Electric vehicle control method and electric vehicle control device

Abstract

To realize a desired acceleration and deceleration state in an electric vehicle in which a winding field type synchronous motor is adopted as a driving motor.SOLUTION: An electric vehicle control method includes: a first torque target value calculating process of calculating a first torque target value based on vehicle information; a disturbance torque estimation process of estimating disturbance torque Td which acts on a motor 101; a second torque target value calculating process of calculating a second torque target value Tm2* from the first torque target value Tm1* and the disturbance torque Td; and a field current command value calculating process of calculating a field current command value if* based on the second torque target value Tm2*. Especially, in the disturbance torque estimation process, an actual field current responce if^ as a response of actual field current for the field current command value if* is defined, a correction second torque target value Tm2_pr* is acquired based on the actual field current response if^, and the disturbance torque Td is calculated based on the correction second torque target value Tm2_pr*.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to an electric vehicle control method and an electric vehicle control device. [Background technology]

[0002] Patent Document 1 describes an electric vehicle control method that estimates the disturbance torque (torque caused by the influence of gradients, etc.) acting on the drive motor and controls the torque of the drive motor based on the estimated disturbance torque. In particular, Patent Document 1 calculates a basic torque target value (torque table target value) determined based on the required driving force for the electric vehicle (such as the amount of accelerator pedal operation by the occupant) and the vehicle speed, and calculates a corrected torque target value (first torque target value) by correcting the basic torque target value based on the estimated disturbance torque. Then, by controlling the power supplied to the drive motor based on this corrected torque target value, a desired acceleration and deceleration state in the electric vehicle is achieved. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] Patent No. 6135775 [Overview of the project] [Problems that the invention aims to solve]

[0004] On the other hand, the inventors of the present invention focused on the fact that in a motor control system employing a so-called wound-field synchronous motor (EESM: Electrically Excited Synchronous Motor) as the drive motor, applying the control described in Patent Document 1 as is would not ensure sufficient accuracy in estimating disturbances. Specifically, in a control system for a wound-field synchronous motor, variations in the motor torque response occur depending on the motor operating point (especially the control state of the field current), leading to errors in the calculation of disturbance torque estimation. As a result, errors also occur in the calculation of the correction torque target value determined from this disturbance torque, resulting in the problem that the desired acceleration and deceleration state in electric vehicles cannot be achieved.

[0005] Therefore, the object of the present invention is to provide an electric vehicle control method and an electric vehicle control device that can achieve a desired acceleration and deceleration state in an electric vehicle in which a wound-field synchronous motor is used as the drive motor. [Means for solving the problem]

[0006] According to one aspect of the present invention, there is a method for controlling an electric vehicle equipped with a drive motor consisting of a wound-field motor as a driving source. This electric vehicle control method includes a first torque target value calculation step of calculating a first torque target value based on vehicle information, a disturbance torque estimation step of estimating the disturbance torque acting on the motor, a second torque target value calculation step of calculating a second torque target value from the first torque target value and the disturbance torque, and a field current command value calculation step of calculating a field current command value based on the second torque target value.

[0007] Then, in the disturbance torque estimation process, the real field current response is defined as the response of the real field current to the field current command value, and based on the real field current response, a corrected second torque target value, which is a correction value of the second torque target value, is determined, and the disturbance torque is calculated based on the corrected second torque target value. Furthermore, in the disturbance torque estimation process, a first torque estimate corresponding to the actual torque response according to the first torque target value is calculated based on a speed parameter proportional to the vehicle speed of the electric vehicle. The calculated second torque target value is used as input, and a response model reflecting the response characteristics of the field current is applied to obtain the actual field current response. Based on the actual field current response, a second torque estimate corresponding to the actual torque response according to the corrected second torque target value is calculated, and the disturbance torque is calculated from the first torque estimate and the second torque estimate. [Effects of the Invention]

[0008] According to the present invention, in an electric vehicle employing a wound-field synchronous motor as the drive motor, a desired acceleration and deceleration state can be achieved. [Brief explanation of the drawing]

[0009] [Figure 1] Figure 1 is a block diagram illustrating the system configuration of an electric vehicle on which an electric vehicle control method according to an embodiment of the present invention is implemented. [Figure 2] Figure 2 shows an example of an accelerator opening-torque table. [Figure 3] Figure 3 shows a model of the power transmission system of an electric vehicle. [Figure 4] Figure 4 is a block diagram showing the configuration of the second torque target value calculation unit. [Figure 5] Figure 5 is a block diagram showing the configuration of the disturbance torque estimation unit. [Figure 6] Figure 6 is a block diagram showing the configuration of the response delay processing unit. [Figure 7] Figure 7 is a timing chart showing the first control results for the examples and comparative examples. [Figure 8] Figure 8 is a timing chart showing the second control results for the examples and comparative examples. [Modes for carrying out the invention]

[0010] The electric vehicle control methods according to each embodiment of the present invention will be described below with reference to the drawings.

[0011] [First Embodiment] Figure 1 is a block diagram illustrating the system configuration of an electric vehicle 10 on which an electric vehicle control method is implemented. In this embodiment, the concept of an electric vehicle 10 includes an electric vehicle (EV) or a hybrid electric vehicle (HEV) equipped with at least one electric motor as a driving source.

