OIL PRESSURE ESTIMATING DEVICE AND STORAGE MEDIUM

Abstract

Oil pressure estimating device, which defines as the simulation target a torque converter (20) comprising two oil chambers (25, 26) and a lock-up clutch (30), wherein the lock-up clutch (30) is switched between an engaged state and a disengaged state depending on a differential pressure between the two oil chambers (25, 26), and which, based on a command differential pressure which is a setpoint of the differential pressure for the torque converter (20), calculates an estimated differential pressure which is an estimated value of the differential pressure generated in the torque converter (20), wherein the oil pressure estimating device comprises: a storage device (95) designed to store map data (D) defining a map (Dx), wherein the map (Dx) outputs, in response to the input of an input variable, an estimated differential pressure variable, which is a variable indicating the estimated differential pressure, wherein the map (Dx) includes a command differential pressure variable, which is a variable indicating the command differential pressure as one of a plurality of input variables; and an execution device (91, 93) designed to perform a detection process for acquiring a value of the input variables and a calculation process for inputting the value of the input variables acquired by the detection process into the characteristic map (Dx) in order to calculate a value of the output variables, characterized by the fact that the characteristic map (Dx) was trained by machine learning; the storage device (95) and the execution device (91, 93) are provided in a vehicle (500) equipped with the torque converter (20); the storage device (95) is designed to store a plurality of characteristic maps (Dx) for respective phases, which are obtained by dividing a period of time from the start of switching the bridging clutch (30) into the engaged state until the start of switching the bridging clutch (30) into the disengaged state into a plurality of phases; the execution device (91, 93) is designed to perform the acquisition process, a selection process and the calculation process; the acquisition process includes acquiring the value of the input variable when the command differential pressure is output to the torque converter (20); the selection process involves selecting the characteristic map (Dx) from the characteristic maps (Dx) for the respective phases, where the characteristic map (Dx) corresponds to a phase in which the command differential pressure is output; and The calculation process involves calculating the value of the output variable by entering the value of the input variable acquired through the acquisition process into the characteristic field (Dx) selected through the selection process.

Description

BACKGROUND OF THE INVENTION 1. Field of the invention

[0001] The present invention relates to an oil pressure estimating device and a storage medium. 2. Description of the state of the art

[0002] JP 2010 - 270 822 A discloses a control system for a torque converter's lock-up clutch that switches between a PD and a PID controller during start-up to improve responsiveness. A characteristic map and oil pressure estimation help control the transition and reduce the number of sensors required.

[0003] US 2005 / 0230207A1 discloses a control system for a lock-up clutch of a torque converter that precisely adjusts the clutch pressure based on a piston back pressure map to reduce vibrations and save fuel.

[0004] A vehicle disclosed in JP 2010-156359A includes an internal combustion engine, a torque converter, and an automatic transmission. An output shaft of the internal combustion engine is connected to an input shaft of the torque converter. The torque converter transmits the torque between the input and output shafts via a fluid. The output shaft of the torque converter is connected to the input shaft of the automatic transmission.

[0005] The torque converter includes a lock-up clutch. The lock-up clutch switches between an engaged state, in which the input and output shafts of the torque converter are directly mechanically connected, and a disengaged state, in which the connection is broken. The lock-up clutch operates dependent on oil pressure.

[0006] The vehicle includes a control device. The control device switches the state of the lock-up clutch by controlling the oil pressure. BRIEF DESCRIPTION OF THE INVENTION

[0007] In a technique like the one described in JP 2010-156359A, which controls the oil pressure relative to the lock-up clutch, a deviation can occur between a target oil pressure and the actual oil pressure. Therefore, to control the oil pressure with high accuracy, it is necessary to estimate the actual oil pressure with high precision and to control the oil pressure taking into account the deviation between the estimated and target values. To accurately estimate the actual oil pressure, it is necessary to understand the relationship between the target and actual oil pressure and to incorporate this relationship into the pressure estimation. This requires deriving relational expressions and the like that express the relationship between the target and actual oil pressure. However, such relational expressions are very complex, and their derivation requires considerable time and effort.

[0008] The purpose of the present disclosure is to control the oil pressure of a lock-up clutch with high accuracy. Specifically, the purpose of the present disclosure is to provide a means of estimating the actual oil pressure with high accuracy, which enables highly accurate control of the lock-up clutch.

[0009] This problem is solved by an oil pressure estimating device according to claim 1 and by a non-transient storage medium according to claim 7.

[0010] A first aspect of the present invention relates to an oil pressure estimating device. The oil pressure estimating device defines as its simulation target a torque converter with two oil chambers and a lock-up clutch, wherein the lock-up clutch switches between an engaged and a disengaged state depending on a differential pressure between the two oil chambers, and calculates, based on a command differential pressure or instruction differential pressure, which is a setpoint or command value of the differential pressure for the torque converter, an estimated differential pressure, which is an estimated value of the differential pressure generated in the torque converter. The oil pressure estimating device includes a storage device and an execution device.The storage device is designed to store map data that defines a characteristic map. This map, in response to an input variable, outputs an estimated differential pressure variable, which represents the estimated differential pressure. The map has been trained using machine learning. Specifically, the map includes a command differential pressure variable, which represents the command differential pressure as one of several input variables. The execution device is designed to perform a capture process to acquire a value from the input variables and a computation process to input the acquired value into the map to calculate a value for the output variable.

[0011] In the oil pressure estimation device of the first aspect, when the characteristic map is used to calculate the estimated differential pressure, it is only necessary to train the map using suitable data to accurately calculate the estimated differential pressure. Therefore, incorporating the relationship between the command differential pressure and the actual differential pressure into the calculation of the estimated differential pressure does not require the effort of deriving a complex relational expression.

[0012] In the oil pressure estimation device of the first aspect, the storage device and the execution device are provided in a vehicle equipped with the torque converter. The storage device is designed to store a plurality of characteristic maps for respective phases, which are obtained by subdividing the period from the start of the lock-up clutch shifting into the engaged state until the start of the lock-up clutch shifting into the disengaged state into a plurality of phases. The execution device is designed to perform the acquisition process, a selection process, and the calculation process. The acquisition process involves acquiring the value of the input variables when the command differential pressure is output to the torque converter.The selection process involves choosing from the available characteristic maps the one corresponding to the phase in which the command differential pressure is output. The calculation process involves calculating the value of the output variable by inputting the value of the input variable acquired during the acquisition process into the characteristic map selected during the selection process.

[0013] In the oil pressure estimating device with the above configuration, a fixed characteristic curve is used for each phase. Therefore, even if the relationship between the command differential pressure and the estimated differential pressure differs depending on the phase, the difference can be included in the output variable.

[0014] In the oil pressure estimator with the above configuration, one of the input variables can be an acceleration variable, which represents the vehicle's acceleration. Even if the relationship between the command differential pressure and the estimated differential pressure differs during vehicle acceleration compared to deceleration, the oil pressure estimator with this configuration can still obtain an accurate output variable that accounts for the difference.

[0015] In the oil pressure estimating device with the above configuration, one of the input variables can be an accelerator pedal variable, which represents the amount of actuation of the vehicle's accelerator pedal. Because the accelerator pedal variable is included as one of the input variables, the oil pressure estimating device with the above configuration can produce an output variable that reflects the torque applied to the torque converter.

[0016] In the oil pressure estimating device with the above configuration, one of the input variables can be a vehicle speed variable, which indicates the vehicle's speed. Even if the relationship between the command differential pressure and the estimated differential pressure differs depending on the vehicle's speed, the oil pressure estimating device with the above configuration can still obtain an accurate output variable.

[0017] In the oil pressure estimating device with the above configuration, one of the input variables can be a switching variable, which is a variable indicating the gear ratio of the vehicle's transmission. In the oil pressure estimating device with the above configuration, an accurate output variable can be obtained even if the relationship between the command differential pressure and the estimated differential pressure differs depending on the transmission's gear ratio.

[0018] In the oil pressure estimating device with the above configuration, one of the input variables can be an oil temperature variable, which represents the temperature of the hydraulic oil supplied to the torque converter. The oil pressure estimating device with the above configuration can also have an output variable that takes the hydraulic oil temperature into account, which can influence the estimated differential pressure.

