Motor control device

Abstract

A motor control device includes a motor that generates a torque corresponding to a current supplied to coils of a plurality of phases, a current sensor that detects a current value of the current supplied to the coils of the plurality of phases, and a controller that calculates the current value flowing in a predetermined coil by adding a zero point learning value to a signal input from the current sensor, and controls the current supplied to the predetermined coil based on the current value, wherein an amplitude of a predetermined order decided in advance is calculated based on the zero point learning value after a change of a predetermined value decided in advance and the signal input from the current sensor, and the zero point learning value at a point in time when a tendency to decrease is switched to a tendency to increase is corrected based on the amplitude.

Description

TECHNICAL FIELD

[0001] The present application relates to a control device of a motor provided as a drive power source of a vehicle. BACKGROUND

[0002] A motor provided as a drive power source in a conventional electric vehicle is driven by applying an alternating voltage from a power source. The current value of each phase in the motor is used as a feedback value to control the alternating voltage. On the other hand, a current sensor outputs a voltage or the like signal corresponding to the detected value to a controller. Therefore, for example, the voltage value input to the controller in the case where the actually flowing current value is 0 A is taken as a reference to judge the current value flowing in the current sensor. That is, the signal (for example, voltage value) output from the current sensor in the case where the current flowing in the current sensor is 0 A is taken as an offset value and is stored in advance to the controller. Therefore, the controller controls the voltage applied to the motor by adding the offset value to the feedback value from the current sensor. Therefore, when the offset value of the sensor that detects the current value of a predetermined phase is changed for some reason, the current value judged by the controller is different from the actual current value. In such a case, the voltage applied to the predetermined phase is changed, so there is a possibility that the output torque of the motor is pulsated to generate vibration called motor noise.

[0003] Therefore, the control device described in Japanese Patent Application Publication No. 2016-119817 is configured to correct the offset value as described above. With regard to the motor described in Japanese Patent Application Publication No. 2016-119817, the torque is controlled by controlling an inverter constituted by a switching element and a diode connected in anti-parallel between a direct current power source. Therefore, at the time point when the switching element is stopped, no current flows in the phase connected to the direct current power source via the switching element. That is, the theoretical current value of the phase connected to the direct current power source via the stopped switching element is 0 A. Therefore, the control device described in Japanese Patent Application Publication No. 2016-119817 is configured to detect the current value of the phase connected to the direct current power source via the stopped switching element using the current sensor at the time of parking or at the timing when the switching element is temporarily stopped during running, and to store the detected current value as an offset value. SUMMARY

[0004] According to the control device described in Japanese Patent Application Publication No. 2016-119817, the detected value of the current sensor at the time point when the current value becomes 0 A in theory is stored as an offset value. However, in the signal input from the current sensor to the controller, noise from the current sensor or the surrounding electrical equipment such as a circuit connecting the current sensor and the controller is included. In addition, an error is generated in the detected value due to the temperature or the like of the current sensor. Therefore, it is difficult to make the offset value a correct value, and there is a possibility that the output torque of the motor is pulsated to generate motor noise.

[0005] The present application has been achieved in view of the above-described technical problems, and has an object to provide a motor control device capable of correcting an origin learning value for suppressing motor noise.

[0006] To achieve the above-described object, the present application is a motor control device characterized by comprising: a motor having coils of a plurality of phases that generate a torque corresponding to a current supplied to the coils of the plurality of phases; a current sensor that detects a current value of a current supplied to a predetermined coil among the coils of the plurality of phases, and outputs a signal corresponding to the detected current value; and a controller that calculates a current value flowing in the predetermined coil by adding an origin learning value to the signal input from the current sensor, and controls the current supplied to the predetermined coil based on the calculated current value, the controller being configured to calculate an amplitude of a predetermined order decided in advance under a q-axis current of the motor based on the origin learning value after a change and the signal input from the current sensor each time the origin learning value is changed by a predetermined value decided in advance, and correct the origin learning value based on the origin learning value at a point of time when a tendency to decrease is switched to a tendency to increase from the amplitude.

