JP6809341B2 - Rotating machine control device - Google Patents

Description

The present invention relates to a control device for a rotary electric machine.

Conventionally, as seen in Patent Document 1 below, for example, a control device that estimates the rotational position information of a rotary electric machine based on the current value of the rotating coordinate system flowing in the winding group wound around the stator of the rotary electric machine is known. Has been done. Further, as a control device, a device that drives a rotary electric machine in a rectangular wave is also known.

Japanese Patent No. 6022951

When the square wave drive is performed, the current value of the rotating coordinate system flowing through the winding group includes a harmonic component in addition to the fundamental wave component. Specifically, for example, 6th-order and 12th-order harmonic components are included. If the rotation position information is estimated based on the current value of the rotating coordinate system including the harmonic component, the estimation accuracy of the rotation position information may decrease. As a result, the controllability of the rotary electric machine may decrease.

An object of the present invention is to provide a control device for a rotary electric machine capable of suppressing a decrease in estimation accuracy of rotation position information.

The first invention is applied to a system including a rotary electric machine having a winding group wound around a stator and a power conversion circuit for applying a current to the winding group, and obtains rotation position information of the rotary electric machine. In the control device of the rotary electric machine including the position estimation unit for estimating, the current acquisition unit that acquires the current value of the fixed coordinate system flowing in the winding group, the current value acquired by the current acquisition unit, and the rotation position information. Based on the current calculation unit that calculates the conversion current value, which is the current value of the rotation coordinate system flowing in the winding group, and the power conversion to perform the rectangular wave drive of the rotation electric machine based on the rotation position information. The position estimation unit includes an operation unit for operating the circuit, the position estimation unit estimates the rotation position information based on the conversion current value calculated by the current calculation unit, and the current calculation unit estimates the position estimation unit. As the conversion current value used in the above, the average value of the conversion current value in one fluctuation cycle is calculated.

In the first invention, the current value of the rotating coordinate system flowing through the winding group is defined as the conversion current value. In the first invention, the current calculation unit calculates the average value of the converted current value in one fluctuation cycle based on the current value acquired by the current acquisition unit and the rotation position information. The above average value is a value in which the harmonic component included in the actual conversion current value is reduced. Therefore, since the rotation position information is estimated based on the above-mentioned average value, it is possible to suppress a decrease in the estimation accuracy of the rotation position information due to the harmonic component.

The second invention comprises a voltage acquisition unit that acquires an input voltage value of the power conversion circuit, a speed calculation unit that calculates the electric angle frequency of the rotating electric machine based on the rotation position information, and the voltage acquisition unit. Based on the acquired input voltage value and the electric angle frequency calculated by the speed calculation unit, the timing at which the actual conversion current value is assumed to be the average value in one fluctuation cycle is calculated as the detection timing. The current acquisition unit acquires the current value of the fixed coordinate system at the detection timing calculated by the timing calculation unit, and the current calculation unit is based on the rotation position information. Then, the current value acquired by the current acquisition unit is converted into the conversion current value used by the position estimation unit.

The amplitude and phase of the current value in the rotating coordinate system change according to the input voltage value of the power conversion circuit and the electric angular frequency of the rotating electric machine. Therefore, when the current value acquired by the current acquisition unit is a current value detected at a predetermined timing regardless of the magnitude of the input voltage value and the electric angular frequency, the current value acquired by the current acquisition unit. The difference between the conversion current value based on the above and the average value of the actual conversion current value in one fluctuation cycle becomes large. When this deviation becomes large, the harmonic component included in the converted current value becomes large, and the estimation accuracy of the rotation position information may decrease.

Therefore, in the second invention, the timing calculation unit averages the actual conversion current value in one fluctuation cycle based on the input voltage value acquired by the voltage acquisition unit and the electric angular frequency calculated by the speed calculation unit. The timing that is expected to be a value is calculated as the detection timing. The current acquisition unit acquires the current value of the fixed coordinate system at the calculated detection timing. The current calculation unit converts the current value acquired by the current acquisition unit into the conversion current value used in the position estimation unit based on the rotation position information. Therefore, the conversion current value used in the position estimation unit can be brought close to the average value, and the harmonic component included in the conversion current value can be reduced. As a result, it is possible to suppress a decrease in the estimation accuracy of the rotation position information.

The third invention comprises a voltage acquisition unit that acquires an input voltage value of the power conversion circuit, a speed calculation unit that calculates the electric angle frequency of the rotating electric machine based on the rotation position information, and the voltage acquisition unit. An average value calculation unit that calculates the average value based on the acquired input voltage value and the electric angle frequency calculated by the speed calculation unit, and the average value calculated by the average value calculation unit. Based on the conversion current value calculated by the current calculation unit, a correction amount for converting the conversion current value calculated by the current calculation unit into the average value calculated by the average value calculation unit is calculated. The correction amount calculation unit and the correction unit for correcting the conversion current value calculated by the current calculation unit with the correction amount are provided, and the position estimation unit includes the conversion current value corrected by the correction unit. The rotation position information is estimated based on.

The amplitude and phase of the current value in the rotating coordinate system change according to the input voltage value of the power conversion circuit and the electric angular frequency of the rotating electric machine. Therefore, the deviation between the converted current value based on the current value acquired by the current acquisition unit and the average value of the actual converted current value in one fluctuation cycle can be large. When this deviation becomes large, the harmonic component included in the converted current value becomes large, and the estimation accuracy of the rotation position information may decrease.

