JP2005199780A - Control device of electric power steering device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve a problem in a motor control wherein the motor control can not be rightly executed when neutral point potential is unstable, and torque ripple and noise of the motor are large and loud and are not desirable. <P>SOLUTION: The motor control for stabilizing the neutral point potential is enabled by performing a control to estimate and compensate counter electromotive voltage of each phase of the motor after reducing an average value of each phase voltage command value in the voltage command value of each phase determined from a current command value to the motor and detected current of the motor, to reduce torque ripple and noise of the motor. In this device using the motor, a steering wheel operation reduced in noise and having good feeling is enabled. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a control device for an electric power steering device, and more particularly to a control device for an electric power steering device provided with motor control for stabilizing the neutral point potential of a motor.

  An electric power steering device for energizing an automotive steering device with a rotational force of a motor applies an auxiliary force to a steering shaft or a rack shaft by a transmission mechanism such as a gear or a belt via a reduction gear. It comes to be energized. The configuration of such an electric power steering apparatus will be described with reference to FIG. A shaft 102 of the steering handle 101 is connected to a tie rod 106 of a steering wheel via a reduction gear 103, universal joints 104a and 104b, and a pinion rack mechanism 105. The shaft 102 is provided with a torque sensor 107 that detects the steering torque of the steering handle 101, and a motor 108 that assists the steering force of the steering handle 101 is connected to the shaft 102 via the reduction gear 103. Yes.

In such an electric power steering apparatus, control of the motor 108 is important. For example, the relationship between the voltage and current of the three-phase motor is expressed by equations (1), (2), and (3).
(Equation 1)
Van = Va−Vn = EMFa + La · (dIa / dt) + Ra · Ia
(Equation 2)
Vbn = Vb−Vn = EMFb + Lb · (dIb / dt) + Rb · Ib
(Equation 3)
Vcn = Vc−Vn = EMFc + Lc · (dIc / dt) + Rc · Ic
Here, definitions relating to the motor voltage, current, and inductance value and resistance value of the motor winding will be described with reference to FIG. Van, Vbn, and Vcn indicate voltages between the neutral point of the motor and terminals of each phase (a phase, b phase, and c phase). Va, Vb, Vc, and Vn represent a voltage between the ground and each phase terminal and a voltage between the neutral points, in other words, a potential. Ia, Ib, and Ic indicate currents of respective phases of the motor. La, Lb, Lc and Ra, Rb, Rc indicate the phase inductance value and resistance value of the motor winding.

  Here, when the neutral point potential Vn varies, the phase voltages Van, Vbn, Vcn vary, and if this is not used as an error, torque ripple becomes large and becomes a problem.

To deal with such a problem, for example, Patent Document 1 discloses a control method for stabilizing the neutral point with respect to control of a motor having a power supply at the neutral point.
JP 2002-84758 A

  However, there is no disclosure of a control method that stabilizes the neutral point by the motor control of a method that does not install a power source at the neutral point, and as described above, if the neutral point potential is unstable, the torque ripple of the motor It becomes large and is not preferable. Furthermore, if a special sensor is required for the control for stabilizing the neutral point potential, there is a problem that the cost of the apparatus increases, and if the special control is performed, the processing load of the CPU increases.

  The present invention has been made for the above-mentioned circumstances, and the object of the present invention is to provide a control system that does not require any special sensor, is simple in control processing, and has a low burden on the CPU. Control of an electric power steering system that can handle the steering wheel with a good feeling by stabilizing the neutral point of a multi-phase motor that is not equipped with a power source and enabling control of the motor with less torque ripple and noise To provide an apparatus.

  The present invention relates to a multiphase motor for applying a steering assist force to a steering system of a vehicle, a current command value calculating means for calculating a current command value for each phase to the motor, and a current for each phase of the motor. The present invention relates to a control device for an electric power steering apparatus, comprising: a current detection unit that performs a current control unit that calculates a voltage command value for each phase based on the current command value and the detected current. The above object is provided with a counter electromotive voltage calculation means for estimating a counter electromotive voltage value induced in the motor, subtracts an average value of the voltage command values of the phases from the voltage command values of the phases, and This is achieved by controlling based on a new voltage command value for each phase obtained by adding back electromotive voltage values for each phase. The above-mentioned object of the present invention is achieved more effectively by the current control means being a feedback control means. The above-mentioned object of the present invention is achieved more effectively by the current control means comprising a feedforward control means and a disturbance observer. The above-mentioned object of the present invention can be achieved more effectively by estimating the back electromotive voltage from the terminal voltage and current of the motor. The above-mentioned object of the present invention is achieved more effectively by estimating the back electromotive force from the angular velocity and motor position of the motor. The above-mentioned object of the present invention is more effectively achieved by the motor being a three-phase motor and the connection system being a star connection or a delta connection.