[0019] A second aspect of the present invention is a non-transitory / non-rewritable storage medium that stores instructions which can be executed by one or more processors contained in an implementing device of an oil pressure estimating device, and which cause the one or more processors to perform the following functions. These functions include a detection process for acquiring the value of an input variable and a computation process for inputting the value of the input variable acquired by the detection process into a characteristic map in order to calculate the value of an output variable.The oil pressure estimator defines a torque converter as its simulation target. This converter contains two oil chambers and a lock-up clutch, where the lock-up clutch switches between an engaged and disengaged state depending on the differential pressure between the two oil chambers. The oil pressure estimator includes a storage device and an execution device. The storage device stores map data that defines a characteristic map. In response to input variables, the characteristic map outputs an estimated differential pressure variable as its output variable. This differential pressure variable represents an estimated value of the differential pressure generated in the torque converter. The characteristic map has been trained using machine learning.Here, the characteristic map includes a command differential pressure variable, which is a variable that indicates a command differential pressure, which is a setpoint of the differential pressure for the torque converter as one of a plurality of input variables.

[0020] In the storage medium of the second aspect, if the characteristic map is used to calculate the estimated differential pressure, it is only necessary to train the characteristic map using suitable data in order to calculate the estimated differential pressure appropriately. Therefore, including the relationship between the command differential pressure and the actual differential pressure in the calculation of the estimated differential pressure does not require the effort of deriving a complex relational expression. BRIEF DESCRIPTION OF THE FIGURES

[0021] Features, advantages and technical and industrial significance of exemplary embodiments of the present invention are described below with reference to the accompanying drawings, in which the same symbols denote the same elements, and wherein: Fig. 1 a schematic configuration diagram of a vehicle according to an embodiment of the present invention; Fig. 2 a schematic configuration diagram of a in Fig. The torque converter shown in section 1 is; Fig. 3 is a diagram that is an example of a in Fig. The bridging characteristic shown in Figure 1 indicates; Fig. 4 is a diagram showing an example of a time series change of a command differential pressure in an indentation operation initiated by a Fig. The control device shown in 1 is implemented; Fig. 5 is a flowchart showing the process flow of a differential pressure estimation process performed by the control device; and Fig. Figure 6 is a schematic diagram showing a modification of the oil pressure estimating device according to the embodiment. DETAILED DESCRIPTION OF THE EXECUTION FORMS

[0022] The following describes an embodiment of an oil pressure estimating device used in a vehicle, with reference to the figures. First, the schematic setup of the vehicle is described. As shown in Fig. As shown in Figure 1, a vehicle 500 comprises an internal combustion engine 10, a torque converter 20, an automatic transmission 80, a differential 71, and drive wheels 72. The internal combustion engine 10 is a power source for the vehicle 500. A crankshaft, which is an output shaft of the internal combustion engine 10, is connected to an input shaft 21 of the torque converter 20. The torque converter 20 is a fluid coupling with a torque amplification function. An output shaft 22 of the torque converter 20 is connected to an input shaft of the automatic transmission 80. The automatic transmission 80 is a multi-stage transmission. An output shaft of the automatic transmission 80 is connected to the drive wheels 72 via the differential 71. The differential 71 allows for a speed difference between the right and left drive wheels 72.

[0023] The torque converter 20 comprises a pump impeller 23 and a turbine impeller 24. Specifically, the pump impeller 23 is connected to the input shaft 21 of the torque converter 20. The pump impeller 23 rotates integrally with the input shaft 21. The turbine impeller 24 is located opposite the pump impeller 23. The turbine impeller 24 is connected to the output shaft 22 of the torque converter 20. The turbine impeller 24 rotates integrally with the output shaft 22. In the torque converter 20, the torque is transmitted between the pump impeller 23 and the turbine impeller 24 via a fluid.

[0024] The torque converter 20 includes a lock-up clutch 30, as used in Fig. Figure 2 shows the bridging clutch 30 as a multi-plate clutch. This means that the bridging clutch 30 comprises a plurality of first friction discs 32 and a plurality of second friction discs 34. In the present embodiment, two first friction discs 32 and two second friction discs 34 are provided. It should be noted that Fig. Figure 2 shows only a first friction plate 32. The first friction discs 32 and the second friction discs 34 are arranged alternately. The adjacent first and second friction discs 32, 34 face each other. The first friction discs 32 and the second friction discs 34 are movable in their direction of arrangement. The first friction discs 32 rotate integrally with the pump impeller 23. The second friction discs 34 rotate integrally with the turbine impeller 24. A group of friction discs consisting of the first friction discs 32 and the second friction discs 34 is called a friction disc group.

[0025] The torque converter 20 includes a front cover 20a and a piston 37. The front cover 20a is located on one side of the friction disc assembly in the direction of the first friction discs 32 and the second friction discs 34. The front cover 20a forms part of the housing of the torque converter 20. The piston 37 is located on the other side of the friction disc assembly in the direction of the first friction discs 32 and the second friction discs 34. The piston 37 is movable in the direction of the first friction discs 32 and the second friction discs 34. When the piston 37 approaches the front cover 20a, the friction disc assembly is clamped between the piston 37 and the front cover 20a. In this case, the adjacent first and second friction discs 32, 34 come into contact with each other.When the piston 37 moves away from the front cover 20a, the state in which the friction disc assembly is clamped between the piston 37 and the front cover 20a is released. In this case, the adjacent first and second friction discs 32, 34 are arranged in separate positions.

[0026] The torque converter 20 includes a control oil chamber 25, a front oil chamber 26, and a rear oil chamber 27. Specifically, the control oil chamber 25 is located near the friction disc assembly in the torque converter 20. The control oil chamber 25 is connected to a control port 25a. Hydraulic oil is supplied to the control oil chamber 25 via the control port 25a. The hydraulic oil in the control oil chamber 25 is discharged to the outside via the control port 25a.

[0027] The front oil chamber 26 is defined near the control oil chamber 25. The front oil chamber 26 is connected to a supply port 26a. The front oil chamber 26 is supplied with hydraulic oil via the supply port 26a. The rear oil chamber 27 is defined within the torque converter 20. The rear oil chamber 27 is connected to the front oil chamber 26. The rear oil chamber 27 is connected to an outlet port 27a. The hydraulic oil in the rear oil chamber 27 is discharged to the outside via the outlet port 27a.

[0028] The front oil chamber 26 borders the control oil chamber 25 in the direction of arrangement of the first friction discs 32 and the second friction discs 34. Part of the piston 37 extends between the front oil chamber 26 and the control oil chamber 25. The front oil chamber 26 is located more towards the front cover 20a than the piston 37, in the direction of arrangement of the first friction discs 32 and the second friction discs 34. Although a detailed illustration is omitted, the piston 37 separates the front oil chamber 26 and the control oil chamber 25. The piston 37 therefore operates in the direction of arrangement of the first friction discs 32 and the second friction discs 34 according to the differential pressure between the front oil chamber 26 and the control oil chamber 25.

[0029] The operating state of the lock-up clutch 30 is switched by actuating the piston 37, depending on the differential pressure between the front oil chamber 26 and the control oil chamber 25. Hereinafter, a value obtained by subtracting the hydraulic oil pressure (hereinafter referred to as oil pressure) PF of the front oil chamber 26 from the oil pressure PC in the control oil chamber 25 is referred to as the lock-up differential pressure ΔP. In the present embodiment, the piston 37 moves away from the front cover 20a when the lock-up differential pressure ΔP is negative, that is, when the oil pressure PC of the control oil chamber 25 is lower than the oil pressure PF of the front oil chamber 26. In this case, the adjacent first and second friction discs 32, 34 are arranged in separate positions. Thus, the operating state of the lock-up clutch 30 is an open state.Since the adjacent first and second friction discs 32, 34 are separated from each other, the pump impeller 23 and the turbine impeller 24 are not directly connected to each other.

[0030] On the other hand, the piston 37 approaches the front cover 20a when the differential pressure ΔP is equal to or greater than zero, that is, when the oil pressure PC of the control oil chamber 25 is equal to or greater than the oil pressure PF of the front oil chamber 26. Then the adjacent first and second friction discs 32, 34 come into contact with each other. This means that the operating state of the lock-up clutch 30 becomes engaged. Since the adjacent first and second friction discs 32, 34 come into contact with each other, the pump impeller 23 and the turbine impeller 24 are directly mechanically connected.

[0031] The modes in which the operating state of the lock-up clutch 30 is engaged include a semi-engaged state and a fully engaged state. When the differential pressure ΔP of the lock-up clutch is sufficiently small, the operating state of the lock-up clutch 30 is semi-engaged. In this case, the adjacent first and second friction discs 32, 34 are in a slip state, with slippage occurring between them. When the differential pressure ΔP of the lock-up clutch increases significantly, i.e., reaches the maximum permissible value, the operating state of the lock-up clutch 30 becomes fully engaged. In this case, no slippage occurs between the adjacent first and second friction discs 32, 34.