[0007] Further, in the present application, the predetermined order can include an electrical first order.

[0008] Further, in the present application, the motor can be provided in a tire wheel.

[0009] Further, the amplitude of the predetermined order can be calculated by performing a fast Fourier transform on the q-axis current.

[0010] Further, in the present application, the motor can have a U-phase coil, a V-phase coil, and a W-phase coil, and a neutral point to which terminals of one of the U-phase coil, the V-phase coil, and the W-phase coil are connected, and the current sensor can include a first current sensor that detects a current flowing in any one of the U-phase coil, the V-phase coil, and the W-phase coil, and a second current sensor that detects a current flowing in the other of the U-phase coil, the V-phase coil, and the W-phase coil.

[0011] According to the present application, the actual current value flowing in the predetermined coil is calculated by adding an origin learning value to the current value detected by the current sensor. The origin learning value is changed by a predetermined value amount every time the origin learning value is changed, and the amplitude of the predetermined order in the q-axis current is calculated from the changed origin learning value and the signal input from the current sensor. Further, the origin learning value is corrected from the origin learning value at the point in time when the amplitude changes from the decreasing tendency to the increasing tendency. That is, the origin learning value is calculated so that the pulsation of the q-axis current is reduced, and the origin learning value is corrected. By thus correcting the origin learning value, the origin learning value can be corrected with high precision without being affected by noise of the surrounding equipment, temperature of the current sensor, and the like. As a result, the motor noise can be reduced, and further, the vibration of the vehicle and the in-vehicle sound can be reduced. BRIEF DESCRIPTION OF DRAWINGS

[0012] Features, advantages, and technical and industrial significance of exemplary embodiments of the present application will be described below with reference to the accompanying drawings, in which like numerals denote like elements, and wherein:

[0013] Figure 1 is a schematic diagram for explaining one example of a vehicle provided with a motor in an embodiment of the present application.

[0014] Figure 2 is a circuit diagram for explaining one example of the structure of a motor and an inverter.

[0015] Figure 3 is a flowchart for explaining a control example executed by the motor control device of the present application.

[0016] Figure 4 is a graph showing the change of the origin learning value and the FFT spectrum in the case where the control example shown in Figure 3 is executed.

[0017] Figure 5 is a graph showing the effect of reducing the pulsation of the q-axis current by correcting the origin learning value.

[0018] Figure 6 is a graph showing the effect of reducing the in-vehicle sound by correcting the origin learning value. DETAILED DESCRIPTION

[0019] Figure 1 is a schematic diagram for explaining one example of a vehicle provided with a motor in an embodiment of the present application. Figure 1The illustrated vehicle 1 is a so-called in-wheel motor vehicle in which a motor (M) 2 is provided in each tire wheel, and a housing that houses the motor 2 is held to a vehicle body via a knuckle and a suspension, which are not illustrated. Further, the motor in the embodiment of the present application can also be a motor that is provided as a driving power source of the vehicle and is supported to the vehicle body via a bracket.

[0020] The above-described motor 2, like a motor provided as a driving power source of a conventional electric vehicle or hybrid vehicle, is configured to have a plurality of phase coils, such as a three-phase alternating-current synchronous motor, an induction motor, or the like, and generate a torque corresponding to a current supplied to the coils. In the following description, a synchronous motor in which permanent magnets are attached to a rotor will be described as an example.

[0021] The above-described motor 2 is configured to output a driving torque by being supplied with electric power from a direct-current power source (BAT) 3 provided in the vehicle 1. In addition, the motor 2 is configured to generate an induced voltage in the motor 2 in a case where a regenerative torque is output from the motor 2 in such a manner that the rotational speed of the motor 2 is reduced, that is, in a case where the motor 2 is rotated, as a result of which the direct-current power source 3 can be charged. That is, the motor 2 is configured to function as a motor generator. Further, in the following description, the direct-current power source 3 will be simply referred to as the power source 3.