  The neutral point potential of the motor can be expressed as the average value of each phase obtained by subtracting the sum of the back electromotive force of each phase from the sum of the terminal voltage of each phase. Compensating the average value of the command value and further compensating the back electromotive force of each phase makes it possible to stabilize the neutral point potential, and as a result, it has motor control with less torque ripple and noise. There is an effect that it is possible to provide a control device for an electric power steering device capable of a steering operation with a good feeling.

  First, the theory on which the present invention is established will be described, and then a preferred embodiment of the present invention will be described.

By adding the left side and the right side of the equations (1), (2), and (3), which are equations related to the voltage and current of the motor, the equation (4) is derived.
(Equation 4)
Van + Vbn + Vcn
= Va + Vb + Vc-3Vn
= EMFa + La. (DIa / dt) + Ra.Ia + EMFb + Lb. (DIb / dt) + Rb.Ib + EMFc + Lc. (DIc / dt) + Rc.Ic
It becomes.

Here, when La = Lb = Lc = L, Ra = Rb = Rc = R, and Ia + Ib + Ic = 0 are substituted into Equation 4, Equations 5 and 6 are derived.
(Equation 5)
Vn = 1/3 ((Va + Vb + Vc)-(Van + Vbn + Vcn))
(Equation 6)
Vn = 1/3 ((Va + Vb + Vc) − (EMFa + EMFb + EMFc))
It becomes.

Here, paying attention to Equation 6, it is considered that the neutral point potential Vn is stabilized at a constant value, for example, 0. For this purpose, the phase voltages Va, Vb, and Vc are controlled to have a relationship as shown in Equation 7.
(Equation 7)
Va = Varef + EMFa
Vb = Vbref + EMFb
Vc = Vcref + EMFc
Then, when Equation 7 is substituted into Equation 6, Equation 8 is obtained.
(Equation 8)
Vn = 1/3 (Varef + Vbref + Vcref)
Here, in order for the neutral point potential Vn to be 0, (Varef + Vbref + Vcref) = 0 must be satisfied. However, in order for the sum of the voltage command values Varef, Vbref, and Vcref of each phase to be zero, it is assumed that there is no variation due to secular change or temperature change in parameters such as L and R of each phase of the motor. Become. Actually, there is variation, and the current control means of each phase performs control to correct this error, so that the sum of the voltage command values of each phase does not become zero. Therefore, the neutral point potential Vn cannot be stabilized even if the control for maintaining the relationship of Equation 7 is performed.

Therefore, an attempt is made to stabilize the neutral point potential Vn by reducing the average value of the voltage command values of each phase. Specifically, control is performed to maintain the relationship of Equation 9.
(Equation 9)
Va = Varef + EMFa-1 / 3 (Varef + Vbref + Vcref)
Vb = Vbref + EMFb-1 / 3 (Varef + Vbref + Vcref)
Vc = Vcref + EMFc-1 / 3 (Varef + Vbref + Vcref)
Substituting Equation 9 into Equation 6, for example, even if (Varef + Vbref + Vcref) = 0 is not satisfied, Vn = 0, and the neutral point potential Vn can be stabilized.

  FIG. 1 is a control block diagram illustrating the theory of the present invention. The torque value detected by the torque sensor 107 and the vehicle speed value Vs detected by the vehicle speed sensor 120 are input to the torque command value calculation unit 122, and the torque command value Tref is output as a result of the calculation. The torque command value Tref is input, and a current command value calculation unit 204 serving as a current command value calculation unit outputs each phase current command value. For example, in the case of a three-phase motor, current command values Iaref, Ibref, and Icref are output. As an example of the current command calculation unit 204, current command values Iaref, Ibref, and Icref are calculated by performing two-phase / three-phase conversion of current command values Idref and Iqref by vector control using the d-axis and q-axis. There is a value calculation unit or a current command value calculation unit for calculating each phase current command value Iaref, Ibref, Icref by a rectangular wave current.