[0012] As shown in the figure, the system of the electric vehicle 10 mainly includes a drive motor 101, an inverter 102, a battery 106, and a power transmission system (reduction gear 103, drive shaft 104, and drive wheels 105).

[0013] The drive motor 101 is composed of an electrically excited synchronous motor (EESM). Specifically, the drive motor 101 generates driving force using alternating current supplied from the inverter 102 and transmits this driving force to each drive wheel 105 via the reduction gear 103 and drive shaft 104. In addition, the drive motor 101 recovers kinetic energy as electrical energy based on the regenerative braking force received from each drive wheel 105 when the vehicle is running.

[0014] The inverter 102 is composed of two pairs of switching elements (power semiconductor elements such as IGBTs and MOS-FETs) for each phase for stator current control. By switching the switching elements ON / OFF in response to the drive signal, it converts the DC current supplied from the battery 106 into AC current and supplies the desired current to the drive motor 101. The inverter 102 also connects two pairs (a total of four) of switching elements (power semiconductor elements such as IGBTs and MOS-FETs) to both ends of the rotor winding for rotor current control. By switching the switching elements ON / OFF in response to the drive signal, it supplies the desired current from the battery 106 to the rotor winding of the drive motor 101. If the direction of current flowing to the rotor winding is only unidirectional, two of the two pairs of switching elements located diagonally opposite each other may be replaced with diodes. Furthermore, during regenerative operation of the drive motor 101, the inverter 102 converts the AC current generated by the drive motor 101 into DC current and supplies it to the battery 106.

[0015] The battery 106 consists of an on-board secondary battery (such as a stacked lithium-ion battery) that enables the supply (discharge) of drive power during traction operation of the drive motor 101 and the acceptance (charging) of regenerative power during regenerative operation.

[0016] The motor controller 107 receives, as digital signals, signals indicating various vehicle variables such as vehicle speed V, accelerator opening APO, rotor phase α of the drive motor 101, and stator current (three-phase alternating current i us , i vs , i ws ), and rotor current (field current i f ), and generates a PWM signal based on the digital signals. Further, the motor controller 107 generates drive signals (D uu * , D ul * , D vu * , D vl * , D wu * , D wl * , D fu * , D fl * ) of the inverter 102 according to the generated PWM signal. The detailed configuration of the motor controller 107 will be described later.

[0017] Also, a current sensor 109 and a magnetic position detector 112 are provided in the system of the electric vehicle 10. The current sensor 109 detects the current flowing through the stator winding of the drive motor 101 (particularly the three-phase alternating current i u , i v , i w ) and the current flowing through the rotor winding (field current i f ) and outputs them to the motor controller 107. Since the sum of the three-phase alternating currents i u , i v , i w is 0, any two-phase currents may be detected, and the remaining one-phase current may be obtained by calculation. The magnetic position detector 112 is, for example, a resolver or an encoder, detects the rotor position (rotor phase α) of the drive motor 101, and outputs A-phase, B-phase, and Z-phase pulses corresponding to the rotor position to the motor controller 107.

[0018] The following describes the configuration of the motor controller 107 in detail. The motor controller 107 includes a pulse counter 113, an angular velocity calculator 114, a look-ahead compensation unit 115, a three-phase / dq coordinate converter 111, a first torque target value calculation unit S115, a second torque target value calculation unit S116, a current command value calculation unit 117, a current control unit 118, a non-interference control unit 119, a voltage command value calculation unit 120, and a dq / three-phase coordinate converter 121.

[0019] The pulse counter 113 calculates the electrical angle θ of the drive motor 101 from the A-phase, B-phase, and Z-phase pulses input from the magnetic position detector 112. re The angular velocity calculator 114 calculates the electrical angle θ. re Using the input, the electrical angle θ re The electrical angular velocity ω is the rate of change over time. re The angular velocity calculator 114 calculates the electrical angular velocity ω. re By dividing by the number of motor pole pairs p, the mechanical angular velocity ω rm Perform the calculation.

[0020] The look-ahead compensation unit 115 controls the electrical angle θ. re and electrical angular velocity ω re Enter the following, and then lookahead compensated electrical angle θ re The calculation is performed. More specifically, the look-ahead compensation unit 115 calculates the electrical angle θ. re , the electrical angular velocity ω re By adding the multiplicative value of the dead time of the control system, the look-ahead compensated electrical angle θ is obtained. re To find '.

[0021] The three-phase / dq coordinate converter 111 uses the electrical angle θ. re Using the following equation (1), the three-phase AC current i detected by the current sensor 109 is calculated. us ,i vs ,i ws dq axis current i d ,i q Convert to.

[0022]

number

[0023] Note that the current sensor 109 detects two-phase current (for example, u-phase current i us ,v phase current i vs If the configuration only detects the currents of the two phases detected, the current of the remaining one phase (for example, the w-phase current i) is calculated based on the following equation (2). ws Applying equation (1), the dq axis current i d ,i q Perform the calculation.

[0024]

number

[0025] The first torque target value calculation unit S115 takes the vehicle speed V and accelerator opening APO as inputs and refers to the accelerator opening-torque table shown in Figure 2 to calculate the first torque target value T m1 * This calculates the first torque target value T. m1 * This is calculated as a torque value corresponding to the required output for the drive motor 101 (required driving force for the electric vehicle 10).