Therefore, in the third invention, the average value calculation unit calculates the average value based on the input voltage value and the electric angular frequency. The correction amount calculation unit calculates the correction amount based on the calculated average value and the conversion current value calculated by the current calculation unit. The correction unit corrects the conversion current value calculated by the current calculation unit with the calculated correction amount. Therefore, the conversion current value used in the position estimation unit can be brought close to the average value, and the influence of the harmonic component included in the conversion current value on the estimation accuracy of the rotation position information can be reduced. As a result, it is possible to suppress a decrease in the estimation accuracy of the rotation position information.

In the fourth invention, a period shorter than the period from the first timing at which the voltage vector of the power conversion circuit is switched to the second timing at which the voltage vector is switched immediately after the first timing is defined as the detection prohibition period. The current acquisition unit acquires the current value of the fixed coordinate system in a period other than the period from the first timing to the elapse of the detection prohibition period.

In the detection prohibition period, which is the period immediately after the voltage vector is switched, the noise caused by the switching may be included in the current value of the fixed coordinate system. If the rotation position information is estimated based on the current value including noise, the estimation accuracy of the rotation position information may decrease. Therefore, in the fourth invention, the current value of the fixed coordinate system used in the current calculation unit is set to the current value in a period other than the period from the first timing to the elapse of the detection prohibition period. Therefore, it is possible to suppress the noise caused by the switching of the voltage vector from being included in the current value acquired by the current acquisition unit, and it is possible to suppress the deterioration of the estimation accuracy of the rotation position information.

In the fifth invention, a period shorter than the period from the first timing at which the voltage vector of the power conversion circuit is switched to the second timing at which the voltage vector is switched immediately after the first timing is defined as the detection prohibition period. ing. A fifth invention comprises a voltage acquisition unit that acquires an input voltage value of the power conversion circuit, a speed calculation unit that calculates the electric angle frequency of the rotating electric machine based on the rotation position information, and the voltage acquisition unit. Based on the acquired input voltage value and the electric angle frequency calculated by the speed calculation unit, the timing at which the actual conversion current value is assumed to be the average value in one fluctuation cycle is calculated as the detection timing. The timing calculation unit, the determination unit for determining that the detection timing calculated by the timing calculation unit is included in the period from the first timing to the elapse of the detection prohibition period, and the determination unit. When it is determined that the current acquisition unit is included, the current acquisition unit includes a shift unit that shifts the detection timing to a period other than the period from the first timing to the elapse of the detection prohibition period. The current value of the fixed coordinate system at the detection timing shifted by the shift unit is acquired, and the current calculation unit converts the current value acquired by the current acquisition unit into the conversion current value based on the rotation position information. Convert to. A fifth invention includes an average value calculation unit that calculates the average value based on the input voltage value acquired by the voltage acquisition unit and the electric angle frequency calculated by the speed calculation unit, and the average value calculation unit. The conversion current value calculated by the current calculation unit was calculated by the average value calculation unit based on the average value calculated by the value calculation unit and the conversion current value calculated by the current calculation unit. The position estimation unit includes a correction amount calculation unit that calculates a correction amount for obtaining the average value, and a correction unit that corrects the conversion current value calculated by the current calculation unit with the correction amount. The rotation position information is estimated based on the conversion current value corrected by the correction unit.

The amplitude and phase of the current value in the rotating coordinate system change according to the input voltage value of the power conversion circuit and the electric angular frequency of the rotating electric machine. Therefore, when the current value acquired by the current acquisition unit is a current value detected at a predetermined timing regardless of the magnitude of the input voltage value and the electric angular frequency, the current value acquired by the current acquisition unit. The difference between the converted current value converted by and the average value of the actual converted current value in one fluctuation cycle becomes large. When this deviation becomes large, the harmonic component included in the conversion current value used for estimating the rotation position information becomes large, and the estimation accuracy of the rotation position information may decrease.

Therefore, in the fifth invention, the timing calculation unit averages the actual conversion current value in one fluctuation cycle based on the input voltage value acquired by the voltage acquisition unit and the electric angular frequency calculated by the speed calculation unit. The timing that is expected to be a value is calculated as the detection timing. Here, in the detection prohibition period, which is the period immediately after the voltage vector is switched, the noise caused by the switching may be included in the current value of the fixed coordinate system. If the rotation position information is estimated based on the current value including noise, the estimation accuracy of the rotation position information may decrease.

Therefore, in the fifth invention, the determination unit determines that the detection timing calculated by the timing calculation unit is included in the period from the first timing to the elapse of the detection prohibition period. When it is determined by the determination unit that the shift unit is included, the shift unit shifts the detection timing to a period other than the period from the first timing to the elapse of the detection prohibition period.

Here, the conversion current value based on the current value of the fixed coordinate system at the shifted detection timing can greatly deviate from the average value in one fluctuation cycle of the actual conversion current value. In this case, the harmonic component included in the converted current value becomes large, and the estimation accuracy of the rotation position information may decrease.