  Each phase current Ia, Ib, Ic detected by the current detector 205 as current detecting means for detecting each phase current Ia, Ib, Ic of the motor 108 is converted into the phase current command values Ia, Ib, Ic is input to a current control unit A as current control means, and voltage command values Varef, Vbref, and Vcref for each phase are output. Specific contents of the current control unit A will be described in detail in an embodiment described later. The voltage command values Varef, Vbref, and Vcref of each phase are input to the average value calculation unit 10, and the average value Vav = 1/3 (Varef + Vbref + Vcref) is output. The subtraction unit 12 outputs the phase voltage command values Varef, Vbref. , Vcref, respectively, the average value Vav is subtracted.

  On the other hand, the counter electromotive voltage eaf, ebf, ecf of each phase induced by the motor 108 is estimated by the counter electromotive voltage calculation unit 20 as the counter electromotive voltage calculation means, and is an output of the subtraction unit 12 (Varef−Vav). ), (Vbref−Vav), and (Vcref−Vav), the above-described counter electromotive voltages eaf, ebf, and ecf are respectively added by the adding unit 14. Then, the voltage command values (Varef−Vav + eaf), (Vbref−Vav + ebf), and (Vcref−Vav + ecf) of new phases that are output values of the adding unit 14 are input to the PWM control unit 210, and the inverter circuit 211 PWM control is performed based on the PWM control signal output from the controller 210 to supply current to the motor 108.

  As a result of such a control method according to the present invention, it is possible to stabilize the neutral point potential of the motor, and as a result, it is possible to control the motor with less torque ripple. The above explanation is the theoretical support of the present invention. Embodiments of the present invention will be described below with reference to the drawings.

  A preferred embodiment of the present invention will be described with reference to FIGS. 2, 3, 4 and 5. FIG. In the above-described theoretical explanation of the present invention, the neutral point potential Vn of the motor has been described as 0. However, in the motor control used in the electric power steering apparatus, the battery voltage Vbat which is the power source of the motor is centered on 0. Since it is not a power source that outputs positive and negative values but a power source that outputs only one of positive and negative values, the neutral point potential Vn to be stabilized is controlled to be Vbat / 2 instead of 0.

From the relationship of terminal voltage Va = Van + Vn, Vbat / 2 may be added to the formula (9). Therefore, the theoretical formula for controlling the neutral point potential Vn to be maintained at Vbat / 2 instead of 0 is expressed by Equation 10.
(Equation 10)
Va = Varef + EMFa-1 / 3 (Varef + Vbref + Vcref) + Vbat / 2
Vb = Vbref + EMFb-1 / 3 (Varef + Vbref + Vcref) + Vbat / 2
Vc = Vcref + EMFc-1 / 3 (Varef + Vbref + Vcref) + Vbat / 2
The present embodiment is an embodiment in which control is performed so as to satisfy the formula (10) in consideration of the battery power source of the electric power steering apparatus. In this embodiment, a case where feedback control is used as an embodiment of the current control unit A will be described.

  First, the rotation angle θ of the motor 108 is detected by a resolver 201 as an example of a position detector that detects the rotation angle θ of the motor 108 and a rotation angle detector 202 that includes an RDC circuit that processes an output signal of the resolver 201. And the angular velocity ω is detected. The torque command value Tref, the rotation angle θ, and the angular velocity ω are input to the current command value calculation unit 204 serving as a current command value calculation unit, and current command values Iaref, Ibref, and Icref for each phase are output.

  In this embodiment, the current control unit A as current control means has a feedback control configuration. The phase currents Ia, Ib, and Ic of the motor 108 are detected as follows. First, currents Ia and Ic are detected by current detectors 205-1 and 205-2 as current detection means. Next, the current Ib is calculated as current Ib = − (Ia + Ic) from the relationship of (Ia + Ib + Ic = 0) by using the adder 205-3 and the polarity inversion unit 205-4 as inputs of the currents Ia and Ic.

  The detected phase currents Ia, Ib, Ic of the motor 108 are fed back to the subtracting units 150-1, 150-2, 150-3 in the current control unit A, respectively, and the phase current command values Iaref, Ibref, Icref Currents Ia, Ib, and Ic are reduced. Deviations (Iaref-Ia), (Ibref-Ib), and (Icref-Ic) as outputs of the subtracting units 150-1, 150-2, and 150-3 are proportional integral control units (PI control units) 152-1, 152. As a result, the voltage command values Varef, Vbref, and Vcref for each phase are output.