[0026] The vehicle speed V can be obtained by communication from controllers other than the motor controller 107 mounted on the electric vehicle 10, such as the meter and brake controller. In addition, the motor controller 107 can obtain the mechanical angular velocity ω of the drive motor 101. rm Alternatively, a configuration may be adopted in which the vehicle speed v (m / s) is calculated by multiplying the tire radius r by the tire radius and dividing the resulting value by the final gear ratio, and then multiplying this by 3600 / 1000 to convert the units and calculate the vehicle speed V (km / h). Furthermore, the accelerator opening APO (%) is obtained from an accelerator opening sensor (not shown) or by communication from another controller such as a vehicle controller (not shown).

[0027] The second torque target value calculation unit S116 calculates the mechanical angular velocity ω rm and the first torque target value T m1 * Using this as input, the second torque target value Tm2 * The second torque target value T is calculated. m2 * The first torque target value T is determined from the viewpoint of achieving the desired driving state (acceleration / deceleration state) of the electric vehicle 10, taking into account the effects of disturbances acting on the drive motor 101. m1 * This is the corrected torque.

[0028] Specifically, the second torque target value calculation unit S116 calculates the disturbance torque T, which indicates the disturbance acting on the drive motor 101. d The second torque target value calculation unit S116 calculates the first torque target value T m1 * The disturbance torque estimate T d_e By correcting with this, the second torque target value T m2 * This is calculated. Details of the processing in the second torque target value calculation unit S116 will be described later.

[0029] The current command value calculation unit 117 calculates the second torque target value T m2 * , mechanical angular velocity ω rm , and DC voltage V dc Using this as input, the system references map data stored in memory beforehand to determine the dq axis current command value i d * ,i q * and field current command value i f * The calculation is performed. Note that the DC voltage V dc (V) can be determined based on the detected value of a voltage sensor (not shown) provided on the connection line of the battery 106, or the estimated power supply voltage value estimated by a battery controller (not shown).

[0030] The current control unit 118 controls the dq axis current command value i d * ,i q * and field current command value i f * The input is the first dq axis voltage command value v d_dsh ,v q_dshand the first field voltage command value v f_dsh are calculated. More specifically, the current control unit 118 calculates the d-axis current command value i d * , the q-axis current command value i q * , and the field current command value i f * respectively, to cause the measured d-axis current i d , the q-axis current i q , and the field current i f to follow with a desired responsiveness without steady-state deviation, and calculates the first d-axis voltage command value v d_dsh , the first q-axis voltage command value v q_dsh , and the first field voltage command value v f_dsh .

[0031] The non-interference control unit 119 takes the electrical angular velocity ω re and the measured d-axis current i d , the q-axis current i q , and the field current i f as inputs, and calculates the non-interference voltages v d_dcpl ,v q_dcpl ,v f_dcpl . The non-interference voltages v d_dcpl ,v q_dcpl ,v f_dcpl are correction voltages for canceling the interference voltages between the d-axis, q-axis, and field (f-axis). Note that instead of the measured d-axis current i d , the q-axis current i q , and the field current i f , a configuration may be adopted in which the d-axis current command value i d * , the q-axis current command value i q * , and the field current command value i f * are used as inputs to calculate the non-interference voltages v d_dcpl ,v q_dcpl ,v f_dcpl . In particular, in this case, the d-axis current command value i d * , the q-axis current command value i q * , and the field current command value i f *From these, the d-axis current reference response i is obtained as a response current based on the reference response characteristics. d_ref , q-axis current reference response i q_ref , and field current reference response i f_ref Then, we determine the non-interference voltage v from these current reference responses. d_dcpl ,v q_dcpl ,v f_dcpl Perform the calculation.

[0032] The voltage command value calculation unit 120 calculates the first d-axis voltage command value v d_dsh , 1st q-axis voltage command value v q_dsh , and the first field voltage command value v f_dsh The d-axis non-interference voltage v d_dcpl q-axis non-interference voltage v q_dcpl f-axis non-interference voltage v f_dcpl By adding this, the second d-axis voltage command value v d_dsh2 , 2nd q-axis voltage command value v q_dsh2 , and the second field voltage command value v f_dsh2 The voltage command value calculation unit 120 then calculates the final d-axis voltage command value v. d * q-axis voltage command value v q * , and field voltage command value v f * Output as follows.

[0033] The dq / three-phase coordinate converter 121 calculates the look-ahead compensated electrical angle θ. re Using ', the dq axis voltage command value v d * ,v q * For this, perform a coordinate transformation based on the following equation (3) to obtain the three-phase voltage command value v u * ,v v * ,v w * We seek.

[0034]

number

[0035] Next, the details of the processing in the second torque target value calculation unit S116 will be described. In particular, in the second torque target value calculation unit S116 of this embodiment, the motor torque T m From the mechanical angular velocity ω rm Transfer characteristics G p (s) is defined, and the transfer characteristic G p Using (s), estimate the disturbance torque T d_e We calculate this. Therefore, first, this transfer characteristic G p (s) will be explained.

[0036] <Transmission Characteristics G p (s)> Figure 3 shows a model of the power transmission system of the electric vehicle 10. The definitions of each parameter, including those already explained, are shown below.

