of the stability of the feedback loop.
tVpri = [Kpc + (Kic / s)] × (tIpri′−Ipri) (3)
tVsec = [Kpc + (Kic / s)] × (tIsec′−Isec) (4)

Then, the duty Dpri and Dsec of the drive pulse signal of the switching element SW are calculated and output to the switching element SW so that the voltage command values tVpri and tVsec are realized by PWM control. V _B is the current drive voltage of the solenoid valves 114 and 115 (FIG. 3).
tDpri = (tVpri / V _B ) × 100 (5)
tDsec = (tVsec / V _B ) × 100 (6)

  11 and 12 are flowcharts showing the flow of dither amplitude learning according to the first embodiment. The transmission controller 12 executes a dither amplitude learning routine by interrupt processing in S103 of the flowchart shown in FIG. In the present embodiment, when learning the dither amplitude, the hydraulic pressure applied to the target solenoid valves 114 and 115 is made lower than that during normal operation. Specifically, the clamping force that does not cause belt slippage at the variator 20 when starting from a stop is reduced below the hydraulic pressure required to form the pulleys 21 and 22. In the present embodiment, in order to secure the driving force at the time of starting, the speed ratio of the variator 20 is changed to the largest value in setting before or during a stop.

In S301, the change rate (actual current change rate) μ of the solenoid actual current Isol is calculated. Specifically, the calculation is performed by subjecting the actual current Isol detected by the current sensor 52 to an approximate differentiation process represented by the following equation (7). _Tdiv is a filter time constant and is set in consideration of measurement noise.
μ = {s / (T _div s + 1)} × Isol (7)

  In S302, it is determined whether or not the solenoid current command value tI is increasing, in other words, whether the solenoid current command value tIsol is increasing or decreasing. When it is increasing, it progresses to S303, and when it is decreasing, it progresses to S401 of the flowchart shown in FIG. Whether the solenoid current command value tIsol is increasing is determined based on the difference between the current value of the solenoid current command value tIsol and the past value (for example, the solenoid current command value calculated in the previous routine). Is possible.

In S303, it is determined whether or not learning of point A (stroke increase start point) in the characteristic diagram shown in FIG. 5 is incomplete. If the learning of the point A is not completed, the process proceeds to S304, and if already completed, the process proceeds to S306. Whether or not the learning of the point A is incomplete is determined according to the value of the point A learning completion flag Flrn _A , for example, when Flrn _A = 0, and when Flrn _A = 1. Determine that it has already been completed.

In S304, it is determined whether or not the actual current change rate μ is equal to or less than the first predetermined value α _A. When the actual current change rate μ is equal to or less than the first predetermined value alpha _A, the process proceeds to S305, is greater than the first predetermined value alpha _A, the process proceeds to S308.

In S305, the solenoid actual current Isol is detected and stored as the current value Isol _A of the point A, and the learning of the point A is completed, and the point A learning completion flag Flrn _A is switched from 0 to 1.

In S306, it is determined whether the actual current change rate μ is equal to or less than a second predetermined value α _B. When the actual current change rate μ is equal to or less than the second predetermined value alpha _B, the process proceeds to S307, is greater than the second predetermined value alpha _B, the process proceeds to S308.

In S307, the solenoid actual current Isol is detected and stored as the current value Isol _B of the point B, and the learning of the point B is completed, and the point B learning completion flag Flrn _B is switched from 0 to 1.

Moving to FIG. 12, in S401, it is determined whether or not learning of point C (stroke reduction start point) in the characteristic diagram shown in FIG. 5 is incomplete. If the learning of the point C is not completed, the process proceeds to S402, and if already completed, the process proceeds to S404. Whether or not the learning of the point C is incomplete is determined according to the value of the point C learning completion flag Flrn _C , for example, when Flrn _C = 0, and when Flrn _C = 1. Determine that it has already been completed.

In S402, it is determined whether or not the actual current change rate μ is equal to or greater than a third predetermined value α _C. When the actual current change rate μ is equal to or greater than the third predetermined value α _C , the process proceeds to S403, and when it is smaller than the third predetermined value α _C , the process proceeds to S308 in the flowchart shown in FIG.