Therefore, in the fifth invention, the average value calculation unit calculates the average value based on the input voltage value and the electric angular frequency. The correction amount calculation unit calculates the correction amount based on the calculated average value and the conversion current value calculated by the current calculation unit. The correction unit corrects the conversion current value calculated by the current calculation unit with the calculated correction amount. Therefore, the conversion current value used in the position estimation unit can be brought close to the average value, and the influence of the harmonic component included in the conversion current value on the estimation accuracy of the rotation position information can be reduced. As a result, it is possible to suppress a decrease in the estimation accuracy of the rotation position information.

The overall block diagram of the motor control system which concerns on 1st Embodiment. The figure which shows the spatial phase difference formed by the 1st winding group and the 2nd winding group. The figure which shows the transition of the voltage vector in the rectangular wave control. The figure which shows the relationship between a voltage vector and a switching mode. Block diagram of motor control in a control device. A time chart showing the current detection timing. The flowchart which shows the procedure of a motor control process. The overall block diagram of the motor control system which concerns on 2nd Embodiment. A time chart showing the current detection timing. The flowchart which shows the procedure of a motor control process. The figure which shows the transition of the modeled voltage which concerns on 3rd Embodiment. The flowchart which shows the procedure of a motor control process. The flowchart which shows the procedure of the motor control processing which concerns on 4th Embodiment. The flowchart which shows the procedure of the 1st correction processing. The flowchart which shows the procedure of the 2nd correction process. The flowchart which shows the procedure of the motor control processing which concerns on 5th Embodiment. The figure which shows the detection prohibition period. A time chart showing the current detection timing according to other embodiments.

<First Embodiment>
Hereinafter, a first embodiment in which the control device according to the present invention is applied to a vehicle having an engine as an in-vehicle main engine will be described with reference to the drawings.

As shown in FIG. 1, the motor 10 is a rotary electric machine having a multi-phase multiple winding, and in the present embodiment, it is a salient pole type synchronous motor having a three-phase double winding. In this embodiment, the motor 10 is assumed to be an ISG (Integrated Starter Generator) in which the functions of the starter and the alternator are integrated. In particular, in the present embodiment, in addition to the initial start of the engine (not shown), the engine is automatically stopped when a predetermined automatic stop condition is satisfied, and then the engine is automatically stopped when a predetermined restart condition is satisfied. The motor 10 also functions as a starter when the idling stop function for restarting is executed.

In this embodiment, the motor 10 is of the permanent magnet field type. The rotor 12 constituting the motor 10 is capable of transmitting power to the crankshaft of the engine.

The stator 13 constituting the motor 10 is wound with two armature winding groups, the first winding group 10A and the second winding group 10B. The rotor 12 is shared with respect to the first and second winding groups 10A and 10B. In the present embodiment, the first winding group 10A corresponds to the first target winding group, and the second winding group 10B corresponds to the second target winding group.

Each winding group 10A and 10B will be described with reference to FIG. Each of the first winding group 10A and the second winding group 10B has a three-phase winding having a different neutral point. The first winding group 10A has windings UA, VA, and WA displaced by 120 ° from each other in the electric angle, and the second winding group 10B has windings UB, VB, WB displaced by 120 ° from each other by the electric angle. have. In the present embodiment, the spatial phase difference Δα, which is the electric angle formed by the first winding group 10A and the second winding group 10B, is set to 30 °. Further, in the present embodiment, the first winding group 10A and the second winding group 10B have the same configuration. Specifically, the number of turns NA of each of the windings UA, VA, and WA constituting the first winding group 10A, and the number of turns NB of each of the windings UB, VB, and WB constituting the second winding group 10B. Are set equally. As a result, the self-inductance of the first winding group 10A and the self-inductance of the second winding group 10B are equalized.

Returning to FIG. 1, the motor 10 is electrically connected to the first inverter 20A corresponding to the first winding group 10A and the second inverter 20B corresponding to the second winding group 10B. A battery 21, which is a common DC power source, is connected in parallel to each of the first inverter 20A and the second inverter 20B. A capacitor 22 is connected in parallel to the battery 21.

The first inverter 20A includes three sets of a series connection body of the first U, V, W phase upper arm switches SUp1, SVp1, SWp1 and the first U, V, W phase lower arm switches SUn1, SVn1, SWn1. The connection points of the series connection bodies in the U, V, and W phases are connected to the U, V, and W phase windings UA, VA, and WA constituting the first winding group 10A. In this embodiment, IGBTs are used as the switches SUP1 to SWn1. Diodes DUp1 to DWn1 are connected in antiparallel to each of the switches SUp1 to SWn1. The switches SUP1 to SWn1 are not limited to IGBTs, and for example, N-channel MOSFETs may be used.

Similar to the first inverter 20A, the second inverter 20B is connected in series with the second U, V, W phase upper arm switches SUp2, SVp2, SWp2 and the second U, V, W phase lower arm switches Sun2, SVn2, SWn2. It has three sets of bodies. The connection points of the series connection bodies in the U, V, and W phases are connected to the U, V, W phase windings UB, VB, and WB constituting the second winding group 10B. In this embodiment, an IGBT is used as each switch SUP2 to SWn2. And the diodes DUp2 and DWn2 are connected in antiparallel to each of the switches SUp2 and SWn2, respectively. The switches SUP2 to SWn2 are not limited to IGBTs, and for example, N-channel MOSFETs may be used.