[0032] The torque transmission mode through the torque converter 20 is switched in accordance with the operating state of the lock-up clutch 30 described above. When the lock-up clutch 30 is in the open position, the amount of torque transmitted through the lock-up clutch 30 is zero. The torque converter 20 then transmits the torque from the input shaft 21 to the output shaft 22 by means of torque transmission between the pump impeller 23 and the turbine impeller 24 via a fluid. When the lock-up clutch 30 is in the partially engaged position, the efficiency of the torque transmission varies depending on the degree of slip of the first friction discs 32 and the second friction discs 34.The lower the slip, the smaller the difference between the rotational speed of the pump impeller 23 and the rotational speed of the turbine impeller 24, and the higher the efficiency of the torque transmission from the input shaft 21 to the output shaft 22. When the lock-up clutch 30 is fully engaged, the pump impeller 23 and the turbine impeller 24 rotate at the same speed. This allows the torque to be transmitted from the input shaft 21 to the output shaft 22 with virtually no loss.

[0033] Vehicle 500 includes a hydraulic circuit 40. Hydraulic circuit 40 is a flow path through which hydraulic oil flows. Hydraulic circuit 40 is connected to the control port 25a, the inlet port 26a, and the outlet port 27a. A plurality of (electro-)solenoid valves 42 are provided midway along hydraulic circuit 40. By switching the solenoid valves 42 on and off, hydraulic oil is supplied to and discharged from the control oil chamber 25, the front oil chamber 26, and the rear oil chamber 27.

[0034] As in Fig. As shown in Figure 1, the vehicle 500 is equipped with a vehicle speed sensor 58, an accelerator pedal sensor 59, an acceleration sensor 64, and an oil temperature sensor 56. The vehicle 500 is also equipped with an accelerator pedal 60. The vehicle speed sensor 58 detects a vehicle speed SP, which is the driving speed of the vehicle 500. The accelerator pedal sensor 59 detects an accelerator pedal actuation amount ACCP, which is the amount of actuation of the accelerator pedal 60 by a driver. The acceleration sensor 64 detects a front-to-rear acceleration W, which is the acceleration of the vehicle 500 in the front-to-rear direction. The oil temperature sensor 56 detects an oil temperature L, which is the temperature of the hydraulic oil flowing through the hydraulic circuit 40.

[0035] Next, an overview of the vehicle control device configuration is described. The vehicle 500 includes a control device 90. The control device 90 can be configured as one or more processors that execute various processes according to a computer program (software). The control device 90 can be configured as one or more dedicated hardware circuits, such as application-specific integrated circuits (ASICs), that execute at least some of the various processes, or as circuits that include a combination of hardware circuits. The processor includes a central processing unit (CPU) 91 and memory, such as random-access memory (RAM) and read-only memory (ROM) 93. The memory stores program code or instructions configured to instruct the CPU 91 to execute processes.The memory, that is, a computer-readable medium, includes any available medium that a general-purpose or specialized computer can access. Furthermore, the control device 90 includes a memory device 95, which is an electrically rewritable non-volatile memory. The CPU 91, the ROM 93, and the memory device 95 can communicate with each other via an internal bus 98.

[0036] Detection signals from various sensors attached to the vehicle 500 are input into the control unit 90. In particular, detection signals for the following parameters are input into the control unit 90: vehicle speed SP, which is detected by the vehicle speed sensor 58; accelerator pedal actuation amount ACCP, which is detected by the accelerator pedal sensor 59; front-to-rear acceleration W, which is detected by the acceleration sensor 64; and oil temperature L, which is detected by the oil temperature sensor 56.

[0037] The CPU 91 controls the internal combustion engine 10, the automatic transmission 80, and the like by executing various programs stored in the ROM 93. Specifically, the CPU 91 calculates a target torque for the internal combustion engine 10 based on the accelerator pedal input (ACCP) and the vehicle speed (SP). The CPU 91 then controls the internal combustion engine 10 so that its output torque becomes the target torque. Furthermore, the CPU 91 calculates a target shift point (SFT) for the automatic transmission 80 based on the accelerator pedal input (ACCP) and the vehicle speed (SP). The CPU 91 then controls the automatic transmission 80 so that the selected shift point (SFT) is reached.

[0038] The CPU 91 controls the bypass clutch 30 by executing the program stored in the ROM 93. During the control of the bypass clutch 30, the CPU 91 performs a target determination process, an engagement process, and a phase specification process. These processes are described sequentially below.

[0039] First, the target determination process is described. The CPU 91 repeatedly executes the target determination process while the vehicle 500 is in motion. The target determination process is a process for determining a target operating state of the lock-up clutch 30. During the target determination process, the CPU 91 determines the target operating state of the lock-up clutch based on the accelerator pedal actuation amount ACCP, the vehicle speed SP, the target shift stage SFT, and the acceleration state of the vehicle 500. In determining this state, the CPU 91 refers to a lock-up characteristic map stored in the ROM 93.

[0040] As in Fig. As shown in Figure 3, the bypass map defines the target operating state for each driving range of the vehicle 500, which is defined by the vehicle speed SP and the accelerator pedal input amount ACCP. The bypass map is created for each target shift stage SFT. For clarity, in each bypass map, the open state is defined in the driving range where the accelerator pedal input amount ACCP is large. In the driving range where the accelerator pedal input amount ACCP is small, the open state is defined on the low vehicle speed side, and the fully engaged state is defined on the high vehicle speed side. Then, the half-engaged state between the open state and the fully engaged state is defined in the driving range where the accelerator pedal input amount ACCP is greater than zero. As the thick solid line in Fig. Figure 3 shows that the half-engaged state is also defined in the driving range where the accelerator pedal input amount (ACCP) is zero. That is, it defines the driving range in which the operating state of the lock-up clutch transitions to the half-engaged state during vehicle deceleration.

[0041] When determining the target operating state of the lock-up clutch 30, the CPU 91 selects a lock-up characteristic map that corresponds to the current target shift stage SFT. The CPU 91 then determines the target operating state according to the current accelerator pedal actuation amount ACCP and the current vehicle speed SP within the selected lock-up characteristic map.

[0042] The engagement process is described next. When the target operating state of the lock-up clutch 30 is switched from the open state to the engaged state during the target determination process, the CPU 91 executes the engagement process. The engagement process is a procedure for switching the operating state of the lock-up clutch 30 to the engaged state and for maintaining the operating state of the lock-up clutch 30 in the engaged state. During the engagement process, the CPU 91 repeatedly calculates a command differential pressure ΔPG, which is a target value of the lock-up differential pressure ΔP for the torque converter 20. When the CPU 91 calculates the command differential pressure ΔPG, it outputs a control signal based on the command differential pressure ΔPG to the hydraulic circuit 40.Each solenoid valve 42 of the hydraulic circuit 40 operates in response to this control signal to supply or discharge hydraulic oil into the control oil chamber 25, the front oil chamber 26, and the rear oil chamber 27. That is, the CPU 91 outputs the command differential pressure ΔPG to the torque converter 20 via the control signal, so that hydraulic oil is supplied to and discharged from each oil chamber of the torque converter 20.

[0043] During the engagement process, the CPU 91 changes the instruction differential pressure ΔPG as follows. The operation of the bypass clutch 30 during the engagement process, along with one method of changing the instruction differential pressure ΔPG, is described below.

[0044] As in Fig. As shown in Figure 4, at time T1 when the engagement process is initiated, the CPU 91 rapidly increases the command differential pressure ΔPG for the torque converter 20 from a value P0, fixed in the open state, to an application differential pressure P1. The CPU 91 then maintains the command differential pressure ΔPG at the application differential pressure P1 until a predetermined initial time interval has elapsed since the start of the engagement process. During this time, the lock-up clutch 30 begins to engage. That is, the distance between the adjacent first and second friction discs 32, 34 begins to decrease. The application differential pressure P1 is predetermined as the lock-up differential pressure ΔP required to initiate engagement. The time at which the CPU 91 initiates the engagement process is the time at which the lock-up clutch 30 begins to transition from the open state to the engaged state.