In S403, the solenoid actual current Isol is detected and stored as the current value Isol _C of the point C, and the point C learning completion flag Flrn _C is switched from 0 to 1 assuming that the learning of the point C is completed.

In S404, the actual current change rate μ is equal to or fourth predetermined value or more alpha _D. Specifically, after learning of the point C is completed, it is determined whether or not the actual current change rate μ has changed from a decrease to an increase and has reached a fourth predetermined value α _D. If the fourth predetermined value α _D has been reached, the process proceeds to S405, and if not, the process proceeds to S308 in the flowchart shown in FIG.

In S405, the solenoid actual current Isol is detected and stored as the current value Isol _D of the point D, and the point D learning completion flag Flrn _D is switched from 0 to 1 assuming that the learning of the point D is completed.

Returning to FIG. 11, in S <b> 308, it is determined whether learning has been completed for all of the points A to D. If the values of the learning completion flags Flrn _A , Flrn _B , Flrn _C , and Flrn _D for each point A to _D are all 1, it is determined that learning has been completed for all of the points A to D, and the process proceeds to S 309. If the learning is not completed at any of the points A to D, the process returns to S301, and the processes of S301 to S308 are repeated.

In S309, the dither amplitude is learned, and the routine returns to the routine shown in the flowchart of FIG. At the time of learning, unlike the normal operation (S208), not feedback control but feed-forward control in which a value obtained by multiplying the learning current command value tIsol by the impedance of the current drive stage is set as a voltage command value. This is because the influence of the stroke of the valve body easily appears in the current. In this embodiment, the amplitude learning value is calculated according to the following formula (8) or (9) according to the following cases (a) and (b).
(A) When the hydraulic pressure applied to the target solenoid valves 114 and 115 is not more than a predetermined value sufficiently lower than the pressure during operation of the automatic transmission 4, amplitude learning value = (Isol _A −Isol _D ) × (Sf− Se) / Sf + (Isol _B- Isol _C ) × Se / Sf (8)
Sf: Full stroke Se: Stroke at pressure equilibrium point (a) When the hydraulic pressure applied to the target solenoid valves 114, 115 is lower than the pressure during operation of the automatic transmission 4, but higher than the predetermined value Amplitude learning Value = Isol _A− Isol _D (9)

  The amplitude learning value may be calculated as a larger value as the hysteresis width increases, with reference to the tendency calculation table shown in FIG. 16 instead of the current difference itself as the hysteresis width.

(Explanation of effects)
The control device for the automatic transmission 4 according to the present embodiment is configured as described above, and effects obtained by the present embodiment will be described below.

  First, by adding a dither to the solenoid current command value tIsol (tIpri, tIsec), hysteresis generated in the solenoid valves 114 and 115 can be suppressed. Here, the current based on the learning current command value is supplied to the solenoid valve, and the dither amplitude is learned based on the solenoid actual current Isol (Ipri, Isec) when the learning current command value is changed. It is possible to deal with a case where there is a secular change due to individual variation or deterioration for each transmission 4, and the hysteresis can be appropriately suppressed. Furthermore, the learning accuracy can be ensured by learning the dither amplitude in a state in which the pressure applied to the solenoid valves 114 and 115 is lower than that during the operation of the automatic transmission 4. In the present embodiment, the “state in which the pressure is reduced” includes a state in which no hydraulic pressure is applied to the solenoid valves 114 and 115 and a state in which the valve body is not in the oil. In this sense, “when the learning condition is satisfied” The term “concept” is not limited to the time when the automatic transmission 4 is in operation, but also includes the time before shipment of the vehicle and the time of maintenance or repair.

  Second, by learning the dither amplitude based on the change rate (actual current change rate) of the solenoid actual current Isol, the influence of the current response component in the solenoid actual current Isol is compensated, and the stroke of the valve body Since it is possible to detect the change in the corresponding solenoid actual current Isol more accurately, learning can be performed with high accuracy.