The positive electrode terminal of the battery 21 is connected to the collector of each upper arm switch, which is the high potential side terminal of the first and second inverters 20A and 20B. The negative electrode terminal of the battery 21 is connected to the emitter of each lower arm switch which is the low potential side terminal of the first and second inverters 20A and 20B.

The control system includes a voltage detection unit 30, a first-phase current detection unit 31A, a second-phase current detection unit 31B, and a temperature detection unit 32. The voltage detection unit 30 detects the input voltage value from the battery 21 to the first and second inverters 20A and 20B as the power supply voltage value VDC. The first-phase current detection unit 31A detects at least two phases of the current values of each phase of the first winding group 10A in the three-phase fixed coordinate system as the first-phase current value Iph1. The second-phase current detection unit 31B detects at least two phases of the current values of each phase of the second winding group 10B in the three-phase fixed coordinate system as the second-phase current value Iph2. The temperature detection unit 32 detects the temperature of the first winding group 10A as the first winding temperature TH1 and the temperature of the second winding group 10B as the second winding temperature TH2. In the present embodiment, the first phase current value Iph1 corresponds to the first current value, and the second phase current value Iph2 corresponds to the second current value.

The detection value of each of the above detection units is input to the control device 40. The control device 40 generates operation signals of the first and second inverters 20A and 20B based on the detection values of the detection units in order to control the control amount of the motor 10 to the command value. In the present embodiment, the control amount is torque, and the command value thereof is command torque Trq *. In FIG. 1, signals for operating the switches SUp1 to SWn1 of the first inverter 20A are shown as first operation signals gUp1 to gWn1, and signals for operating the switches SUp2 to SWn2 of the second inverter 20B are shown as second operation signals gUp2. It is shown as ~ gWn2.

The function provided by the control device 40 can be provided, for example, by software recorded in a substantive memory device, a computer that executes the software, hardware, or a combination thereof.

The control device 40 generates each operation signal based on the rectangular wave drive control. As shown in FIG. 3, the square wave drive control is a control in which the effective voltage vectors V1 to V6 are sequentially switched every 60 ° of the electric angle with 360 ° of the electric angle θe as one cycle. In FIG. 3, Vct1 indicates the first voltage vector of the first inverter 20A, and Vct2 indicates the second voltage vector of the second inverter 20B. Further, FIG. 4 shows the correspondence between each voltage vector and the operating state of each phase switch. In the figure, V0 and V7 indicate invalid voltage vectors.

Subsequently, the torque control of the motor 10 will be described with reference to FIG. In this control, position sensorless control is used. In the position sensorless control, the detection value of an angle detector such as a resolver that directly detects the electric angle which is the magnetic pole position of the motor 10 is not used, but the estimated electric angle is used.

The control device 40 includes a first processing unit 41 corresponding to the first inverter 20A. In the first processing unit 41, the first current conversion unit 41a has a first estimated angle θγ1 which is an estimated value of the electric angle corresponding to the first winding group 10A, and a first phase current detecting unit 31A detected by the first phase current detecting unit 31A. Based on the one-phase current value Iph1, the U, V, W phase current values of the first winding group 10A in the three-phase fixed coordinate system are set to the first γ-axis current value Iγ1r and the first δ-axis current value Iδ1r in the γδ coordinate system. Convert to. Here, the γδ coordinate system is an estimated coordinate system of the dq coordinate system, which is a two-phase rotating coordinate system of the motor 10.

The first command current setting unit 41b sets the first γ-axis command current value Iγ1 * and the first δ-axis command current value Iδ1 * based on the command torque Trq *. The first γ-axis deviation calculation unit 41c calculates the first γ-axis deviation ΔIγ1 as a value obtained by subtracting the first γ-axis current value Iγ1r from the first γ-axis command current value Iγ1 *. The 1st δ axis deviation calculation unit 41d calculates the 1st δ axis deviation ΔIδ1 as a value obtained by subtracting the 1st δ axis current value Iδ1r from the 1st δ axis command current value Iδ1 *.

The first command voltage setting unit 41e uses the first γ-axis command voltage value Vγ1 as an operation amount for feedback-controlling the first γ-axis current value Iγ1r to the first γ-axis command current value Iγ1 * based on the first γ-axis deviation ΔIγ1. * Calculate. Further, the first command voltage setting unit 41e uses the first δ-axis command voltage as an operation amount for feedback-controlling the first δ-axis current value Iδ1r to the first δ-axis command current value Iδ1 * based on the first δ-axis deviation ΔIδ1. Calculate the value Vδ1 *. As the feedback control, for example, proportional integration control can be used.

The first voltage conversion unit 41f is γδ based on the first γ-axis command voltage value Vγ1 *, the first δ-axis command voltage value Vδ1 *, the power supply voltage value VDC detected by the voltage detection unit 30, and the first estimated angle θγ1. The first γ, δ-axis command voltage values Vγ1 *, Vδ1 * in the coordinate system are converted into the first U, V, W phase command voltage values VU1, VV1, VW1 in the three-phase fixed coordinate system. In the present embodiment, the first U, V, and W phase command voltage values VU1, VV1, and VW1 are sinusoidal signals having a median value of 0 and a phase shift of 120 ° at the electrical angle.