  Next, the voltage command values Varef, Vbref, and Vcref of each phase are input to the average value calculation unit 10, and the average value Vav = 1/3 (Varef + Vbref + Vcref) is calculated. The voltage command values Varef, Vbref, Vcref and average value Vav of each phase are input to the subtraction units 12-1, 12-2, 12-3, respectively, and as a result of the subtraction, the subtraction units 12-1, 12-2, 12 -3 outputs (Varef-Vav), (Vbref-Vav), and (Vcref-Vav).

  On the other hand, the counter electromotive voltages eaf, ebf, and ecf of each phase are a-phase counter-electromotive voltage calculation unit 20-1, b-phase counter-electromotive voltage calculation unit 20-2, and c-phase counter-electromotive voltage calculation as counter-electromotive voltage calculation means, respectively. Estimated by the unit 20-3. First, a-phase motor terminal voltage Va, b-phase motor terminal voltage Vb, and c-phase motor terminal voltage Vc are detected by voltage detectors 220-1, 220-2, and 220-3, respectively. Then, the current Ia and the voltage Va are input to the a-phase counter electromotive voltage calculation unit 20-1, and as a result, the counter electromotive voltage eaf is output. Similarly, the current Ib and the voltage Vb are input to the b-phase counter electromotive voltage calculation unit 20-2, and as a result, the counter electromotive voltage ebf is output. The voltage Vc is input, and as a result, the back electromotive voltage ecf is output.

  Specific contents of the a-phase counter electromotive voltage calculation unit 20-1, the b-phase counter electromotive voltage calculation unit 20-2, and the c-phase counter electromotive voltage calculation unit 20-3 will be described with reference to FIG. It is calculated from the equation V = EMF + (R + s · L) · I indicating the relationship between the voltage and current of the motor. By transforming this equation, the back electromotive force EMF is calculated as EMF = V− (R + s · L) · I. Specifically, in FIG. 3, the a-phase current Ia is input to the transfer function unit 22-1 (the transfer function is (Ra + s · La) / (s · Tf + 1)). The denominator of the transfer function means a low-pass filter (LPF) for absorbing noise or the like in current detection. Then, in the subtraction unit 24-1, the output of the transfer function unit 22-1 is subtracted from the terminal voltage Va of the motor, so that the a-phase counter electromotive voltage eaf becomes eaf = Va− (Ra + s · La) · Ia / ( s · Tf + 1).

  Similarly, the b-phase counter electromotive voltage ebf is input to the transfer function unit 22-2, the output and the b-phase terminal voltage Vb are input to the subtracting unit 24-2, and the counter electromotive voltage ebf is output. Similarly, the c-phase counter electromotive voltage ecf is input to the transfer function unit 22-3, the output thereof and the c-phase terminal voltage Vc are input to the subtracting unit 24-3, and the counter electromotive voltage ecf is output.

  Since an example of a specific procedure for estimating the back electromotive force of each phase has been described with reference to FIG. 3, the description returns to the control block diagram of FIG. In the adding unit 14-1, the counter electromotive voltage eaf, which is the output of the a-phase counter electromotive voltage calculation unit 20-1, is added to (Varef-Vav), which is the output of the subtracting unit 12-1. Voltage command value (Varef−Vav + eaf) is output. Similarly, with respect to the b phase, the b phase counter electromotive force estimated by the b phase counter electromotive voltage calculation unit 20-2 with respect to (Vbref−Vav) that is the output of the subtraction unit 14-2 in the adding unit 14-2. The voltage ebf is added, and a new b-phase voltage command value (Vbref−Vav + ebf) is output. Similarly, for the c-phase, the c-phase counter electromotive force estimated by the c-phase counter electromotive voltage calculation unit 20-3 with respect to (Vcref−Vav), which is the output of the subtraction unit 14-3, in the adding unit 14-3. The voltage ecf is added, and a new c-phase voltage command value (Vcref−Vav + ecf) is output.

  Then, Vbat / 2, which is half of the battery voltage value indicated by the target neutral point potential setting unit 40, is a voltage command value for each new phase in the adding units 15-1, 15-2, and 15-3 (Varef). -Vav + eaf), (Vbref-Vav + ebf), and (Vcref-Vav + ecf) are added to (Vref-Vav + eaf + Vbat / 2), (Vbref-Vav + ebf + Vbat / 2), and (Vcref-Vav + ec / cf + cf + cf + cf + cf + cf + cf + cf + cf2). These outputs from the adders 15-1, 15-2, and 15-3 are input to the PWM controller 210, converted into PWM control signals, and the inverter circuit 211 supplies current to the motor 108.