[0022] Figure 1 The illustrated vehicle 1 has an inverter (INV) 4 between each motor 2 and the power source 3, which inverts a direct-current voltage output from the power source 3 to an alternating-current voltage and outputs the same to each motor 2, and inverts an alternating-current induced voltage generated by each motor 2 to a direct-current voltage and outputs the same to the power source 3.

[0023] In Figure 2 The motor 2 is a three-phase alternating-current motor as described above, and thus has a U-phase coil 2u, a V-phase coil 2v, and a W-phase coil 2w. One-side terminals of these coils 2u, 2v, 2w are connected to a neutral point 5, and the other-side terminals are connected to the inverter 4.

[0024] Figure 2 The illustrated inverter 4 is configured by an upper branch switch (element) 8 and a lower branch switch (element) 9 provided between a positive bus 6 connected to a positive terminal of the power source 3 and a negative bus 7 connected to a negative terminal of the power source 3. The collector of the upper branch switch 8, which is a high-potential side terminal, is connected to the positive bus 6, the emitter, which is a low-potential side terminal, is connected to the collector of the lower branch switch 9, which is a high-potential side terminal, and the emitter of the lower branch switch 9, which is a low-potential side terminal, is connected to the negative bus 7. That is, the upper branch switch 8 and the lower branch switch 9 are connected in series.

[0025] As described above, the motor 2 is a three-phase alternating-current motor, and therefore the upper branch switch 8 and the lower branch switch 9 are each composed of three switches. Specifically, the upper branch switch 8 is composed of a first switch Ql connected to the U-phase coil 2u, a third switch Q3 connected to the V-phase coil 2v, and a fifth switch Q5 connected to the W-phase coil 2w. In addition, the lower branch switch 9 is composed of a second switch Q2 connected to the U-phase coil 2u, a fourth switch Q4 connected to the V-phase coil 2v, and a sixth switch Q6 connected to the W-phase coil 2w.

[0026] These switches Ql to Q6 are each composed of an insulated gate bipolar transistor (IGBT) known in the art, to which a freewheeling diode Dl to D6 is connected in anti-parallel. In addition, the switches Ql to Q6 are not limited to IGBTs, but can be composed of other switching elements such as a metal oxide semiconductor field effect transistor (MOSFET), for example.

[0027] One end of the U-phase coil 2u is connected to the connection point of the first switch Ql and the second switch Q2, one end of the V-phase coil 2v is connected to the connection point of the third switch Q3 and the fourth switch Q4, and one end of the W-phase coil 2w is connected to the connection point of the fifth switch Q5 and the sixth switch Q6.

[0028] In the above-described circuit, current sensors 10v, 10w are provided for detecting the currents flowing in the V-phase coil 2v and the W-phase coil 2w. In addition, as described above, the terminal of one of the phase coils 2u, 2v, 2w is connected at the neutral point 5, and therefore by detecting the currents flowing in the V-phase coil 2v and the W-phase coil 2w, the current flowing in the U-phase coil 2u can be calculated and obtained. The current sensor 10v that detects the current flowing in the V-phase coil 2v corresponds to the "first current sensor" in the embodiment of the present application, and the current sensor 10w that detects the current flowing in the W-phase coil 2w corresponds to the "second current sensor" in the embodiment of the present application.

[0029] An electronic control device (hereinafter referred to as ECU) 11 is provided, which is inputted with signals from the above-described current sensors 10v, 10w, a resolver or the like rotational speed sensor that detects the rotational speed of the motor 2, a throttle pedal opening sensor that detects the operation amount of a not-shown throttle pedal, a brake sensor that detects the operation amount of a brake pedal, and the like various sensors, and outputs a command signal to the motor 2 in accordance with these signals and a map or the like stored in advance. The ECU 11 can be configured in the same manner as the electronic control device known in the art, with a microcomputer as the main body.