The first generation unit 41g generates the first operation signals gUp1 to gWn1 based on the first U, V, W phase command voltage values VU1, VV1, VW1 output from the first voltage conversion unit 41f. The first generation unit 41g outputs the generated first operation signals gUp1 to gWn1 to each switch SUp1 to SWn1. Here, the first operation signal may be generated by, for example, PWM control based on a magnitude comparison between a carrier signal such as a triangular wave signal and each phase command voltage value VU1, VV1, VW1. The first operation signal on the upper arm side and the corresponding first operation signal on the lower arm side are complementary signals to each other. Therefore, the upper arm switch and the corresponding lower arm switch are turned on alternately.

The control device 40 includes a second processing unit 42 corresponding to the second inverter 20B. Since the second processing unit 42 has the same configuration as the first processing unit 41, detailed description thereof will be omitted as appropriate.

In the second processing unit 42, the second current conversion unit 42a has a second estimated angle θγ2 which is an estimated value of the electric angle corresponding to the second winding group 10B, and a second phase current detecting unit 31B detected by the second phase current detecting unit 31B. Based on the two-phase current Iph2, the U, V, and W phase current values corresponding to the second winding group 10B are converted into the second γ-axis current value Iγ2r and the second δ-axis current value Iδ2r in the γδ coordinate system. .. Here, the second estimated angle θγ2 is calculated by the angle adding unit 51 as an added value of the first estimated angle θγ1 and the spatial phase difference Δα.

The second command current setting unit 42b sets the second γ-axis command current value Iγ2 * and the second δ-axis command current value Iδ2 * based on the command torque Trq *. The command current values Iγ2 *, Iδ2 *, Iγ1 *, and Iδ1 * are set to values required for the actual torque of the motor 10 to be the command torque Trq *.

The second γ-axis deviation calculation unit 42c calculates the second γ-axis deviation ΔIγ2 as a value obtained by subtracting the second γ-axis current value Iγ2r from the second γ-axis command current value Iγ2 *. The second δ-axis deviation calculation unit 42d calculates the second δ-axis deviation ΔIδ2 as a value obtained by subtracting the second δ-axis current value Iδ2r from the second δ-axis command current value Iδ2 *.

The second command voltage setting unit 42e calculates the second γ-axis command voltage value Vγ2 * based on the second γ-axis deviation ΔIγ2. Further, the second command voltage setting unit 42e calculates the second δ-axis command voltage value Vδ2 * based on the second δ-axis deviation ΔIδ2. As the feedback control used in the second command voltage setting unit 42e, for example, proportional integration control can be used.

The second voltage conversion unit 42f is based on the second γ-axis command voltage value Vγ2 *, the second δ-axis command voltage value Vδ2 *, the power supply voltage value VDC, and the second estimated angle θγ2, and the second γ, δ-axis command in the γδ coordinate system. The voltage values Vγ2 * and Vδ2 * are converted into the second U, V, W phase command voltage values VU2, VV2, VW2 in the three-phase fixed coordinate system. In the present embodiment, the second U, V, and W phase command voltage values VU2, VV2, and VW2 are sinusoidal signals having a median value of 0 and being 120 ° out of phase with the electrical angle.

The second generation unit 42g generates the second operation signal gUp2 to gWn2 based on the second U, V, W phase command voltage values VU2, VV2, VW2 output from the second voltage conversion unit 42f. The second generation unit 42g outputs the generated second operation signal gUp2 to gWn2 to each switch SUp2 to SWn2.

In this embodiment, the above-described configurations 41b to 41g and 42b to 42g correspond to the operation unit.

The control device 40 has a speed calculation unit 43, a γ-axis voltage unit 44, a δ-axis voltage unit 45, a γ-axis current unit 46, a δ-axis current unit 47, and an induced voltage calculation as a configuration for calculating the first estimated angle θγ1. A unit 48, a parameter setting unit 49, and a position estimator 50 are provided. Hereinafter, after explaining the principle of estimating the electric angle according to the present embodiment, each configuration 43 to 50 will be described.

The voltage equation of the motor 10 is expressed by the following equation (eq1).

In the above equation (eq1), Vd1 and Vq1 indicate the d and q-axis voltage in the first winding group 10A, Vd2 and Vq2 indicate the d and q-axis voltage in the second winding group 10B, and Id1 and Iq1 are the first. The d and q-axis current values flowing in the first winding group 10A are shown, and Id2 and Iq2 indicate the d and q-axis current values flowing in the second winding group 10B. R indicates the winding resistance values of the first and second winding groups 10A and 10B, and s indicates the differential operator in the Laplace transform. Ld1 and Lq1 indicate the d and q-axis inductance of the first winding group 10A, Ld2 and Lq2 indicate the d and q-axis inductance of the second winding group 10B, and Md and Mq are between the winding groups 10A and 10B. Indicates the mutual inductance of the d and q axes, Ke indicates the induced voltage constant, and ωe indicates the electric angular frequency of the motor 10.

From the above equation (eq1), the voltage equation of the γδ coordinate system corresponding to the first winding group 10A is derived as in the following equation (eq2), and the extended induced voltage of the γδ coordinate system corresponding to the first winding group 10A is derived. Is derived as in the following equation (eq3).