[0045] The CPU 91 reduces the instruction differential pressure ΔPG to a ready differential pressure P2 at time T2, after the first time interval / period has elapsed from time T1. The CPU 91 then maintains the instruction differential pressure ΔPG at the ready differential pressure P2 until a predetermined second time interval has elapsed since the instruction differential pressure ΔPG was reduced to the ready differential pressure P2. During this time, the engagement of the bridging clutch 30 is completed. The ready differential pressure P2 is preset as the bridging differential pressure ΔP shortly before the adjacent first and second friction discs 32, 34 come into contact with each other.

[0046] At time T3 after the second time interval from time T2, the CPU 91 stops maintaining the command differential pressure ΔPG at the standby differential pressure P2. From time T3, the CPU 91 increases the command differential pressure ΔPG at a differential pressure change rate α until a predetermined third time interval has elapsed. During this time, the slip of the lock-up clutch 30 gradually decreases. When the second time interval has elapsed, the CPU 91 calculates a final command differential pressure P3 based on the accelerator pedal actuation amount ACCP, the vehicle speed SP, the target shift stage SFT, the oil temperature L, and the front-to-rear acceleration W of the vehicle 500.The final instruction differential pressure P3 is the bridging differential pressure ΔP required to achieve the indented state defined as the desired operating state, namely the fully indented state, or the half-indented state with a specific slip amount. When the CPU 91 calculates the final instruction differential pressure P3, it divides the value obtained by subtracting the ready differential pressure P2 from the final instruction differential pressure P3 by the third time interval and determines this value as the differential pressure change rate α. That is, the differential pressure change rate α is the rate of change of the instruction differential pressure ΔPG per unit of time required to bring the instruction differential pressure ΔPG to the final instruction differential pressure P3 during the third time interval. The CPU 91 increments the instruction differential pressure ΔPG based on the differential pressure change rate α.

[0047] At time T4, after the third time interval from time T3 has elapsed, the CPU 91 stops increasing the instruction differential pressure ΔPG. Subsequently, from time T4 onwards, the CPU 91 maintains the instruction differential pressure ΔPG at the final instruction differential pressure P3 until the target operating state of the bypass clutch 30 switches from the engaged state to the open state. During this process, the bypass clutch 30 is held with a specific amount of slip in either the fully engaged or the half-engaged state.

[0048] When the target operating state of the bypass clutch 30 switches from the engaged state to the open state, the CPU 91 terminates the engagement process. The CPU 91 then initiates a process to reduce the instruction differential pressure ΔPG as part of the process to switch the operating state of the bypass clutch 30 to the open state. That is, the point at which the CPU 91 terminates the engagement process is the point at which the bypass clutch 30 begins to transition from the engaged state to the open state.

[0049] Next, the phase specification process is described. The CPU 91 repeatedly executes the phase specification process during the execution of the indentation operation. The phase specification process is a process for specifying multiple phases, obtained by subdividing the execution process of the indentation operation, and for calculating phase variables (PHS) corresponding to the specified phases. The phase variables (PHS) are variables used to specify the phases of the indentation operation.

[0050] In the present embodiment, the CPU 91 specifies four phases. As in Fig. As shown in Figure 4, a first phase Q1 represents the period from time T1 to time T2, during which the instruction differential pressure ΔPG is maintained at the application differential pressure P1 in the sequence of operations of the induction process. A second phase Q2 represents the period from time T2 to time T3, during which the instruction differential pressure ΔPG is maintained at the readiness differential pressure P2. A third phase Q3 represents the period from time T3 to time T4, during which the instruction differential pressure ΔPG is increased at the differential pressure change rate α. A fourth phase Q4 represents the period from time T4 onward, during which the instruction differential pressure ΔPG is maintained at the final instruction differential pressure P3. The CPU 91 refers to the content of the processes executed during the induction process and specifies the phase within the induction process.

[0051] When the CPU 91 specifies the phase, it calculates the phase variable PHS. In this embodiment, the phase variable PHS is defined as a positive integer that identifies each phase, for example, "1" for the first phase Q1 and "2" for the second phase Q2. The CPU 91 calculates a value for the phase variable PHS that corresponds to the specified phase.

[0052] Next, the configuration of the control device for calculating the estimated differential pressure is described. The control device 90 acts as an oil pressure estimator that calculates an estimated differential pressure ΔPS, which is an estimated value of the bridging differential pressure ΔP that actually occurs in the torque converter 20. That is, the control device 90 sets the torque converter 20 as the simulation target and simulates the actual bridging differential pressure ΔP of the torque converter 20.

[0053] As in Fig. As shown in Figure 1, the storage device stores 95 map data D. The map data D are data that define a map Dx which outputs a variable when input variables are provided. In the present embodiment, the input variables include a command differential pressure variable, which is a variable that specifies the command differential pressure ΔPG; an acceleration variable, which is a variable that specifies the front-to-rear acceleration W; an accelerator pedal variable, which is a variable that specifies the amount of accelerator pedal actuation ACCP; a vehicle speed variable, which is a variable that specifies the vehicle speed SP; a shift variable, which is a variable that specifies the gear ratio of the automatic transmission 80; and an oil temperature variable, which is a variable that specifies the oil temperature L.The output variable is an estimated differential pressure variable, which is a variable that specifies the estimated differential pressure ΔPS.

[0054] The characteristic map data D stored in the storage device 95 includes the characteristic map Dx for each phase of the engagement process. That is, the storage device 95 stores four characteristic maps Dx, which belong to the first phase Q1 to the fourth phase Q4, obtained by dividing the execution process of the engagement process into four sections.

[0055] The CPU 91 can execute the differential pressure estimation process, which is a process for calculating the estimated differential pressure ΔPS during the execution of the engagement process. The CPU 91 implements each step of the differential pressure estimation process by executing the program stored in the ROM 93. In the present embodiment, the CPU 91 and the ROM 93 form an execution device.

[0056] The CPU 91 performs a data acquisition process, a selection process, and a calculation process within the differential pressure estimation process. In the data acquisition process, the CPU 91 acquires various input variables required for calculating the estimated differential pressure ΔPS, such as the command differential pressure ΔPG. The CPU 91 acquires these input variables when the command differential pressure ΔPG is output to the torque converter 20, that is, when the control signal for the hydraulic circuit 40 is output during the background engagement process. In the selection process, the CPU 91 selects from the characteristic maps Dx stored in the memory device 95 the characteristic map Dx that corresponds to the phase in which the command differential pressure ΔPG is output.In the calculation process, the CPU 91 calculates the value of the output variables by entering the values ​​of the input variables acquired through the acquisition process into the characteristic field Dx selected in the selection process.

[0057] Next, the differential pressure estimation process is described. The CPU 91 repeatedly executes the differential pressure estimation process during the execution of the indentation operation. As shown in Fig. As shown in Figure 5, when CPU 91 starts the differential pressure estimation process, it executes the process from step S10. In step S10, CPU 91 acquires various variables required for the processes from step S20. Specifically, CPU 91 acquires the phase variable PHS, the command differential pressure ΔPG, an acceleration characteristic WD, the accelerator pedal actuation amount ACCP, the vehicle speed SP, the target shift stage SFT, and the oil temperature L.

[0058] For the phase variable PHS, CPU 91 obtains the most recent value calculated in the phase specification process. For the command differential pressure ΔPG, CPU 91 obtains the most recent value calculated during the engagement process. The command differential pressure ΔPG is the command differential pressure variable. For the accelerator pedal actuation amount ACCP, CPU 91 obtains the current value input by the accelerator pedal sensor 59 to the control device 90. The accelerator pedal actuation amount ACCP is the accelerator pedal actuation amount variable. For the vehicle speed SP, CPU 91 obtains the most recent value input by the vehicle speed sensor 58 to the control device 90. For the target shift stage SFT, CPU 91 obtains the most recent calculated value for controlling the automatic transmission 80. The target shift stage SFT is the shift variable. For the oil temperature L, the CPU 91 obtains the most up-to-date value from the oil temperature sensor 56, which is then entered into the control unit 90.The oil temperature L is the oil temperature variable.