[0037] J m :Motor Inertia J w Inertia of the drive wheels M: Vehicle weight K D Torsional rigidity of the drive system K T : Coefficient related to the friction between the tire and the road surface N: Overall Gear Ratio r: Tire load radius ω rm : Motor mechanical angular velocity T m : Motor Torque T D Torque of the drive wheels F: Forces applied to the vehicle V:Vehicle speed ω w Angular velocity of the drive wheels

[0038] From Figure 3, the equations of motion for the electric vehicle 10 are expressed by the following equations (4) to (8).

[0039]

number

[0040] Based on the above equations of motion (4) to (8), the transfer characteristic G p (s) can be expressed as shown in equations (9) to (17) below.

[0041]

number

[0042] Note that "s" in equation (9) represents the Laplace operator. Furthermore, by transforming equation (9), we obtain the following equation (18).

[0043]

number

[0044] Note that the coefficients a'1 to a'3 and b'0 to b'2 in equation (18) are determined by the coefficients a1 to a4 and b0 to b3 specified in equations (10) to (17).

[0045] Here, the transfer characteristic G p When we examine the poles and zeros of (s), they show extremely close values. That is, this means that α and β in equation (18) are very close to each other. Therefore, by performing pole-zero cancellation (approximation process where α=β) on equation (18), the transfer characteristic G p (s) can be expressed as a rational function with a quadratic numerator and a cubic denominator, as shown in equation (19) below.

[0046]

number

[0047] Now, if we rewrite equation (19), the transfer characteristic G p (s) can be expressed as shown in equation (20).

[0048]

number

[0049] However, in equation (20) "M p " is a constant that does not depend on the Laplace operator s. Also, "ζ z '', ζ p "ω" z ", and "ω p The value of '' is determined by the following equations (21) to (24).

[0050]

number

[0051] <Second Torque Target Value Calculation Process> Figure 4 is a block diagram showing the configuration of the second torque target value calculation unit S116. As shown in the figure, the second torque target value calculation unit S116 includes a disturbance torque estimation unit S401, a gain processing unit S402, and an addition unit S403.

[0052] The disturbance torque estimation unit S401 calculates the mechanical angular velocity ω rm and the second torque target value T m2 * Using as input, estimate disturbance torque T d_e Perform the calculation.

[0053] Figure 5 is a block diagram showing the configuration of the disturbance torque estimation unit S401. As shown in the figure, the disturbance torque estimation unit S401 includes a first filter processing unit S501, a response delay processing unit S502, a second filter processing unit S503, and an adder S504. In this embodiment, the first filter processing unit S501 functions as the first torque estimate calculation unit, and the response delay processing unit S502 and the second filter processing unit S503 function as the second torque estimate calculation unit.

[0054] The first filter processing unit S501 determines the mechanical angular velocity ω which indicates the vehicle speed V of the electric vehicle 10. rm A predetermined first filter is applied to the first torque estimate T m1 ^ The following is calculated. In particular, the first filter is the motor torque T mentioned above. m From the mechanical angular velocity ω rmTransfer characteristics G p H(s) / G is defined as (s) as the denominator and the low-pass filter H(s) as the numerator. p It is composed of (s). Note that the order of the low-pass filter H(s) is the first filter H(s) / G p The order of the denominator of (s) is determined to be equal to or greater than the order of the numerator. The first torque estimate T is calculated in this manner. m1 ^ This is the first torque target value T m1 * The actual motor torque T output relative to this is... m (This corresponds to the actual torque response.)

[0055] The response delay processing unit S502 sets the second torque target value T m2 * (In particular, the second torque target value T) m2 * The previous value of the input is used to determine the response delay processing value T. m2_pr * We seek.

[0056] Figure 6 is a block diagram showing the configuration of the response delay processing unit S502. As shown in the figure, the response delay processing unit S502 includes a current command value calculation unit S601, a response model unit S602, and a response delay processing value calculation unit S603.

[0057] The current command value calculation unit S601 calculates the second torque target value T m2 * , mechanical angular velocity ω rm , and DC voltage V dc Using this as input, the d-axis current command value i is calculated using the same logic as the calculation in the current command value calculation unit 117 described above. d * q-axis current command value i q * , and field current command value i f * Perform the calculation.

[0058] The response model unit S602 determines the d-axis current command value i d * q-axis current command value i q *, and field current command value i f * Using these as inputs, a low-pass filter is applied to them to estimate the d-axis current i d ^ , q-axis current estimate i q ^ , and estimated field current i f ^ Calculate the d-axis current estimate i. d ^ , q-axis current estimate i q ^ , and estimated field current i f ^ This corresponds to an estimated value of the actual current value (actual current response) output for each command value.

[0059] In a wound-field synchronous motor, there is a significant difference in inductance characteristics between the stator and the rotor. Therefore, the d-axis current i d and q-axis current i q And, field current i f There is a difference in the current response to the command value between and . As a result, the q-axis current i q and field current i f The actual torque response determined based on the product of the drive motor 101 (especially when the field current i f The control state will vary. Taking this into consideration, the response model section S602 of this embodiment sets the d-axis current i d , q-axis current i q , and field current i f Using the individually configured response models S6021, S6022, and S6023 for each, the estimated d-axis current i d ^ , q-axis current estimate i q ^ , and estimated field current i f ^ Perform the calculation.