  That is, the final voltage command value of each phase that is an input of the PWM control unit 210 is (Vref−Vav + eaf + Vbat / 2), (Vbref−Vav + ebf + Vbat / 2), (Vcref−Vav + ecf + Vbat / 2), A value indicated by the equation is input to the PWM control unit 210. As a result, the neutral point potential Vn of the motor 108 is controlled to be maintained at Vbat / 2, which is half the battery voltage value. That is, in the motor control, the neutral point potential is stabilized and the torque ripple and noise of the motor are suppressed.

  In the above description, the embodiment in which the back electromotive voltage of the motor is calculated from the terminal voltage and current of the motor has been described. However, the back electromotive voltage can also be estimated from the rotation angle θ as the motor position and the angular velocity ω of the motor. The embodiment will be described with reference to FIG. The rotation angle θ and the angular velocity ω output from the rotation angle detector 202 are input to the a-phase counter electromotive voltage calculation unit 30-1. The counter electromotive voltage induced by the rotational angle θ indicating the rotor position of the motor is as follows: If the rotational speed N [rpm] = ω × 60 / 2π calculated from the angular speed ω [rad / s] is 1000 [rpm], Since it is determined by the characteristics of the motor, the normalized back electromotive force eafk at 1000 rpm can be estimated by inputting to the standardized back electromotive force calculating unit 32-1 prepared as a table. Next, in order to correct the counter electromotive voltage in consideration of the rotational speed N [rpm] with respect to the standardized counter electromotive voltage eafk at 1000 [rpm], the detected angular speed ω is converted into the rotational speed N, and Then, the normalized back electromotive force eafk is input to the rotation speed correction unit 34-1 that multiplies the rotation speed N by a relative speed (N / 1000) based on 1000 [rpm]. Electromotive voltage eaf = eafk × (N / 1000) is calculated.

  Next, in order to calculate the b-phase counter electromotive voltage ebf, the rotation angle θ as the motor position and the motor angular velocity ω are input to the b-phase counter electromotive voltage calculation unit 30-2. Since the rotation angle of the b phase is 120 degrees behind the a phase, the input rotation angle θ is input to the phase shift unit 36-2 that delays the 120 ° phase, It is output as a certain rotation angle (θ−120). Thereafter, like the a phase, it is input to the normalized counter electromotive voltage calculation unit 32-2, and the output ebfk is input to the rotation speed correction unit 34-2 and output as the b phase counter electromotive voltage ebf. The c-phase counter electromotive force ecf is calculated because the rotation angle of the c-phase is 120 ° ahead of the a-phase, so that the input rotation angle θ is advanced by 120 ° to the phase shift unit 36-3. It is input and output as a rotation angle (θ + 120) which is the rotation angle of the c phase. Thereafter, like the a phase, it is input to the normalized counter electromotive voltage calculation unit 32-3, and its output ecfk is input to the rotation speed correction unit 34-3 and output as the c phase counter electromotive voltage ecf.

  The back electromotive voltage of each phase can be estimated from the terminal voltage and current of the motor, and can also be estimated from the rotation angle and angular velocity of the motor. It is possible to select in consideration of accuracy, detection speed, etc. depending on the performance of the installed detector. Further, it is needless to say that the counter electromotive voltage can be detected by a method other than the above-described method, and the present invention can be applied even if the counter electromotive voltage detected by the method is used.

  Details of the embodiment in the case where the proportional integral control unit 152 of FIG. 2 is specifically digitally processed will be described with reference to FIG. FIG. 5A is a block diagram when the control block diagram shown in FIG. 2 is directly digitally processed. As a procedure for calculating the average value of the voltage command values of each phase and subtracting the average value from the voltage command value of each phase, in FIG. 5A, in the PI control units 152-1, 152-2, and 152-3. Voltage command values Varef, Vbref, and Vcref, which are the outputs of the output, are calculated first, and then the average value Vav is calculated through the adder 10-1 and the divider 10-2, and the subtractors 12-1, 12-2 12-3, the average value Vav is subtracted from the voltage command values Varef, Vbref, Vcref of each phase. If digital processing is executed in this procedure, an overflow occurs in the middle of processing, or a value sticks to the limiter, which is not preferable. Therefore, as shown in FIG. 5B, an average value is calculated in the middle of the proportional integration process, and a process of subtracting the average value from the value of each phase is also calculated in the middle of the proportional integration process.