  Third, in learning, the learning current command value tIsol is changed so that the valve element moves over the entire stroke region, thereby evaluating the hysteresis as compared with a case where the valve body is moved only in a part of the stroke region. Can be performed more appropriately, and learning can be performed with higher accuracy.

  Fourth, by making the speed at which the learning current command value is changed slower than the current response of the solenoid valves 114 and 115, the time difference between the movement (stroke) of the valve body and the change in the solenoid actual current Isol can be reduced. It is possible to suppress and improve the accuracy of learning.

  Fifth, after the ignition switch is turned off, in other words, learning is performed after the operation of the vehicle is stopped, so that the current flowing through the solenoid valves 114 and 115 affects the operation of the vehicle during learning (for example, the driver feels uncomfortable) ) Can be avoided. On the other hand, it is possible to prevent the learning accuracy from being impaired by the current for driving the vehicle, and to ensure the learning accuracy.

  Sixth, learning can be performed at more opportunities by performing learning when the vehicle is stopped in a brake-on state or when the P range of the automatic transmission 4 is selected. On the other hand, when shifting from the brake-on state to the off-state or switching from the P range to the travel range, it is possible to secure a time margin for stopping learning and returning to normal control.

  Seventh, when learning, the period of the PWM carrier is made shorter than the current response time constant of the solenoid valves 114 and 115, so that a high-frequency current response component resulting from PWM control can be removed from the solenoid actual current Isol. Thus, it is possible to accurately detect a change in the solenoid actual current Isol according to the operation of the valve body, and to perform learning accurately.

(Description of other embodiments)
FIGS. 13 and 14 are flowcharts showing the flow of dither amplitude learning according to the second embodiment, and the same reference numerals as those shown in FIGS. 11 and 12 are used for steps for performing the same processing as in the first embodiment. The description is omitted. As in the case of the first embodiment, the transmission controller executes a dither amplitude learning routine by interrupt processing.

In S501, it is determined whether or not the actual current deviation amount D increases while the learning current command value tIsol increases and reaches a predetermined value β _A greater than 0. If the actual current deviation amount D reaches the predetermined value β _A , the process proceeds to S305, and if not, the process proceeds to S308.

In S502, the learning current command value actual current deviation amount D during an increase in tIsol further increases from the predetermined value beta _A, determines whether the reached positive peak. Specifically, it is determined whether or not the actual current deviation amount D is smaller than the actual current deviation amount Dn−1 calculated in the previous routine. If the actual current divergence amount D is smaller than the previous value Dn−1 and reaches the positive peak, the process proceeds to S307, and if not, the process proceeds to S308.

  In S601, after the actual current deviation amount D reaches the positive peak, it is determined whether or not the current current deviation amount D has once stabilized and started to decrease again. Specifically, it is determined whether the actual current deviation amount D has a negative value and is smaller than the previous value Dn−1. If the actual current divergence amount D is stabilized and then starts decreasing again, the process proceeds to S403. If the actual current deviation amount D does not start decreasing, the process proceeds to S308 in the flowchart shown in FIG.

In S602, started to increase the actual current deviation amount D from decreasing, it is determined whether reaches a smaller predetermined value beta _D than 0. If the actual current deviation amount D has reached the predetermined value β _D , the process proceeds to S405, and if not, the process proceeds to S308 in the flowchart shown in FIG.

  In this way, the dither amplitude can also be learned by the deviation amount (actual current deviation amount) of the solenoid actual current Isol from the learning current command value tIsol, and the influence of the current response component on the solenoid actual current Isol is compensated. Learning can be performed with high accuracy.

  In the above description, the case where the variator 20 of the automatic transmission 4 is configured by a belt-type continuously variable transmission mechanism has been described. However, the variator 20 is not limited to a belt-type and is configured by a toroidal-type continuously variable transmission mechanism. It may be.