[0030] The ECU 11 calculates the required driving force from the accelerator pedal opening degree detected by the accelerator pedal opening degree sensor, or the required braking force from the brake operation amount detected by the brake sensor. Also, to output a torque corresponding to these required driving force and required braking force from the motor 2, the d-axis current and the q-axis current are calculated in the same manner as the torque control of the motor known in the art, and the voltage applied to the U-phase coil 2u, the V-phase coil 2v, and the W-phase coil 2w is controlled in accordance with the rotational angle of the motor 2 and the q-axis current. Specifically, the on-off duty ratio of each switch Ql to Q6 is controlled.

[0031] By controlling each switch Ql to Q6 as described above, the current values flowing in each phase coil 2u, 2v, 2w are detected by the current sensors 10v, 10w, and the difference between the detected value detected by the current sensors 10v, 10w and the target value is fed back.

[0032] The current sensors 10v, 10w are configured to output a signal (for example, a voltage) corresponding to the current value flowing in the current sensors 10v, 10w to the ECU 11. In a specific example, in the case where the current value flowing in the current sensor 10v is 0 A, a predetermined voltage is output from the current sensor 10v to the ECU 11. Also, it is configured that as the current value flowing in the current sensor 10v increases, a voltage obtained by adding a predetermined multiple of the amount of increase to the above-described predetermined voltage is input from the current sensor 10v to the ECU 11. That is, the ECU 11 judges the current value flowing in the current sensor 10v using the voltage value input from the current sensor 10v in the case where the current value flowing in the current sensor 10v is 0 A as a reference value. Also, the current sensor 10w is configured in the same manner. In the following description, the above-described reference value is described as an origin learning value.

[0033] Therefore, in the case where the origin learning value of the current sensor 10v, 10w changes due to some reason, the current value judged by the ECU 11 is different from the current value actually flowing in the current sensor 10v, 10w. In such a case, in order to make the current value energizing the coil 2v (2w) corresponding to the current sensor 10v (10w) whose origin learning value changes follow the target value, the voltage value applied to the coil 2v (2w) is changed. Therefore, the current value actually flowing in the coil corresponding to the sensor 10v (10w) whose origin learning value changes is different from the target value, resulting in that the current values (peak values) energizing the three coils 2u, 2v, 2w respectively are different. As a result, the q-axis current pulsates, so there is a possibility that vibration is transmitted to the vehicle 1 due to the pulsation of the output torque of the motor 2.

[0034] Accordingly, the motor control device in the embodiment of the present application is configured to correct the origin learning value of the current sensor 10v, 10w in accordance with the vibration of the q-axis current. In Figure 3 In the following, a flowchart for explaining the control example is shown. Figure 3 The control example shown is executed by the ECU 11, and first, the origin learning value is acquired (step S1). In this step S1, the origin learning value stored in the ECU 11 is read in.

[0035] Next, the motor speed and the q-axis current are acquired (step S2). The motor speed can be found from the signal detected by the motor speed sensor. Also, regarding the q-axis current, first, the detection values of the sensors 10v, 10w of each phase are stored to the ECU 11 for a predetermined period decided in advance, and the origin learning value acquired in step S1 is added to the above detection values to find the current value for the predetermined period. Next, the progression of the q-axis current for the predetermined period is found from the current value of each phase and the motor speed, as known in the art.

[0036] Next, the q-axis current found in step S2 is subjected to fast Fourier transform (FFT), and the amplitude a of the electric first order is calculated (step S3). Thereafter, only the origin learning value with respect to the current sensor 10v is increased by a predetermined value α decided in advance, and the same processing as in steps S2 and S3 is performed again to calculate the amplitude b of the electric first order (step S4). Also in step S4, the detection values of the motor 2, the current sensors 10v, 10w can be acquired again after the origin learning value is increased to find the q-axis current, or the detection values of the motor speed, the current sensors 10v, 10w acquired in step S2 can be used to find the q-axis current.