[0060] More specifically, in the response model section S602, the d-axis current command value i d * The d-axis response model S6021 simulates the response of the actual d-axis current to the q-axis current command value i. q *Q-axis response model S6022 that simulates the response of the actual q-axis current to, and field current command value i f * A field current response model S6023, which simulates the response of the actual field current to the given parameters, is individually set. The d-axis response model S6021, the q-axis response model S6022, and the field current response model S6023 are each constructed using transfer functions that reflect the actual d-axis current response, the actual q-axis current response, and the actual field current response, based on the motor control system in this embodiment. In particular, the d-axis response model S6021, the q-axis response model S6022, and the field current response model S6023 each have a time constant τ d ,τ q ,τ f It is composed of a first-order low-pass filter having τ. d , τ q , and τ f These are the d-axis current i in the motor control system, respectively. d , q-axis current i q , and field current i f It is set to an appropriate value based on the response characteristics. In particular, the time constant τ in the field current response model S6023 f This takes into account the difference in inductance characteristics between the stator and rotor as described above, and the time constant τ in the d-axis response model S6021. d and the time constant τ in the q-axis response model S6022 q It is set to a value greater than this. This results in the stator current, i, being the dq-axis current. d ,i q and the rotor current, which is the field current i f It is possible to determine each current estimate (each current response) that appropriately reflects the differences in response characteristics.

[0061] The response delay processing value calculation unit S603 calculates the d-axis current estimate i d ^ , q-axis current estimate i q ^ , and estimated field current i f ^ Using the input, the system references map data stored in memory beforehand to calculate a response delay value T with torque units. m2_pr *The response delay processing value T is calculated. m2_pr * The estimated value of the d-axis current i d ^ , q-axis current estimate i q ^ , and estimated field current i f ^ These are parameters corresponding to the command torque when these are considered as the actual d-axis current response, actual q-axis current response, and actual field current response, respectively. That is, the response delay processing value T m2_pr * is the d-axis current i d , q-axis current i q , and field current i f The second torque target value T was determined considering the difference in response characteristics. m2 * This is the correction value.

[0062] Instead of using a configuration that references map data, the response delay processing value T can be determined by applying each current estimate to the following equation (25), which defines the relationship between torque and current in a wound-field synchronous motor. m2_pr * You may adopt a configuration that calculates it.

[0063]

number

[0064] The parameters included in equation (25), including those already explained, are defined as follows.

[0065] p: polar logarithm M: Mutual inductance between stator and rotor L d d-axis self-inductance L q q-axis self-inductance

[0066] Returning to Figure 5, the second filter processing unit S503 processes the response delay processing value T, which corresponds to the torque command value obtained from each current estimate. m2_pr *The second torque estimate T is obtained by processing it with the same low-pass filter H(s) as the first filter processing unit S501. m2 ^ This determines the second torque estimate T. m2 ^ This is the response delay processing value T. m2_pr * This value is calculated as the value corresponding to the actual torque response when the torque command value is set to [value].

[0067] The addition unit S504 calculates the second torque estimate T. m2 ^ From the first torque estimate T m1 ^ Subtracting this gives the estimated disturbance torque T. d_e The result is calculated and output to the gain processing unit S402.

[0068] In this embodiment, disturbances include air resistance, modeling errors due to fluctuations in vehicle mass caused by the number of occupants and load, tire rolling resistance, gradient resistance, and the first torque estimate T. m1 ^ and the second torque estimate T m2 ^ This includes various disturbance factors such as phase shifts between them. Furthermore, the dominant factor among these disturbance factors differs depending on the driving conditions, including the driving environment of the electric vehicle 10. In contrast, the disturbance torque estimation unit S401 of this embodiment sets the second torque target value T m2 * and mechanical angular velocity ω rm From the vehicle model, transmission characteristics G p Based on (s), the disturbance torque estimate T d_e To calculate this, the disturbance torque estimate T comprehensively suggests the above disturbance factors. d_e It is possible to determine this.

[0069] Returning to Figure 4, the gain processing unit S402 processes the disturbance torque estimate T. d_e By multiplying this by a predetermined gain, the torque correction value T is obtained. c We will determine the first torque target value T. m1 * Torque correction value T cAdding this to the second torque target value T m2 * The current command value is calculated and output to the current command value calculation unit 117.

[0070] By performing the above process, the stator current (dq axis current i) d ,i q The difference in the response characteristics of the operating point (especially the field current i) f Even if variations in torque response occur depending on the control state, the decrease in disturbance estimation accuracy can be suppressed.

[0071] [Second Embodiment] In the electric vehicle control of this embodiment, the response delay processing unit S502 (see Figure 5) sets the second torque target value T m2 * Instead, the measured d-axis current i d , q-axis current i q , and field current i f The input is the response delay processing value T. m2_pr * The difference lies in the fact that it seeks a specific result, but all other processes are performed similarly.

[0072] In particular, the response delay processing unit S502 of this embodiment calculates the response delay processing value S603 based on the measured d-axis current i d , q-axis current i q , and field current i f Then, referencing predetermined map data or equation (25), the response delay processing value T is determined. m2_pr * This determines the response delay processing value T. In other words, the response delay processing unit S502 of this embodiment directly calculates the response delay processing value T from the above-mentioned current measurement values ​​without going through the current command value calculation unit S601 and the response model unit S602 (see Figure 6). m2_pr * To define.