[0059] The acceleration characteristic WD is a characteristic used to detect whether the vehicle 500 is accelerating or decelerating. In the present embodiment, the acceleration characteristic WD is "1" when the vehicle 500 is accelerating, "2" when the vehicle 500 is decelerating, and "3" when the vehicle 500 is traveling at a constant speed. In step S10, the CPU 91 acquires the current / last value input by the acceleration sensor 64 to the control device 90. The CPU 91 calculates the acceleration characteristic WD as "1" if the front-to-rear acceleration W is a positive value, i.e., when the vehicle 500 is accelerating. The CPU 91 calculates the acceleration characteristic WD as "2" if the front-to-rear acceleration W is a negative value, i.e., when the vehicle 500 is decelerating. The CPU 91 calculates the acceleration parameter WD as "3" when the front-rear acceleration W is zero, that is, when the vehicle is traveling at a constant speed of 500.The calculation of the acceleration parameter WD by the CPU 91 can be viewed as the acquisition of the acceleration parameter WD by the CPU 91. The acceleration parameter WD is the acceleration variable.

[0060] When CPU 91 acquires the various variables, it proceeds to step S20. The process of step S10 is the acquisition process. In step S20, CPU 91 selects a characteristic map Dx based on the phase variable PHS, which is used to calculate the estimated differential pressure ΔPS. Storage device 95 stores a selection characteristic map that links the phase variable PHS with the characteristic maps Dx for the phases. CPU 91 refers to the selection characteristic map to select, from the characteristic maps Dx for the phases stored in storage device 95, the characteristic map Dx for the phase that corresponds to the phase variable PHS acquired in step S10. When CPU 91 selects the characteristic map Dx, the process proceeds to step S30. The process of step S20 is the selection process.

[0061] In step S30, as preprocessing for the calculation of the estimated differential pressure ΔPS, the CPU 91 replaces the values ​​of the various variables acquired in step S10 in the input variables x(1) to x(6) for input into the characteristic map. Specifically, the CPU 91 inserts the instruction differential pressure ΔPG into input variable x(1). The CPU 91 inserts the acceleration characteristic WD into input variable x(2). The CPU 91 inserts the accelerator pedal actuation amount ACCP into input variable x(3). The CPU 91 inserts the vehicle speed SP into input variable x(4). The CPU 91 inserts the target shift stage SFT into input variable x(5). The CPU 91 inserts the oil temperature L into input variable x(6). The CPU 91 then proceeds to step S40.

[0062] In step S40, the CPU 91 calculates an output variable y by inputting the input variables x(1) to x(6) into the characteristic map Dx selected in step S20. The output variable y is the estimated differential pressure ΔPS.

[0063] The characteristic curve Dx is configured as a fully interconnected feedforward neural network with an intermediate layer. The neural network includes an activation function h(x) as an input-side nonlinear characteristic curve that nonlinearly transforms each of the input-side coefficients wFjk (j = 0 to n, k = 0 to 6) and the output of the input-side linear characteristic curve, which is a linear characteristic curve defined by the input-side coefficient wFjk. In the present embodiment, the hyperbolic tangent "tanh(x)" is shown as an example activation function h(x). Furthermore, the above neural network includes an activation function f(x) as an output-side nonlinear characteristic field that nonlinearly transforms each of the output-side coefficients wSj (j = 0 to n) and the output of the output-side linear characteristic field, which is a linear characteristic field defined by the output-side coefficient wSj.In the present embodiment, the hyperbolic tangent "tanh(x)" is shown as an example activation function f(x). The value n indicates the dimension of the intermediate layer. The input-side coefficient wFj0 ​​is a bias parameter and is a coefficient of the input variable x(0). The input variable x(0) is defined as "1". The output-side coefficient wS0 is a bias parameter.

[0064] The map Dx is a learned model trained using machine learning by employing a power transmission device, including the internal combustion engine 10, the torque converter 20, the automatic transmission 80, the hydraulic circuit 40, and the like, which is mounted on the vehicle 500, before it was installed on the vehicle 500. For learning / training the map Dx, training and instruction data are acquired in advance. That is, the training and instruction data are generated by connecting the power transmission device to the roller dynamometer and simulating the vehicle's operation. During the creation of the training and instruction data, various states of the power transmission device are defined, and the engagement process is performed.The state of the power transmission device is defined by a combination of the vehicle's acceleration state, the accelerator pedal actuation amount (ACCP), the vehicle speed (SP), the gear position, and the oil temperature (L). The vehicle's acceleration state can be defined as acceleration, deceleration, or constant speed driving. That is, for each of the three vehicle acceleration state variants, the engagement process is performed by simulating the situation in which the values ​​of the other parameters are combined in different ways, and the actual differential pressure in each situation (hereinafter referred to as the actual differential pressure, ΔPR) is recorded as the teaching data. Furthermore, at the same time as the actual differential pressure ΔPR is recorded, the values ​​of the various variables that are the input variables for the characteristic map (Dx) are also recorded as training data.At this point, the values ​​of the various variables are recorded in the same way as in step S10. The actual differential pressure ΔPR can be calculated based on the detected value from the oil pressure sensor by connecting the oil pressure sensor to hydraulic circuit 40. Specifically, in hydraulic circuit 40, the oil pressure of the oil channel connected to control port 25a is detected as oil pressure PC of control oil chamber 25. Furthermore, in hydraulic circuit 40, the oil pressure of the oil channel connected to supply port 26a is detected as oil pressure PF of the front oil chamber 26. From these detected values, the actual differential pressure ΔPR can then be calculated.In this way, after acquiring the training and teaching data for each state of the power transmission device, the training of the characteristic map Dx is performed using a set of training and teaching data for each state of the power transmission device. That is, the input and output variables are adjusted with respect to the different states of the power transmission device such that the difference between the value output by the characteristic map Dx with the training data as input and the teaching data, which represents the actual differential pressure ΔPR, is equal to or less than a predetermined value. The training is then considered complete when the upper difference becomes equal to or less than the predetermined value. The training of the characteristic map Dx is performed for the phase corresponding to the characteristic map Dx.

[0065] When CPU 91 calculates the estimated differential pressure ΔPS as the output variable y in step S40, it temporarily terminates the differential pressure estimation process. Then, CPU 91 re-executes the process from step S10, provided the indentation operation is performed. The process from step S40 is the calculation process.

[0066] The processes of the embodiment are described. The CPU 91 calculates the estimated differential pressure ΔPS of the torque converter 20 using the characteristic map data D during the execution of the engagement process. During the calculation, the CPU 91 selects the characteristic map Dx that corresponds to the phase of the engagement process and inputs the various input variables into the selected characteristic map Dx to calculate the estimated differential pressure ΔPS.

[0067] The effects of the embodiment are described below: (1) As indicated by a long double-short dashed line in Fig. As shown in Figure 4, the actual differential pressure ΔPR in the torque converter 20 can differ from the command differential pressure ΔPG. For example, the actual differential pressure ΔPR may begin to change with a delay relative to the command differential pressure ΔPG, or it may change slowly while the command differential pressure ΔPG changes suddenly. Because of these circumstances, for an accurate calculation of the estimated differential pressure ΔPS, it is necessary to adequately incorporate the relationship between the command differential pressure ΔPG and the actual differential pressure ΔPR into the calculation of the estimated differential pressure ΔPS.

[0068] To incorporate the relationship between the command differential pressure ΔPG and the actual differential pressure ΔPR into the calculation of the estimated differential pressure ΔPS, it is conceivable to derive the relational expression that expresses the relationship between the command differential pressure ΔPG and the actual differential pressure ΔPR for use in calculating the estimated differential pressure ΔPS. However, when attempting to derive a relational expression suitable for various driving conditions of the vehicle 500, it is difficult to derive such an expression because its content becomes very complex or lacks precision.Even when attempting to derive the above relational expression for each driving state of vehicle 500, it is necessary to analyze the relationship between the command differential pressure ΔPG and the actual differential pressure ΔPR individually for each driving state of vehicle 500 in order to derive a relational expression suitable for each driving state, which requires time and effort.

[0069] In this respect, when calculating the estimated differential pressure ΔPS using the characteristic map Dx as in the present embodiment, if adequate training and instruction data can be prepared, it is possible to include the relationship between the command differential pressure ΔPG and the actual differential pressure ΔPR in the calculation of the estimated differential pressure ΔPS without the time and effort required to derive a complex relational expression. Furthermore, when calculating the estimated differential pressure ΔPS using the derivative Dx, the accuracy of the estimated differential pressure ΔPS can be ensured if a certain amount of training and instruction data can be prepared. Moreover, in the present embodiment, not only the command differential pressure ΔPG but also the majority of the variables are taken as input variables.Therefore, when calculating the estimated differential pressure ΔPS, it is possible to calculate the estimated differential pressure ΔPS by considering the relationship between the variables and the actual differential pressure ΔPR. This allows for a highly accurate calculation of the estimated differential pressure ΔPS. The estimated differential pressure ΔPS calculated in this way can be used for the learning process of the control program with regard to the oil pressure supply to the lock-up clutch 30, for modifying the control program, for designing the lock-up clutch of another vehicle, and the like.