[0037] Then, it is judged whether the amplitude b found in step S4 is greater than the amplitude a found in step S3 (step S5). That is, whether the amplitude of the electric first order of the q-axis current is increased after the origin learning value is changed is judged, as compared to before the origin learning value is changed. In other words, in order to reduce the offset of the target value of the origin learning value which should judge the actual current value from the origin learning value set at the current time point, it is judged whether the origin learning value is changed.

[0038] In the case where it is judged negative in step S5 because the amplitude b found in step S4 is smaller than the amplitude a found in step S3, it is considered that the origin learning value is changed toward the target value of the origin learning value. Therefore, only the origin learning value with respect to the current sensor 10v is further increased by the predetermined value α, and the same processing as in steps S2 and S3 is performed again to calculate the amplitude c of the electric first order (step S6). Next, it is judged whether the amplitude c found in step S6 is greater than the amplitude b found in step S4 (step S7).

[0039] In the case where the determination in step S7 is negative because the amplitude c found in step S6 is smaller than the amplitude b found in step S4, the process returns to step S6. That is, until the amplitude of the electrical one order at the origin learning value set immediately before increases (i.e., until the determination in step S7 is affirmative), the following steps are repeated: the origin learning value is increased by a predetermined value a, the amplitude of the electrical one order at the origin learning value after the increase is found, and the found amplitude of the electrical one order is compared with the amplitude of the electrical one order at the origin learning value set immediately before. That is, until the amplitude of the q-axis current switches from a decreasing tendency to an increasing tendency, the origin learning value is increased.

[0040] In the case where the determination in step S7 is affirmative because the amplitude of the electrical one order increases, the origin learning value set immediately before the amplitude of the electrical one order is about to increase is stored to the ECU 11 (step S8).

[0041] On the other hand, in the case where the determination in step S5 is affirmative because the amplitude b found in step S4 is larger than the amplitude a found in step S3, the amplitude a found in step S3 is decreased by a predetermined value a, and the amplitude d of the electrical one order is calculated again by the same process as in step S3 (step S9). Next, it is determined whether the amplitude d found in step S9 is larger than the amplitude a found in step S3 (step S10).

[0042] In the case where the determination in step S10 is negative because the amplitude d found in step S9 is smaller than the amplitude a found in step S3, the process returns to step S9. That is, until the amplitude of the electrical one order at the origin learning value set immediately before increases (i.e., until the determination in step S10 is affirmative), the following steps are repeated: the origin learning value is decreased by a predetermined value a, the amplitude of the electrical one order at the origin learning value after the decrease is found, and the found amplitude of the electrical one order is compared with the amplitude of the electrical one order at the origin learning value set immediately before.

[0043] In the case where the determination in step S10 is affirmative because the amplitude of the electrical one order increases, the origin learning value set immediately before the amplitude of the electrical one order is about to increase is stored (learned) to the ECU 11 (step S8).

[0044] Then, after the origin learning value with respect to the current sensor 10v is stored to the ECU 11 by step S8, the origin learning value with respect to the current sensor 10w is learned and stored to the ECU 11 (step Sll). In this step Sll, the same steps as in the above steps S4 to S10 are performed by rewriting the origin learning value to which the predetermined value a is added or subtracted to the W-phase.

[0045] Figure 4is a graph for illustrating the relationship of the FFT spectrum of the electric 1st order at the q-axis current and the origin learning value in the case where the above-described steps S1 to S8 are executed, the horizontal axis indicates the origin learning value, and the vertical axis indicates the FFT spectrum. In Figure 4 In the example shown, in the case where the motor 2 is driven using the origin learning value (initial value) acquired in step S1, the FFT spectrum calculated in step S2 becomes a. By increasing the origin learning value by a predetermined value a each time from this state, the FFT spectrum gradually decreases. And, as a result of increasing the initial origin learning value by the predetermined value a each time, the FFT spectrum increases at a predetermined origin learning value.