  That is, for example, the average values of the output values of the a-phase subtracting unit 154-1, the b-phase subtracting unit 154-2, and the c-phase subtracting unit 154-3 are added to the adding unit 10-1 and the dividing unit 10-2. Further, immediately after that, the subtracting units 12-1, 12-2, and 12-3 execute processing for subtracting the average value from the outputs of the subtracting units 154-1, 154-2, and 154-2. That is, the output values of the subtraction units 12-1, 12-2, 12-3 are compared with the output values of the subtraction units 12-1, 12-2, 12-3 in FIG. The output values of the subtracting units 12-1, 12-2 and 12-3 in (B) are small as absolute values, and the possibility of causing an overflow or sticking to the limiter is reduced. FIG. 5 shows an embodiment of the present invention. The configuration of (B) is a preferable configuration.

  In the above-described embodiment, the current control unit A as the current control means is in the case of feedback control. However, the present invention can also be applied to the case where the current control means is a combination of feedforward control and a disturbance observer. With reference to FIG. 6, the Example which applied this invention to the combination with feedforward control and a disturbance observer is described. Only the current control unit A is different from the feedback control embodiment described above, and the calculation procedure of the back electromotive voltages eaf, ebf, and ecf of each phase is the same as that of the feedback control embodiment. Only the control unit A will be described in detail.

  The voltage command values Varef, Vbref, and Vcref for each phase are output to the outputs of the feedforward (FF) control units 160-1, 160-2, and 160-3 that receive the current command values Iaref, Ibref, and Icref for each phase. Can be obtained by adding the outputs of the disturbance observer computing units 164-1, 164-2, and 164-3 in the adding units 162-1, 162-2, and 163-3.

  For example, the current Ia and the voltage command value Varef detected by the current detector 205-1 are input to the a-phase disturbance observer calculation unit 164-1. The current Ia is input to the transfer function unit 167-1 in the disturbance observer calculation unit 164-1. This transfer function is (Ra + s · La) / (1 + s · Tf), and the first-order lag function of the denominator represents LPF for noise absorption at the time of current detection. On the other hand, the transfer function of the transfer function unit 165-1 having the voltage command value Varef as an input is (K / 1 + s · Tf). The denominator (1 + s · Tf) is provided to balance the first-order lag and the time lag of the transfer function unit 167-1. The subtraction unit 166-1 performs subtraction between the output of the transfer function unit 165-1 and the output of the transfer function unit 167-1, and the output becomes the output Vadis of the disturbance observer calculation unit 164-1.

  Similarly, the current Ib and the voltage command value Vbref which are the outputs of the polarity reversing unit 205-4 are input to the b-phase disturbance observer calculation unit 164-2. Similarly to the a-phase, the signals are input to the transfer function unit 167-2 and the transfer function unit 165-2, their outputs are input to the subtracting unit 166-2, and the output of the subtracting unit 166-2 is the disturbance observer calculating unit 164. -2 output Vbdis.

  Similarly, the current Ic and the voltage command value Vcref which are the outputs of the current detector 205-2 are input to the c-phase disturbance observer calculation unit 164-3. Similarly to the a phase, the signals are input to the transfer function unit 167-3 and the transfer function unit 165-3, their outputs are input to the subtraction unit 166-3, and the output of the subtraction unit 166-3 is the disturbance observer calculation unit 164. -3 output Vcdis.

  FF control units 160-1 and 160-2 that receive disturbance values Vadis, Vbdis, and Vcdis that are outputs of the disturbance observer calculation units 164-1, 164-2, and 164-3 and current command values Iaref, Ibref, and Icref as inputs. , 160-3 are added by the adders 162-1, 162-2, and 162-3, respectively, to calculate voltage command values Varef, Vbref, and Vcref.