[0070] (2) In the present embodiment, the characteristics of the method of providing the command differential pressure ΔPG differ depending on the phase of the engagement process. For example, in the first phase Q1, the command differential pressure ΔPG is rapidly increased, while in the second phase Q2, the command differential pressure ΔPG is rapidly decreased. In the third phase Q3, the command differential pressure ΔPG is gradually increased. The method of providing the command differential pressure ΔPG thus differs depending on the phase. Accordingly, the response behavior of the actual differential pressure ΔPR also differs depending on the phase. Therefore, to accurately calculate the estimated differential pressure ΔPS, it is necessary to include the relationship between the command differential pressure ΔPG and the actual differential pressure ΔPR for each phase in the calculation of the estimated differential pressure ΔPS for that phase.

[0071] Since the dedicated characteristic map Dx is used for each phase in the present embodiment, the relationship between the command differential pressure ΔPG and the actual differential pressure ΔPR for each phase can be included in the calculation of the estimated differential pressure ΔPS. This results in a high accuracy in the calculation of the estimated differential pressure ΔPS in each phase.

[0072] (3) The magnitude of the torque acting on the torque converter 20 during the acceleration of the vehicle 500 differs from the magnitude of the torque during the deceleration of the vehicle 500. It is assumed here that when the lock-up clutch 30 is in the engaged state, the torque is transmitted with the same amount of slip during both the acceleration and deceleration of the vehicle 500. During acceleration of the vehicle 500, the power of the internal combustion engine 10 is applied to the input shaft 21 of the torque converter 20. The output shaft 22 then rotates following the input shaft 21 of the torque converter 20. Conversely, during deceleration of the vehicle 500, the power of the internal combustion engine 10 decreases. The input shaft 21 then rotates following the output shaft 22 of the torque converter 20.The magnitude of the torque transmitted between the input shaft 21 and the output shaft 22 during deceleration of the vehicle 500 is an order of magnitude smaller compared to the magnitude of the torque transmitted during acceleration of the vehicle 500, that is, the magnitude of the torque corresponding to the power output of the internal combustion engine 10. Due to these differences, the required magnitude of the differential pressure ΔP also differs, even if the same amount of slip is realized during both acceleration and deceleration of the vehicle 500.

[0073] Even if the slip amount in the target operating state of the lock-up clutch is the same during both acceleration and deceleration of vehicle 500, the final command differential pressure P3 differs. Because the final command differential pressure P3 differs, the nature of the change in command differential pressure ΔPG also differs, such as the differential pressure change rate α that causes the command differential pressure ΔPG to reach the final command differential pressure P3. Since the nature of the change in command differential pressure ΔPG differs, the ratio between the command differential pressure ΔPG and the actual differential pressure ΔPR during vehicle 500 acceleration may also differ from that during vehicle 500 deceleration.

[0074] In the present embodiment, the acceleration characteristic WD is included as one of the input variables for the characteristic map Dx. Based on this, the estimated differential pressure ΔPS can be calculated, taking into account the response behavior of the actual differential pressure ΔPR according to the respective possible changes of the command differential pressure ΔPG during the acceleration of vehicle 500 and during the deceleration of vehicle 500.

[0075] (4) The magnitude of the torque input to the torque converter 20 varies depending on the amount of accelerator pedal input ACCP. Therefore, as in (3) above, the way in which the command differential pressure ΔPG changes differs depending on the amount of accelerator pedal input ACCP. If, as in the present embodiment, the amount of accelerator pedal input ACCP is included as one of the input variables for the characteristic map Dx, it is possible to calculate the estimated differential pressure ΔPS, taking into account the response characteristics of the actual differential pressure ΔPR, according to the way in which the command differential pressure ΔPG changes for each amount of accelerator pedal input ACCP.

[0076] (5) In the lock-up characteristic map, the driving range in which the target operating state of the lock-up clutch 30 is the fully engaged state is defined more towards the higher vehicle speed range than the driving range in which the target operating state is the half-engaged state. That is, the amount of slip, which is defined as the target operating state of the lock-up clutch 30, varies depending on the vehicle speed SP. Therefore, the way in which the final command differential pressure P3 and the command differential pressure ΔPG change until the final command differential pressure P3 is reached differs depending on the vehicle speed SP.Since the vehicle speed SP is included as one of the input variables of the characteristic map Dx in the present embodiment, it is possible to calculate the estimated differential pressure ΔPS taking into account the response behavior of the actual differential pressure ΔPR according to the change behavior of the command differential pressure ΔPG for each vehicle speed SP.

[0077] (6) The amount of slip defined as the target operating state of the lock-up clutch 30 differs depending on the target shift stage SFT, as does the vehicle speed SP. Therefore, the way in which the command differential pressure ΔPG changes also differs depending on the target shift stage SFT. Since, in the present embodiment, the target shift stage SFT is included as one of the input variables of the characteristic map Dx, it is possible to calculate the estimated differential pressure ΔPS, taking into account the response behavior of the actual differential pressure ΔPR, according to the way in which the command differential pressure ΔPG changes for each target shift stage SFT.

[0078] (7) The lower the oil temperature L, the higher the viscosity of the hydraulic oil. Due to the relationship between the oil temperature L and the viscosity of the hydraulic oil, the oil temperature L can influence the response of the actual differential pressure ΔPR to the command differential pressure ΔPG. If the oil temperature L is included as one of the input variables, as in the present embodiment, it is possible to calculate the estimated differential pressure ΔPS taking into account the response of the actual differential pressure ΔPR to the command differential pressure ΔPG corresponding to the oil temperature L.

[0079] Modifications to the present embodiment are described. These modifications are implemented by the following modification of the embodiment described above. The present embodiment and the following modifications can also be implemented in combination within a technically consistent range.

[0080] Part of the differential pressure estimation process can be performed by a computer outside the vehicle. For example, as in Fig. Figure 6 shows that a Server 600 is provided outside the vehicle 500. A configuration can be designed in which the Server 600 performs the selection process and the calculation process of the differential pressure estimation process. In this case, the Server 600 can be designed as one or more processors that execute various processes according to a computer program (software). The Server 600 can be designed as one or more dedicated hardware circuits, such as application-specific integrated circuits (ASICs), that execute at least some of the various processes, or as a circuit that includes a combination of hardware circuits. The processor includes a CPU 602 and memory, such as RAM and ROM 604. The memory stores program code or instructions configured to instruct the CPU 602 to execute the processes.The memory, that is, a computer-readable medium, includes any available medium that a general-purpose or specialized computer can access. Furthermore, the server 600 includes a memory device 606, which is an electrically rewritable non-volatile memory. The memory device 606 stores the characteristic map data D described in the embodiment above, that is, the characteristic map Dx for each phase. The server 600 also includes a communication device 610 for connecting to the environment of the server 600 via an external communication network 700. The CPU 602, the ROM 604, the memory device 606, and the communication device 610 can communicate with each other via an internal bus 608.

[0081] When the server 600 executes the selection and calculation processes of the differential pressure estimation process, the control device 90 of the vehicle 500 includes a communication device 99 for communicating with the outside / environment of the control device 90 via the external communication network 700. The configuration of the control device 90 is the same as in the embodiment described above, except that it includes the communication device 99. Therefore, a detailed description of the control device 90 is omitted. Fig. 6 are the parts that have the same functions as in Fig. 1 have, with the same reference symbols as in Fig. 1 marked. The control device 90, together with the server 600, forms an oil pressure estimating device Z.

[0082] When the selection and calculation processes of the differential pressure estimation process are executed by the server 600, the control device 90 of the vehicle 500 first performs the acquisition process, which is the process of step S10 of the embodiment above. When the control device 90 acquires the various variables in the process of step S10, it sends the values ​​of the acquired variables to the server 600. When the values ​​of the various variables are received, the CPU 602 of the server 600 calculates the estimated differential pressure ΔPS by executing the processes of steps S20, S30, and S40 of the embodiment above. The CPU 602 of the server 600 executes the processes of steps S20, S30, and S40 by running the program stored in the ROM 604.

[0083] When the differential pressure estimation process is carried out by the control device 90 of the vehicle 500 and the server 600 as in the present modification, the CPU 91 and the ROM 93 of the control device 90 of the vehicle 500 and the CPU 602 and the ROM 604 of the server 600 form the execution device.