[0073] This allows for the acquisition of parameters suggesting the actual stator current response and the actual field current response using a simpler logic, while appropriately processing the response delay value T. m2_pr * This allows for the implementation of control logic for performing calculations.

[0074] [Control result] Hereinafter, the control results by the electric vehicle control method (example) of each embodiment will be described while comparing with the control results by the reference example. As the reference example, the second filter processing unit S503 directly determines the second torque estimated value T from the second torque target value T without going through the processing in the response delay processing unit S502 in FIG. 5 with respect to the example, and assumes an electric vehicle control method that executes other processes in the same manner. m2 * from the second torque target value T m2 ^ is different in that it is determined, and it is assumed that other processes are executed in the same manner.

[0075] (Example) FIG. 7 is a timing chart showing the first control results of the example and the comparative example. In FIG. 7 and FIG. 8 described later, the control results of the example and the comparative example are shown by a solid line and a broken line, respectively. In particular, as the first control result, a first control scene is assumed in which the first torque target value T is input at a predetermined timing (time t2) from a state where the field current i is not flowing. f flows and the first torque target value T m1 * is input.

[0076] As shown in the figure, from the situation where the first torque target value T is not input and the field current i is 0 (time t0 to t2), when a positive first torque target value T m1 * is input (time t2), the field current i f and the q-axis current i m1 * change from 0 to positive values. At this time, due to the difference in the response characteristics of the stator current and the field current i f described above, the field current i q changes with a delay with respect to the q-axis current i f is different from the q-axis current i f changes with a delay with respect to the q-axis current i q .

[0077] Here, the actual torque response of the drive motor 101 is the field current i f and the q-axis current i qis determined based on the product (see Equation (25)). Also, since the inductance of the field winding is larger than the inductance of the stator winding, the response speed of the field current i f is slower than the response speed of the q-axis current i q . Therefore, in the first control scenario, the field current i q lags behind the q-axis current i f , and the actual torque response determined by these products will show a deviation of a certain amount or more from the desired torque response (the reference torque response).

[0078] And in the control of the comparative example, without considering the difference in the response characteristics of the stator current and the field current i f , the second torque estimated value T m2 * is determined from the second torque target value T m2 ^ , and the disturbance torque estimated value T d_e is calculated. Therefore, at least in a certain period after the time t2 when the first torque target value T m1 * is input, the accuracy of the disturbance torque estimated value T d_e decreases (the disturbance is misestimated). As a result, the correction for the first torque target value T d_e using this disturbance torque estimated value T m1 * is not appropriately performed, and unintended vibrations of the vehicle longitudinal acceleration occur between time t2 and time t3. m2 * On the other hand, in the embodiment, based on the control logic described in FIGS. 5 and 6 etc., the disturbance torque estimated value T

[0079] is calculated taking into account the difference in the response characteristics of the stator current and the field current i f . Therefore, even in the first control scenario, the estimation accuracy of the disturbance is ensured and the correction for the first torque target value T d_e is appropriately performed. As a result, the unintended vibrations of the vehicle longitudinal acceleration between time t2 and time t3 are suppressed, and a suitable acceleration feeling of the electric vehicle 10 can be realized. m1 * ​

[0080] Figure 8 is a timing chart showing the second control results for the examples and comparative examples. In particular, the second control result is the field current i f With the current flowing, the first torque target value T is set at a predetermined timing (time t2). m1 * Let's consider a second control scene where the input is received.

[0081] In this case, the positive first torque target value T is set between times t0 and t2. m1 * The input is, and the field current i f This is maintained at a predetermined value. From this state, at time t2, the first estimated torque T m1 ^ As it decreases, the field current i decreases accordingly. f and q-axis current i q It also decreases. At this time, similar to the first control scene described above, the field current i f is the q-axis current i q It changes with a delay.

[0082] Here, in the second control scene, the field current i f and q-axis current i q A decrease in is anticipated. Therefore, the characteristics of the actual torque response determined based on the product of these factors take on a relatively large value and have a faster response speed than the q-axis current i. q The response characteristics of the field current i become dominant. f and q-axis current i q Disturbance torque estimate T without considering the difference in response characteristics d_e Even when controlling the comparative example that calculates this, no substantial error occurs, and the accuracy of disturbance estimation is maintained.

[0083] On the other hand, field current i f and q-axis current i q Considering the differences in response characteristics, the disturbance torque estimate T d_e In the control of the embodiment that calculates this, the accuracy of disturbance estimation in the second control scene is inevitably ensured.

[0084] Therefore, in the control of this embodiment, the accuracy of disturbance estimation can be ensured in both the first and second control scenes, where the characteristics of the actual torque response differ. In other words, in the control of the electric vehicle 10 equipped with a wound-field synchronous motor, the accuracy of disturbance estimation can be ensured regardless of the motor operating point, and a suitable acceleration feeling can be achieved.

[0085] [Effects and Effects] The configurations of each embodiment described above and their effects will be summarized below.

[0086] In the above embodiment (first or second embodiment), an electric vehicle control method is provided for controlling an electric vehicle 10 equipped with a drive motor 101 composed of an induction motor as a driving source.

[0087] This electric vehicle control method uses vehicle information (vehicle speed V and accelerator opening APO) to determine a first torque target value T. m1 * The first torque target value calculation step (S115) calculates the disturbance torque T acting on the drive motor 101. d The disturbance torque estimation process (S401) estimates the first torque target value T m1 * and disturbance torque T d From the second torque target value T m2 * The second torque target value calculation step (S116) calculates the second torque target value T m2 * Based on the field current command value i f * This includes a field current command value calculation step (117) for calculating the value of the field current command.