In the above equation (eq2), Vγ1 and Vδ1 indicate the γ and δ-axis voltage values in the first winding group 10A, and Iγ1 and Iδ1 indicate the γ and δ-axis current values flowing in the first winding group 10A. Iδ2 indicates the γ and δ axis current values flowing in the second winding group 10B. Further, p indicates a differential operator, eγ1 indicates a first γ-axis induced voltage value, eδ1 indicates a first δ-axis induced voltage value, and Δθ indicates a position estimation error of the γδ coordinate system with respect to the dq coordinate system.

Further, from the above equation (eq1), the voltage equation of the γδ coordinate system corresponding to the second winding group 10B is derived as in the following equation (eq4), and the extension of the γδ coordinate system corresponding to the second winding group 10B. The induced voltage is derived as shown in the following equation (eq5).

In the above equation (eq4), Vγ2 and Vδ2 indicate the γ and δ-axis voltage values in the second winding group 10B, eγ2 indicates the second γ-axis induced voltage value, and eδ2 indicates the first δ-axis induced voltage value.

Here, the γ-axis voltage value Vγm is expressed by the following equation (eq6), the δ-axis voltage value Vδm is expressed by the following equation (eq7), the γ-axis current value Iγm is expressed by the following equation (eq8), and the δ-axis current value Iδm. Is expressed by the following equation (eq9). Further, the γ-axis induced voltage value eγm is expressed by the following equation (eq10), and the δ-axis induced voltage value eδm is expressed by the following equation (eq11).

The following equation (eq12) is derived based on the above equations (eq2), (eq4), (eq6) to (eq9) and the conditions of “Ld1 = Ld2 = Ld” and “Lq1 = Lq2 = Lq”.

The following equation (eq13) is derived based on the above equations (eq3), (eq5), (eq6) to (eq9) and the conditions of “Ld1 = Ld2 = Ld” and “Lq1 = Lq2 = Lq”.

The γ and δ axis induced voltage values eγm and eδm shown in the above equation (eq13) include information on the position estimation error Δθ. Therefore, the electric angle of the motor 10 can be estimated based on the γ and δ axis induced voltage values eγm and eδm calculated based on the above equation (eq12).

Subsequently, each configuration 43 to 50 will be described.

The speed calculation unit 43 calculates the electric angular frequency ωe of the motor 10 based on the first estimated angle θγ1. The velocity calculation unit 43 may calculate the electric angular frequency ωe based on the second estimated angle θγ2.

The γ-axis voltage unit 44 includes the first γ-axis command voltage value Vγ1 * calculated by the first command voltage setting unit 41e and the second γ-axis command voltage value Vγ2 * calculated by the second command voltage setting unit 42e. The γ-axis voltage value Vγm is calculated based on (eq6).

The δ-axis voltage unit 45 includes the first δ-axis command voltage value Vδ1 * calculated by the first command voltage setting unit 41e and the second δ-axis command voltage value Vδ2 * calculated by the second command voltage setting unit 42e. The δ-axis voltage value Vδm is calculated based on (eq7).

The γ-axis current unit 46 is based on the first γ-axis current value Iγ1r calculated by the first current conversion unit 41a, the second γ-axis current value Iγ2r calculated by the second current conversion unit 42a, and the above equation (eq8). Then, the γ-axis current value Iγm is calculated.

The δ-axis current unit 47 is based on the first δ-axis current value Iδ1r calculated by the first current conversion unit 41a, the second δ-axis current value Iγ2r calculated by the second current conversion unit 42a, and the above equation (eq9). Then, the δ-axis current value Iδm is calculated.

The induced voltage calculation unit 48 calculates the γ-axis induced voltage value eγm based on the γ-axis voltage value Vγm, the γ-axis current value Iγm, the δ-axis current value Iδm, the electric angular frequency ωe, and the above equation (eq12). .. Further, the induced voltage calculation unit 48 calculates the δ-axis induced voltage value eδm based on the δ-axis voltage value Vδm, the γ-axis current value Iγm, the δ-axis current value Iδm, the electric angular frequency ωe, and the above equation (eq12). To do.

The parameter setting unit 49 calculates each motor parameter R, Ld, Lq, Md, Mq used in the induced voltage calculation unit 48 based on the first winding temperature TH1, the second winding temperature TH2, and the current amplitude Iamp. .. The current amplitude Imp is the magnitude of the current vector flowing through the winding groups 10A and 10B. The current amplitude Imp may be calculated based on, for example, the respective phase current values Iph1 and Iph2.

The position estimator 50 calculates the first estimated angle θγ1 based on the γ-axis induced voltage value eγm and the δ-axis induced voltage value eδm. The position estimator 50 may calculate the first estimated angle θγ1 based on, for example, a well-known extended induced voltage observer.

Subsequently, with reference to FIG. 6, the detection timing of the first and second phase current values Iph1 and Iph2 used in the first and second current conversion units 41a, 42a and the like will be described. FIG. 6 shows the transition of the actual first and second axis current values Iδ1 and Iδ2. In FIG. 6, it is assumed that the first and second δ-axis current values Iδ1 and Iδ2 include harmonic components such as 6th and 12th orders.