[0046] Therefore, at the point in time when the predetermined origin learning value is increased, the affirmative is judged in step S7. As a result, by step S8, the origin learning value before the FFT spectrum is about to increase is stored to the ECU 11.

[0047] Figure 5 And Figure 6 is a graph for illustrating the effect of the case where the above-described control example is executed to correct the origin learning value, Figure 5 is a graph in which the horizontal axis indicates the vibration frequency and the vertical axis indicates the FFT spectrum, Figure 6 is a graph in which the horizontal axis indicates the vehicle speed and the vertical axis indicates the in-vehicle sound of the electric 1st order.

[0048] In the case where the vibration of the electric 1st order is significantly larger than the vibration of other orders as shown in Figure 5 and the origin learning value is corrected by executing the above-described control example (solid line), the FFT spectrum decreases compared to before the above-described control example is executed and before the origin learning value is corrected (dotted line). That is, the pulsation (amplitude) of the q-axis current decreases, and as a result, it is known that the motor noise decreases.

[0049] It is known that by decreasing the motor noise as described above, in the case where the origin learning value is corrected by executing the above-described control example as shown in Figure 6 (solid line), the in-vehicle sound decreases from the low vehicle speed range to the high vehicle speed range compared to before the above-described control example is executed and before the origin learning value is corrected (dotted line).

[0050] By calculating the origin learning value at which the pulsation of the q-axis current decreases as described above and correcting it, the origin learning value can be corrected with high accuracy without being affected by the noise of the surrounding equipment, the temperature of the current sensor, and the like. As a result, it is possible to decrease the motor noise, and further, it is possible to decrease the vibration of the vehicle and the in-vehicle sound.

[0051] Further, in the embodiment of the present application, the origin learning value is corrected based on the change in the amplitude of the electric first order under the q-axis current caused by changing the origin learning value, but the origin learning value can be corrected based on the amplitude of the electric second order or other orders. In addition, the motor is not limited to one having three phases of coils, but can be a motor having four or more phases of coils. That is, the motor control device in the embodiment of the present application can be configured as follows: whenever the origin learning value is changed, the amplitude of a predetermined order under the q-axis current is calculated, and the origin learning value is corrected from the origin learning value at the point in time when the tendency to decrease is switched to the tendency to increase based on the amplitude.

Claims

1. A motor control device characterized by comprising: Possessing: a motor having a plurality of phase coils that generate a torque corresponding to a current applied to the plurality of phase coils; a current sensor that detects a current value applied to at least any one predetermined coil of the plurality of phase coils and outputs a signal corresponding to the detected current value; and a controller that calculates a current value flowing in the predetermined coil by adding an origin learning value to the signal input from the current sensor, the origin learning value being a voltage value input from the current sensor when the current value flowing in the current sensor is 0 A, and controls the current applied to the predetermined coil based on the calculated current value, the controller is configured to: calculate an amplitude of a predetermined order of a q-axis current of the motor based on the origin learning value after the origin learning value is changed by a predetermined value amount and the signal input from the current sensor each time the origin learning value is changed by the predetermined value amount, the amplitude of the predetermined order being calculated by performing a fast Fourier transform on the q-axis current, correct the origin learning value based on the origin learning value at a point in time when the amplitude switches from a decreasing tendency to an increasing tendency.

2. The motor control device according to claim 1, wherein the predetermined order includes an electrical first order.

3. The motor control device according to claim 1 or 2, wherein the motor is provided in a tire wheel.

4. The motor control device according to claim 1 or 2, wherein the motor includes a U-phase coil, a V-phase coil, and a W-phase coil, and a neutral point to which a terminal of one of the U-phase coil, the V-phase coil, and the W-phase coil is connected, the current sensor includes: a first current sensor that detects a current flowing in any one of the U-phase coil, the V-phase coil, and the W-phase coil; and a second current sensor that detects a current flowing in the other of the U-phase coil, the V-phase coil, and the W-phase coil. ​

Images