[0084] The entire differential pressure estimation process can be performed outside of the vehicle 500. For example, if the server 600 is located outside of the vehicle 500, as described in the modification above, the vehicle 500's control unit 90 transmits detection signals from various sensors attached to the vehicle 500 to the server 600. The vehicle 500's control unit 90 also transmits other variables used in the differential pressure estimation process, such as the phase variable PHS and the setpoint SFT, to the server 600. The server 600's CPU 602 then executes the process corresponding to step S10 of the embodiment above to acquire the values ​​of the various variables. Subsequently, the server 600's CPU 602 executes the processes corresponding to steps S20, S30, and S40, as described in the modification above. In this configuration, the server 600 performs the acquisition, selection, and calculation processes.

[0085] If the differential pressure estimation process is executed outside of the vehicle 500, it can be run independently of when the command differential pressure ΔPG is issued during the engagement process. That is, the differential pressure estimation process can be executed on its own, instead of together with the engagement process. For example, by pre-generating time series data of the various variables required to calculate the estimated differential pressure ΔPS, it is possible to calculate the estimated differential pressure ΔPS by executing only the differential pressure estimation process, independent of the engagement process.

[0086] In the embodiment described above, the estimated differential pressure ΔPS can be calculated using the same characteristic map Dx in each phase of the engagement process. It is assumed that in each phase of the engagement process, the behaviors, such as the delay of the actual differential pressure ΔPR relative to the command differential pressure ΔPG, are approximately the same. Therefore, even when the same characteristic map Dx is used in each phase, a certain degree of accuracy regarding the estimated differential pressure ΔPS can be expected.

[0087] The estimated differential pressure ΔPS can be calculated when the operating state of the bypass clutch 30 switches from the engaged state to the open / disengaged state. In this case, a characteristic map Dx can be used that is intended for the case of switching the operating state of the bypass clutch 30 to the disengaged state, or the same characteristic map Dx as for the case of switching the operating state of the bypass clutch 30 to the engaged state can be used. Furthermore, the characteristic map Dx can be used for each phase of the plurality of phases resulting from the division of the time interval for switching the operating state of the bypass clutch 30 to the disengaged state.

[0088] The way in which the phase is specified in the phase specification process is not limited to the example of the embodiment above. For example, when specifying the end time of the third phase Q3, a speed difference can be used, which is a value obtained by subtracting the speed of the output shaft 22 from the speed of the input shaft 21 of the torque converter 20. In particular, the time at which the speed difference reaches a value defined by the target operating condition can be set as the end time of the third phase Q3. In this case, a crank angle sensor can be provided to detect the angle of rotation of the crankshaft of the internal combustion engine 10 in the vehicle 500, and the rotational speed of the crankshaft based on the detected value of the sensor can be set as the rotational speed of the input shaft 21.Furthermore, a rotation angle sensor can be provided in the vehicle 500 to detect the rotation angle of the input shaft of the automatic transmission 80, and the speed of the input shaft of the automatic transmission 80 based on the detected value of the sensor can be set as the speed of the output shaft 22 of the torque converter 20.

[0089] The method of dividing the process into phases during the engagement process is not limited to the example of the embodiment above. For instance, a phase for maintaining the command differential pressure ΔPG at the application differential pressure P1 and a phase for maintaining the command differential pressure ΔPG at the standby differential pressure P2 can be combined into a single phase. If the method of phase division is changed, the characteristic curve Dx can be created for each modified phase.

[0090] The temporal change of the command differential pressure ΔPG during the engagement process is not limited to the example of the embodiment above. For instance, the first, second, and third periods can be variably set depending on the driving state of the vehicle 500, and so on. Furthermore, the application differential pressure P1 and the standby differential pressure P2 can be variably set depending on the driving state of the vehicle 500, and so on. Additionally, the differential pressure change rate α can be changed in the middle of the third period instead of keeping it constant throughout the third period. The temporal change of the command differential pressure ΔPG during the engagement process only needs to be sufficient to engage the lock-up clutch 30.If the content of the time series change of the command differential pressure ΔPG is changed compared to the content of the embodiment described above, the way of dividing the process into phases and the way of determining the phase can be adequately determined according to the change.

[0091] As described in the section on the effects of the embodiment, the magnitude of the torque acting on the torque converter 20 differs depending on the acceleration state of the vehicle 500. Therefore, a characteristic map Dx can be created for each acceleration state of the vehicle 500. In particular, a separate characteristic map Dx can be created for each of the cases in which the vehicle 500 is accelerating, decelerating, and traveling at a constant speed. Then, the estimated differential pressure ΔPS can be calculated using the dedicated characteristic map Dx for each acceleration state of the vehicle 500.

[0092] The accelerator pedal input amount ACCP determined in step S10 is not limited to the most recent value at the time of execution of the process in step S10. For example, the maximum value of the accelerator pedal input amount ACCP in the period from the execution of step S10 at the previous time until the execution of the next step S10 can be recorded. Furthermore, instead of recording the instantaneous value, the average value of the accelerator pedal input amounts ACCP for a specific period can also be recorded. The same applies to the vehicle speed SP and the oil temperature L. In addition, from the same perspective, the acceleration characteristic WD can be calculated based on the average value of the front-to-rear acceleration W for a specific period, and so on, instead of the instantaneous value of the front-to-rear acceleration W.

[0093] The variable assumed to be the command differential pressure variable is not limited to the example of the embodiment above. For instance, a value obtained by multiplying the command differential pressure ΔPG by a correction coefficient or similar factor suitable for accurately calculating the estimated differential pressure ΔPS can be assumed to be the command differential pressure variable. The command differential pressure variable can be any variable that specifies the command differential pressure ΔPG.

[0094] The variable assumed to be the acceleration variable is not limited to the example of the embodiment above. For example, the speed difference, which is a value obtained by subtracting the speed of output shaft 22 from the speed of input shaft 21 of the torque converter 20, can be used as the acceleration variable. When the lock-up clutch 30 is in the partially engaged state, the rotational speed of input shaft 21 becomes higher than the rotational speed of output shaft 22 while the vehicle 500 accelerates. Conversely, the rotational speed of output shaft 22 becomes higher than the rotational speed of input shaft 21 when the vehicle 500 decelerates. Therefore, the sign of the speed difference is reversed between the speed difference during the acceleration of the vehicle 500 and the speed difference during the deceleration of the vehicle 500.Thus, the speed difference can be an index indicating the acceleration state of vehicle 500. As described in the modification above, the speed of input shaft 21 can be the speed of the crankshaft. The rotational speed of output shaft 22 can be the rotational speed of the input shaft of the automatic transmission 80. The acceleration variable can be any variable that specifies the front-to-rear acceleration W of vehicle 500.

[0095] The variable assumed to represent the accelerator pedal actuation amount is not limited to the example of the embodiment above. The throttle valve opening degree of the internal combustion engine 10 has a positive correlation with the accelerator pedal actuation amount ACCP. Therefore, for example, the throttle valve opening degree of the internal combustion engine 10 can be used as the variable for the accelerator pedal actuation amount. In this case, the internal combustion engine 10 can be equipped with an opening degree sensor that detects the throttle valve opening degree. The accelerator pedal actuation amount variable can be any variable that specifies the accelerator pedal actuation amount ACCP.

[0096] The variable assumed to be the vehicle speed variable is not limited to the example of the embodiment above. For example, the rotational speed of the output shaft of the automatic transmission 80 can be assumed to be the vehicle speed variable. In this case, a rotation angle sensor can be provided in the vehicle 500 to detect the rotational angle of the output shaft of the automatic transmission 80, and the rotational speed of the output shaft 22 of the automatic transmission 80 can be calculated based on the detected value of the sensor. The vehicle speed variable can be any variable that indicates the vehicle speed SP.

[0097] The variable assumed to be the switching variable is not limited to the example of the embodiment above. For example, the ratio between the rotational speed of the input shaft of the automatic transmission 80 and the rotational speed of the output shaft, that is, the actual gear ratio, can be assumed to be the switching variable. The switching variable can be any variable that specifies the gear ratio of the transmission.

[0098] The variable assumed to be the oil temperature variable is not limited to the example of the embodiment above. For example, the oil temperature L can be divided into a plurality of stages / levels, and a value indicating such a stage can be assumed to be the oil temperature variable. The oil temperature variable can be any variable that indicates the oil temperature L.