In the present embodiment, as described above, the spatial phase difference Δα is set to 30 °. In this configuration, when the rectangular wave drive control is performed, as shown in FIG. 6, the transition of the first δ-axis current value Iδ1 and the transition of the second δ-axis current value Iδ2 are the first and second-axis current values Iδ1. , Iδ2 is symmetric with respect to the average value Iδave of one fluctuation period. In the present embodiment, one fluctuation period of the first and second δ-axis current values Iδ1 and Iδ2 is 60 ° or approximately 60 °. Here, the first and second axis current detection values Iδ1d (θe) and Iδ2d (θe), which are the first and second axis current values Iδ1 and Iδ2 at the same timing, are expressed by the following equation (eq14).

In the above equation (eq14), Iδave1 indicates the average value of the first δ-axis current Iδ1 in one fluctuation cycle, Iδave2 indicates the average value of the second δ-axis current value Iδ2 in one fluctuation cycle, and β represents each δ-axis current value Iδ1. , The amount of deviation from the average values Iδave1 and Iδave2 of Iδ2 are shown. In the present embodiment, since the self-inductances of the winding groups 10A and 10B are the same and the common command torque Trq * is set for each of the inverters 20A and 20B, the average values Iδave1 and Iδave2 are set. The values Iδave are equal to each other. The sum of the first and second axis current detection values Iδ1d (θe) and Iδ2d (θe) gives the following equation (eq15).

According to the above equation (eq15), when the detection timing of the first δ-axis current value Iδ1 and the detection timing of the second δ-axis current value Iδ2 are the same timing, the first and second δ-axis current values Iδ1 and Iδ2 It can be seen that 1/2 of the total value is equal to the δ-axis current average value Iδave. Therefore, in the present embodiment, as shown in FIG. 6, the detection timing of the first δ-axis current value Iδ1 and the detection timing of the second δ-axis current value Iδ2 are set to the same timing Td. That is, the detection timing of the first phase current value Iph1 and the detection timing of the second phase current value Iph2 are set to the same timing. In particular, in the present embodiment, the detection timing is set every 60 ° of the electric angle.

Further, in the present embodiment, the detection timing Td is set to a period Tag other than the detection prohibition period Tnd, as illustrated in FIG. The detection prohibition period Tnd is from the first timing in which the first and second voltage vectors Vct1 and Vct2 of the first and second inverters 20A and 20B are switched to the second timing in which the voltage vector is switched immediately after the first timing. It is set to a period shorter than the period. In the detection prohibition period Tnd, noise caused by switching of the voltage vector may be included in the phase current value. When the first estimated angle θγ1 is calculated based on the phase current value including noise, the calculation accuracy may be lowered. Therefore, by avoiding the detection prohibition period Tnd and setting the detection timing Td, it is possible to avoid a decrease in the calculation accuracy of the first estimated angle θγ1.

Incidentally, the transition of the first γ-axis current value Iγ1 and the transition of the second γ-axis current value Iγ2 are also symmetrical with respect to the average value Iγave of one fluctuation period of the first and second γ-axis current values Iγ1 and Iγ2. In the present embodiment, this one fluctuation period is 60 ° or approximately 60 °. Therefore, the following equation (eq16) is established for the first and second γ-axis current detection values Iγ1d (θe) and Iγ2d (θe), which are the first and second γ-axis current values Iγ1 and Iγ2 at the same timing.

FIG. 7 shows a procedure of torque control processing according to this embodiment. This process is repeatedly executed by the control device 40 at predetermined process cycles.

In this series of processes, first, in step S10, it is determined whether or not the detection timings Td of the first phase current value Iph1 and the second phase current value Iph2 are set. In the present embodiment, as described above, the detection timing Td is set at intervals of 60 ° of electrical angles.

If it is determined in step S10 that the detection timing is Td, the process proceeds to step S11 to acquire the first phase current value Iph1 and the second phase current value Iph2 detected in the detection timing Td. This process corresponds to the current acquisition unit.

Then, based on the first estimated angle θγ1 calculated last time, the acquired first phase current value Iph1 is converted into the first γ, δ-axis current detection values Iγ1d (θe) and Iδ1d (θe). This conversion is performed by the first current conversion unit 41a. Further, the acquired second phase current value Iph2 is converted into the second γ and δ axis detection current values Iγ2d (θe) and Iδ2d (θe) based on the second estimated angle θγ2 calculated last time. This conversion is performed by the second current conversion unit 42a.

In the following step S12, the first and second γ-axis current values Iγ1r are based on the first and second γ-axis current detection values Iγ1d (θe) and Iγ2d (θe) converted in the current processing cycle and the following equation (eq17). , Iγ2r is calculated.

Further, in step S12, the first and second axis current values Iδ1r are based on the first and second axis current detection values Iδ1d (θe) and Iδ2d (θe) converted in the current processing cycle and the following equation (eq18). , Iδ2r is calculated.

The first γ, δ-axis current values Iγ1r, Iδ1r calculated in step S12 are output from the first current conversion unit 41a, and the second γ, δ-axis current values Iγ2r, Iδ2r calculated in step S12 are the second current conversion unit 42a. Is output from. In this embodiment, the conversion process in step S11 and the process in step S12 correspond to the current calculation unit. Further, the first γ, δ-axis current values Iγ1r, Iδ1r calculated in step S12 correspond to the first conversion current value, and the second γ, δ-axis current values Iγ2r, Iδ2r calculated in step S12 are the second conversion current values. Corresponds to.