[0088] Then, in the disturbance torque estimation process, the field current command value i f * The field current response as the response to the real field current (field current estimate i) f ^ or field current i f The measured value is defined, and the second torque target value T is determined based on the field current response.m2 * The corrected value is the second corrected torque target value (response delay processing value T). m2_pr * (S502) calculates the response delay processing value T m2_pr * Based on the disturbance torque T d Perform the calculation (S504).

[0089] This results in stator current (d-axis current i d or q-axis current i q ) and field current i f Considering the difference in response characteristics between them, the disturbance torque T d The estimated disturbance torque T d A second torque target value T defines the power that should ultimately be supplied to the drive motor 101. m2 * Therefore, the operating point (in particular, the field current i) in a motor control system assuming a wound-field motor can be determined. f Even if variations in torque response (nonlinear torque response delay) occur depending on the control state, the disturbance torque T remains constant regardless of the operating point. d This ensures the accuracy of the estimation.

[0090] Furthermore, in the disturbance torque estimation step of the first embodiment described above, a speed parameter (mechanical angular velocity ω) proportional to the vehicle speed V of the electric vehicle 10 is used. rm Based on this, the first torque target value T m1 * The first estimated torque value T corresponds to the actual torque response. m1 ^ The second torque target value T is calculated (S501). m2 * The input is field current i f Applying a response model (S602) that reflects the response characteristics, the real field current response (field current estimate i) is obtained. f ^ ) is calculated. Then, the estimated field current i f ^ Based on this, the response delay processing value T m2_pr * The second torque estimate T corresponds to this. m2 ^The first estimated torque T is calculated (S603). m1 ^ and the second torque estimate T m2 ^ From the disturbance torque T d Perform the calculation (S504).

[0091] This enables a more specific control logic for estimating disturbances. In particular, the first torque estimate T m1 ^ This is the first torque target value T m1 * It is determined as the actual torque response according to the (basic torque target value based on required driving force). Furthermore, the second torque estimate value T is also determined. m2 ^ The second torque target value T m2 * (The torque target value obtained by adding the effect of disturbances to the basic torque target value) is corrected to the value (response delay processing value T) taking into account the differences in response characteristics described above. m2_pr ) can be determined from this. Therefore, the first torque estimate T m1 ^ The second torque estimate T has less phase shift compared to this. m2 ^ This allows us to define the parameters and further improve the accuracy of disturbance estimation.

[0092] In particular, in the disturbance torque estimation step of the first embodiment, the response model (S602) is defined as the stator current command value (dq axis current command value i d * ,i q * Stator current response models (S6021, S6022) that simulate the response of the actual stator current to ) and field current command value i f * A field current response model S6023 is set, which simulates the response of the actual field current to the given value T. Then, the second torque target value T is set. m2 * Based on the previous value, the dq axis current command value i d * ,i q * and field current command value i f *The dq axis current command value i is calculated (S601). d * ,i q * By applying the stator current response model, the estimated stator current (dq-axis current estimate i) is obtained as the actual stator current response. d ^ ,i q ^ ) is calculated (S6021, S6022), and the field current command value i f * By applying the field current response model, the estimated field current i as the actual field current response is obtained. f ^ The calculation is performed (S6023). Furthermore, the estimated value of the dq axis current i d ^ ,i q ^ and estimated field current i f ^ From the second torque estimate T m2 ^ Perform the calculation (S603).

[0093] This results in the dq axis current i d ,i q and field current i f The second torque estimate T is derived from individual parameters that suggest the response characteristics of each. m2 ^ Therefore, the stator current and field current i can be calculated. f While more favorably reflecting the differences in response characteristics, the disturbance torque T d A calculation logic capable of defining this is realized.

[0094] In particular, the stator current response models (S6021, S6022) and the field current response model S6023 each have different time constants (τ d or τ q and τ f It is composed of a low-pass filter with ).

[0095] This results in the dq axis current i d ,i q and field current i fAfter reflecting the differences in the respective response characteristics, a simpler arithmetic logic for calculating the second torque estimation value T m2 ^ can be realized.

[0096] Also, in the disturbance torque estimation step of the second embodiment, based on a speed parameter (mechanical angular velocity ω rm ) proportional to the vehicle speed V of the electric vehicle 10, a first torque estimation value T m1 * corresponding to the actual torque response corresponding to the first torque target value T m1 ^ is calculated (S501). Further, the measured values of the stator current (dq-axis currents i d , i q ) are defined as the actual stator current response, and the measured value of the field current i f is defined as the actual field current response. Then, based on the measured values of the dq-axis currents i d , i q and the measured value of the field current i f , a second torque estimation value T m2_pr * corresponding to the actual torque response corresponding to the response delay processing value T m2 ^ is calculated (S603), and the disturbance torque T m1 ^ is calculated from the first torque estimation value T m2 ^ and the second torque estimation value T d (S504).

[0097] As a result, while obtaining parameters suggesting the actual stator current response and the actual field current response with simpler processing, an arithmetic logic for determining a second torque estimation value T m1 ^ with less phase shift with respect to the first torque estimation value T m2 ^ can be realized.