  For the voltage command values of each phase determined in this way, the average value Vav is calculated by the average value calculation unit 10, and the average value Vav is subtracted from each phase voltage command value by the subtraction units 12-1 and 12. Subtraction is performed at −2 and 12-3, respectively, and the back electromotive voltages eaf, ebf, and ecf of the respective phases are added to the outputs by the adders 14-1, 14-2, and 14-3, respectively. It was obtained by adding the target value Vbat / 2 indicated by the target neutral point potential setting unit 40 for obtaining the neutral point potential in consideration of the battery voltage by the adding units 15-1, 15-2, and 15-3, respectively. The resulting voltage command values for each phase are (Varef−Vav + eaf + Vbat / 2), (Vbref−Vav + ebf + Vbat / 2), and (Vcref−Vav + ecf + Vbat / 2), which are values satisfying the equation (10).

  As a result, even when the current control unit A is a combination configuration of feedforward control and disturbance observer, it is possible to control the neutral point potential of the motor, which is the effect of the present invention, to be constant, and thus the torque of the motor. Motor control that suppresses ripple is possible. In applying the present invention, the configuration of the current control unit A is not limited to the combination of the above-described feedback control, feedforward control, and disturbance observer.

  As described above, by using the present invention, the neutral point potential of the motor can be constantly stabilized without using a special sensor and without increasing the burden of arithmetic processing such as a CPU. Therefore, motor control that suppresses torque ripple and noise of the motor is possible.

  Further, in the case of a three-phase motor, the neutral point exists in the case of star connection, but does not actually exist in the case of delta connection. However, even if the present invention is applied to the motor control of the delta connection, the potential of the virtual neutral point in the case of the delta / star equivalent conversion in which the three terminals of the delta connection are regarded as the three terminals of the star connection is stabilized. An effect is exhibited and interference of each phase current can be suppressed. Therefore, even in the case of a delta-connected motor that does not actually have a neutral point, motor control that can suppress torque ripple is possible.

  In the above description of the embodiment, a three-phase motor has been described. However, the present invention can also be applied to three or more multi-phase motors.

  Furthermore, if the present invention is used, the neutral point potential can be stabilized in the motor control and the torque ripple can be suppressed. Therefore, in the electric power steering device using the motor, the electric power steering can be operated with a low noise and a good feeling. The effect which can provide the control apparatus of an apparatus can be anticipated.

It is a basic control block diagram of the present invention. It is an Example in case electric current control is feedback control. It is a figure which shows the back electromotive force calculation using the terminal voltage and electric current of a motor. It is a figure which shows the back electromotive force calculation using the rotation angle and angular velocity of a motor. It is an Example which shows the digital processing of proportional integral control. It is an Example in case electric current control is feedforward control and a disturbance observer. It is a block diagram of an electric power steering device. It is a figure which shows the relationship between the voltage of a motor, an electric current, the inductance value of a coil | winding, and resistance value.

Explanation of symbols

10 Average Value Calculation Unit 12 Subtraction Unit 14 Addition Unit 20 Back Electromotive Voltage Calculation Unit (Voltage, Current)
30 Back electromotive force calculation unit (angular velocity, rotational position)
40 Target neutral point potential setting unit 15 Addition unit 204 Current command value calculation unit A Current control unit 150 Subtraction unit 152 Proportional integration control unit (PI control unit)
160 Feedforward control unit (FF control unit)
164 Disturbance observer calculation unit

Claims (6)

A multiphase motor for applying a steering assist force to a steering system of a vehicle, a current command value calculation means for calculating a current command value for each phase to the motor, and a current detection means for detecting a current for each phase of the motor And a control device for an electric power steering apparatus comprising: current control means for calculating a voltage command value for each phase based on the current command value and the detected current;
Back electromotive force calculating means for estimating a counter electromotive voltage value induced in the motor, subtracting an average value of the voltage command values of the phases from the voltage command values of the phases, and further back electromotive force of the phases. A control device for an electric power steering apparatus, wherein control is performed based on a voltage command value for each new phase obtained by adding voltage values. The control device for an electric power steering apparatus according to claim 1, wherein the current control means is a feedback control means. The control device for an electric power steering apparatus according to claim 1, wherein the current control means includes a feedforward control means and a disturbance observer. 4. The control device for an electric power steering apparatus according to claim 1, wherein the back electromotive voltage is estimated from a terminal voltage and a current of the motor. 4. The control device for an electric power steering apparatus according to claim 1, wherein the counter electromotive voltage is estimated from an angular velocity and a motor position of the motor. The control device for an electric power steering apparatus according to any one of claims 1 to 5, wherein the motor is a three-phase motor and a connection method is a star connection or a delta connection.

Images