Furthermore, in order to detect hysteresis occurring in the solenoid valves 114 and 115, the case where the four points (point A to point D) in the characteristic diagram shown in FIG. 5 are specified has been described as an example. The hysteresis can be detected and learning can be achieved simply by specifying this point. For example, with specifying the point A shown in FIG. 5, the solenoid actual current Isol _A at the point A, the spring constant K of Itoguchimoto (return spring of the solenoid valves 114 and 115 of interest, setting the stroke STR of the valve body, From the solenoid current-thrust conversion coefficient COEFsol), the hysteresis width at the pressure equilibrium point is calculated by the following equation (10), and the amplitude learning value is calculated based on this. As a result, learning can be completed in a shorter time than when all of the four corner points A to D are specified, and energy consumption required for learning can be reduced.
Dither amplitude = Isol _A −K × STR / COEFsol (10)

  Although the embodiments of the present invention have been described above, the present invention is not limited to these embodiments, and various changes and modifications can be made within the scope of the matters described in the claims. Not too long.

PT ... vehicle drive system 1 ... engine 2 ... torque converter 3 ... first gear train 4 ... automatic transmission 5 ... second gear train (final gear)
6 ... Differential 7 ... Drive wheel 10 ... Oil pump 11 ... Hydraulic control circuit 12 ... Transmission controller 20 ... Variator (main transmission)
21 ... Primary pulley 22 ... Secondary pulley 23 ... Belt 30 ... Sub-transmission

Claims (10)

Supply hydraulic oil to the automatic transmission via a hydraulic control circuit with a solenoid valve,
A control method for an automatic transmission, wherein during operation of the automatic transmission, a dither is added to a solenoid current command value for the solenoid valve, and the solenoid valve is driven based on the solenoid current command value to which the dither is added. ,
When certain learning conditions are met,
Lowering the pressure applied to the solenoid valve than during operation of the automatic transmission,
Supplying a current to the solenoid valve based on a learning current command value, and changing the learning current command value;
Learning the dither amplitude based on the actual solenoid current flowing through the solenoid valve;
Control method of automatic transmission.   The method of controlling an automatic transmission according to claim 1, wherein the learning current command value is changed in a ramp shape.   The method of controlling an automatic transmission according to claim 1 or 2, wherein the amplitude of the dither is learned based on an actual current change rate that is a change amount per unit time of the solenoid actual current.   The method of controlling an automatic transmission according to claim 1 or 2, wherein the dither amplitude is learned based on an actual current deviation amount that is a deviation amount between the learning current command value and the solenoid actual current.   5. The method of controlling an automatic transmission according to claim 1, wherein the learning current command value is changed so that the valve body of the solenoid valve moves over the entire stroke region. 6.   The method for controlling an automatic transmission according to any one of claims 1 to 5, wherein a speed at which the learning current command value is changed is slower than a current response of the solenoid valve.   The automatic transmission according to any one of claims 1 to 6, wherein when the automatic transmission is mounted on a vehicle, the predetermined learning condition is satisfied after the vehicle is stopped. Control method.   When the automatic transmission is mounted on a vehicle, the predetermined learning condition is satisfied when the vehicle is stopped in a brake-on state or when the P range of the automatic transmission is selected. The method of controlling an automatic transmission according to one item.   9. The method according to claim 1, wherein when a current is supplied to the solenoid valve by PWM control based on the learning current command value, a cycle of the PWM carrier is made shorter than a current response time constant of the solenoid valve. The control method of the automatic transmission as described. A control device for an automatic transmission to which hydraulic oil is supplied via a hydraulic control circuit having a solenoid valve,
A current command value setting unit for setting a solenoid current command value for the solenoid valve;
A current command value correction unit for adding dither to the solenoid current command value set by the current command value setting unit;
A solenoid drive unit that outputs a drive signal based on a solenoid current command value to which the dither is added to the solenoid valve;
A dither learning unit for learning the dither amplitude;
With
The dither learning unit supplies current to the solenoid valve based on a learned current command value in a state where pressure applied to the solenoid valve is lower than that during operation of the automatic transmission, and changes the learned current command value. Learning the dither amplitude based on the actual solenoid current flowing through the solenoid valve.
Control device for automatic transmission.

Images