[0098] Also, in the above embodiment (the first or second embodiment), a motor controller 107 that functions as an electric vehicle control device suitable for executing the above electric vehicle control method is provided.

[0099] In particular, this motor controller 107 sets a first torque target value T based on vehicle information (vehicle speed V and accelerator opening APO). m1 * The first torque target value calculation unit S115 calculates the external torque T acting on the drive motor 101. d The disturbance torque estimation unit S401 estimates the first torque target value T m1 * and disturbance torque T d From the second torque target value T m2 * The second torque target value calculation unit S116 calculates the second torque target value T m2 * Based on the field current command value i f * It has a field current command value calculation unit (current command value calculation unit 117) that calculates the value.

[0100] Then, the disturbance torque estimation unit S401 calculates the field current command value i f * The field current response as the response to the real field current (field current estimate i) f ^ or field current i f The detected value is defined, and the second torque target value T is determined based on the field current response. m2 * The corrected value is the second corrected torque target value (response delay processing value T). m2_pr * (S502) calculates the response delay processing value T m2_pr * Based on the disturbance torque T d Perform the calculation (S504).

[0101] Although embodiments of the present invention have been described above, these embodiments only represent a part of the application examples of the present invention, and are not intended to limit the technical scope of the present invention to the specific configurations of the above embodiments. [Explanation of Symbols]

[0102] 10 Electric vehicle, 101 Drive motor, 107 Motor controller, S115 First torque target value calculation unit, S116 Second torque target value calculation unit, S401 Disturbance torque estimation unit, S502 Response delay processing unit, S601 Current command value calculation unit, S602 Current response calculation unit, S603 Processing value calculation unit

Claims

1. A method for controlling an electric vehicle equipped with a drive motor consisting of a wound-field motor as a driving source, A first torque target value calculation step that calculates a first torque target value based on vehicle information, A disturbance torque estimation step for estimating the disturbance torque acting on the drive motor, A second torque target value calculation step, which calculates a second torque target value from the first torque target value and the disturbance torque, A field current command value calculation step for calculating a field current command value based on the second torque target value, Includes, In the aforementioned disturbance torque estimation process, The real field current response is defined as the response of the real field current to the aforementioned field current command value, Based on the field current response, a corrected second torque target value, which is a corrected value of the second torque target value, is determined. The disturbance torque is calculated based on the corrected second torque target value. Based on a speed parameter proportional to the vehicle speed of the electric vehicle, a first torque estimate value corresponding to the actual torque response according to the first torque target value is calculated. Using the calculated second torque target value as input, the actual field current response is obtained by applying a response model that reflects the response characteristics of the field current. Based on the actual field current response, a second torque estimate value corresponding to the actual torque response according to the corrected second torque target value is calculated. The disturbance torque is calculated from the first torque estimate and the second torque estimate. Electric vehicle control method.

2. The electric vehicle control method according to claim 1, As the response models, a stator current response model that simulates the response of the actual stator current to the stator current command value and a field current response model that simulates the actual field current response are set. Based on the previous value of the second torque target value, the stator current command value and the field current command value are calculated. By applying the stator current response model to the stator current command value, an estimated stator current value is calculated as the actual stator current response to the stator current command value. By applying the field current response model to the field current command value, the estimated field current value as the actual field current response is calculated. The second torque estimate is calculated from the stator current estimate and the field current estimate. Electric vehicle control method.

3. The electric vehicle control method according to claim 2, The stator current response model and the field current response model are each composed of low-pass filters with different time constants. Electric vehicle control method.

4. The electric vehicle control method according to claim 1, In the aforementioned disturbance torque estimation process, Based on a speed parameter proportional to the vehicle speed of the electric vehicle, a first torque estimate value corresponding to the actual torque response according to the first torque target value is calculated. The measured value of the stator current is defined as the actual stator current response. The measured value of the field current is defined as the actual field current response. Using the measured values ​​of the stator current and the field current as inputs, a second torque estimate value corresponding to the actual torque response according to the corrected second torque target value is calculated. The disturbance torque is calculated from the first torque estimate and the second torque estimate. Electric vehicle control method.

5. An electric vehicle control device for controlling an electric vehicle equipped with a drive motor consisting of a wound-field motor as a driving source, A first torque target value calculation unit calculates a first torque target value based on vehicle information, A disturbance torque estimation unit that estimates the disturbance torque acting on the drive motor, A second torque target value calculation unit calculates a second torque target value from the first torque target value and the disturbance torque, A field current command value calculation unit calculates a field current command value based on the second torque target value, It has, The disturbance torque estimation unit, The real field current response is defined as the response of the real field current to the aforementioned field current command value, Based on the field current response, a corrected second torque target value, which is a corrected value of the second torque target value, is determined. The disturbance torque is calculated based on the corrected second torque target value. Based on a speed parameter proportional to the vehicle speed of the electric vehicle, a first torque estimate value corresponding to the actual torque response according to the first torque target value is calculated. Using the calculated second torque target value as input, the actual field current response is obtained by applying a response model that reflects the response characteristics of the field current. Based on the actual field current response, a second torque estimate value corresponding to the actual torque response according to the corrected second torque target value is calculated. The disturbance torque is calculated from the first torque estimate and the second torque estimate. Electric vehicle control device.

Images