[0099] Similar to the oil temperature variable in the modification, a plurality of levels can be defined for the other input variables, specifying the range, and a value indicating such a level can be adopted. The types of input variables are not limited to the examples of the embodiment above. Other variables can also be used as input variables instead of or in addition to the variables mentioned in the embodiment above. Furthermore, the number of input variables can be reduced compared to the embodiment described above. The number of input variables can be two or more. The input variables need only include the command differential pressure variable as one of the two or more input variables.

[0100] The acceleration variable, the throttle pedal variable, the vehicle speed variable, the shift variable, and the oil temperature variable are not strictly required as input variables. Even without these variables, the estimated differential pressure ΔPS can be calculated with sufficiently high accuracy as long as two or more variables, including the command differential pressure variable, are assumed as input variables.

[0101] A variable other than those mentioned above can be used as the input variable. For example, a variable indicating the degree of aging of hydraulic circuit 40 can be used as the input variable. In particular, the total mileage of the vehicle 500 or a similar value can be used as the variable indicating the degree of aging of hydraulic circuit 40. The response characteristic of the actual differential pressure ΔPR to the command differential pressure ΔPG can change depending on the degree of aging of hydraulic circuit 40. Therefore, if the variable indicating the degree of aging of hydraulic circuit 40 is used as one of the input variables, the estimated differential pressure ΔPS can be calculated taking into account the degree of aging of hydraulic circuit 40.

[0102] The variable assumed to be the estimated oil pressure variable is not limited to the example of the embodiment above. For example, an estimated value of the difference in flow rate, calculated by converting the bridging differential pressure ΔP into the difference in the hydraulic oil flow rate, can be assumed to be the estimated oil pressure variable. The estimated oil pressure variable can be a quantity that specifies the estimated differential pressure ΔPS.

[0103] The configuration of the characteristic map Dx is not limited to the example of the embodiment above. For example, the number of intermediate layers in the neural network can be two or more.

[0104] A recurrent neural network can be assumed as the neural network. In this case, the values ​​of past input variables are included when the value of the output variable is recalculated, which is suitable for calculating the estimated differential pressure ΔPS while incorporating past history.

[0105] The method of obtaining the training data and the teaching data used to train the characteristic map Dx is not limited to the example of the embodiment above. For example, a vehicle with the same specifications as vehicle 500 can actually be driven to obtain the training and teaching data.

[0106] As shown in the following equation (1), the bridging differential pressure ΔP can be obtained by subtracting the mean of the oil pressure PF of the front oil chamber 26 and the oil pressure PB of the rear oil chamber from the oil pressure PC of the control oil chamber 25. ΔP=PC−((PF+PB) / 2)

[0107] The bridging differential pressure ΔP only needs to specify the differential pressure between the control oil chamber 25 and the front oil chamber 26.

[0108] The overall configuration of vehicle 500 is not limited to the example of the embodiment above. For example, instead of or in addition to the internal combustion engine 10, an (electric) motor / generator can be provided as the drive source for vehicle 500. A continuously variable transmission (CVT) can be provided as the automatic transmission.

[0109] The type of connection between the torque converter 20, the internal combustion engine 10, and the automatic transmission 80 is not limited to the example of the embodiment above. For example, the input shaft of the automatic transmission 80 can be the output shaft 22 of the torque converter 20 itself.

[0110] The number of first friction discs 32 and second friction discs 34 is not limited to the example of the embodiment above. The number of first friction discs 32 and second friction discs 34 can each be one or more. The content of the bridging characteristic map is not limited to the example of the embodiment above. The bridging characteristic map only needs to contain content that enables adequate switching of the operating state of the bridging clutch 30.

Claims

[1] Oil pressure estimating device, which specifies as the simulation target a torque converter (20) comprising two oil chambers (25, 26) and a lock-up clutch (30), wherein the lock-up clutch (30) is switched between an engaged state and a disengaged state depending on a differential pressure between the two oil chambers (25, 26), and which, based on a command differential pressure which is a setpoint of the differential pressure for the torque converter (20), calculates an estimated differential pressure which is an estimated value of the differential pressure generated in the torque converter (20), wherein the oil pressure estimating device comprises: a storage device (95) designed to store map data (D) defining a map (Dx), wherein the map (Dx) outputs, in response to the input of an input variable, an estimated differential pressure variable, which is a variable indicating the estimated differential pressure, wherein the map (Dx) includes a command differential pressure variable, which is a variable indicating the command differential pressure as one of a plurality of input variables; and an execution device (91, 93) designed to perform a detection process for acquiring a value of the input variables and a calculation process for inputting the value of the input variables acquired by the detection process into the characteristic map (Dx) in order to calculate a value of the output variables, characterized by , that the characteristic map (Dx) was trained by machine learning; the storage device (95) and the execution device (91, 93) are provided in a vehicle (500) equipped with the torque converter (20); the storage device (95) is designed to store a plurality of characteristic maps (Dx) for respective phases, which are obtained by dividing a period of time from the start of switching the bridging clutch (30) into the engaged state until the start of switching the bridging clutch (30) into the disengaged state into a plurality of phases; the execution device (91, 93) is designed to perform the acquisition process, a selection process and the calculation process; the acquisition process includes acquiring the value of the input variable when the command differential pressure is output to the torque converter (20); the selection process involves selecting the characteristic map (Dx) from the characteristic maps (Dx) for the respective phases, where the characteristic map (Dx) corresponds to a phase in which the command differential pressure is output; and The calculation process involves calculating the value of the output variable by entering the value of the input variable acquired through the acquisition process into the characteristic field (Dx) selected through the selection process. [2] Oil pressure estimating device according to claim 1, wherein one of the input variables is an acceleration variable which is a variable that indicates an acceleration of the vehicle (500). [3] Oil pressure estimating device according to claim 1 or 2, wherein one of the input variables is an accelerator pedal variable which specifies an actuation value of an accelerator pedal of the vehicle (500). [4] Oil pressure estimating device according to one of claims 1 to 3, wherein one of the input variables is a vehicle speed variable which is a variable that indicates a driving speed of the vehicle (500). [5] Oil pressure estimating device according to one of claims 1 to 4, wherein one of the input variables is a switching variable which is a variable that specifies a gear ratio of a transmission of the vehicle (500). [6] Oil pressure estimating device according to one of claims 1 to 5, wherein one of the input variables is an oil temperature variable which is a variable indicating a temperature of hydraulic oil supplied to the torque converter (20). [7] Non-transitory storage medium that stores instructions that can be executed by one or more processors contained in an execution device of an oil pressure estimating device and that cause the one or more processors to perform functions that include: a process of capturing the value of an input variable; and a calculation process of entering the value of the input variables acquired by the acquisition process into a characteristic map (Dx) in order to calculate a value of an output variable, wherein the oil pressure estimating device defines as a simulation target a torque converter (20) which includes two oil chambers (25, 26) and a lock-up clutch (30), wherein the lock-up clutch (30) is switched between an engaged state and a disengaged state depending on a differential pressure between the two oil chambers (25, 26), and the oil pressure estimating device includes a storage device (95) and the execution device (91, 93), and wherein the storage device (95) stores map data (D) defining a map (Dx), wherein the map (Dx) outputs, in response to the input of the input variables as the output variable, an estimated differential pressure variable, which is a variable that specifies an estimated differential pressure, which is an estimated value of the differential pressure generated in the torque converter (20), wherein the map (Dx) includes a command differential pressure variable as one of a plurality of input variables, which is a variable that specifies a command differential pressure, which is a setpoint of the differential pressure for the torque converter (20), characterized by , that the characteristic map (Dx) was trained by machine learning; the storage device (95) and the execution device (91, 93) are provided in a vehicle (500) equipped with the torque converter (20); the storage device (95) is designed to store a plurality of characteristic maps (Dx) for respective phases, which are obtained by dividing a period of time from the start of switching the bridging clutch (30) into the engaged state until the start of switching the bridging clutch (30) into the disengaged state into a plurality of phases; the execution device (91, 93) is designed to perform the acquisition process, a selection process and the calculation process; the acquisition process includes acquiring the value of the input variable when the command differential pressure is output to the torque converter (20); the selection process involves selecting the characteristic map (Dx) from the characteristic maps (Dx) for the respective phases, where the characteristic map (Dx) corresponds to a phase in which the command differential pressure is output; and The calculation process involves calculating the value of the output variable by entering the value of the input variable acquired through the acquisition process into the characteristic field (Dx) selected through the selection process.

Images