In the following step S13, the γ-axis current value Iγm is calculated by the γ-axis current unit 46 based on the first and second γ-axis current values Iγ1r and Iγ2r calculated in step S12 and the above equation (eq8). Further, the δ-axis current value Iδm is calculated by the δ-axis current unit 47 based on the first and second δ-axis current values Iδ1r and Iδ2r calculated in step S12 and the above equation (eq9). Further, the γ-axis voltage value Vγm is calculated by the γ-axis voltage unit 44 based on the first and second γ-axis command voltage values Vγ1 * and Vγ2 * calculated last time and the above equation (eq6). Then, based on the calculated γ-axis voltage value Vγm, γ-axis current value Iγm, δ-axis current value Iδm, the electric angular frequency ωe calculated by the velocity calculation unit 43, and the above equation (eq12), the induced voltage calculation unit 48 The γ-axis induced voltage value eγm is calculated.

In step S13, the δ-axis voltage value Vδm is calculated by the δ-axis voltage unit 45 based on the first and second δ-axis command voltage values Vδ1 * and Vδ2 * calculated last time and the above equation (eq7). Then, based on the calculated δ-axis voltage value Vδm, γ-axis current value Iγm, δ-axis current value Iδm, electric angular frequency ωe, and the above equation (eq12), the induced voltage calculation unit 48 calculates the δ-axis induced voltage value eδm. calculate.

In the following step S14, the first estimated angle θγ1 is calculated by the position estimator 50 based on the calculated γ-axis induced voltage value eγm and the δ-axis induced voltage value eδm. Then, the angle addition unit 51 calculates the second estimated angle θγ2 based on the first estimated angle θγ1. The first estimated angle θγ1 and the second estimated angle θγ2 calculated in step S14 are used in the next conversion process in step S11. In this embodiment, the processes of steps S13 and S14 correspond to the position estimation unit.

In the following step S15, based on the first γ and δ axis current values Iγ1r and Iδ1r calculated in step S12, the first γ axis deviation calculation unit 41c, the first δ axis deviation calculation unit 41d and the first command voltage setting unit 41e 1γ, δ-axis command voltage values Vγ1 *, Vδ1 * are calculated. Further, in step S15, based on the second γ and δ axis current values Iγ2r and Iδ2r calculated in step S12, the second γ axis deviation calculation unit 42c, the second δ axis deviation calculation unit 42d and the second command voltage setting unit 42e perform the second operation. 2γ, δ-axis command voltage values Vγ2 *, Vδ2 * are calculated. Each command voltage value Vγ1 *, Vδ1 *, Vγ2 *, Vδ2 * calculated in step S15 is used in the next process of step S13.

As described above, in the present embodiment, the detection timings of the first phase current value Iph1 and the second phase current value Iph2 are set to the same timing every 60 °. Then, based on the detected first phase current value Iph1 and second phase current value Iph2, the first γ, δ axis current values Iγ1r, Iδ1r and the second γ, δ axis current values Iγ2r, Iδ2r are obtained by the processing of steps S11 and S12. It is calculated. The deviations of the respective current values Iγ1r, Iδ1r, Iγ2r, and Iδ2r calculated in this way from the corresponding average values are reduced, and thus the harmonic components are reduced. Therefore, the first and second estimated angles θγ1 and θγ2 are calculated based on the current values Iγ1r, Iδ1r, Iγ2r, and Iδ2r calculated by the processes of steps S11 and S12, so that the first and second estimated angles are calculated. It is possible to suppress a decrease in the calculation accuracy of θγ1 and θγ2.

<Second Embodiment>
Hereinafter, the second embodiment will be described with reference to the drawings, focusing on the differences from the first embodiment. In this embodiment, as shown in FIG. 8, shunt resistors for current detection are provided in the inverters 20A and 20B instead of the first and second phase current detection units 31A and 31B. In FIG. 8, the same members as those shown in FIG. 1 above are designated by the same reference numerals for convenience.

The first inverter 20A includes first U, V, W phase shunt resistors 50U, 50V, 50W. The 1st U phase shunt resistor 50U is connected between the emitter of the 1st U phase lower arm switch SUn1 and the negative electrode terminal of the battery 21, and detects the current value flowing through the 1st U phase lower arm switch Sun1 as the 1st U phase current value IU1. To do. The first V phase shunt resistor 50V is connected between the emitter of the first V phase lower arm switch SVn1 and the negative electrode terminal of the battery 21, and detects the current value flowing through the first V phase lower arm switch SVn1 as the first V phase current value IV1. To do. The 1W phase shunt resistor 50W is connected between the emitter of the 1W phase lower arm switch SWn1 and the negative electrode terminal of the battery 21, and detects the current value flowing through the 1W phase lower arm switch SWn1 as the 1W phase current value IW1. To do.

The second inverter 20B includes second U, V, W phase shunt resistors 60U, 60V, 60W. The shunt resistors 60U, 60V, 60W constituting the second inverter 20B are the same as the shunt resistors 50U, 50V, 50W constituting the first inverter 20A, and thus detailed description thereof will be omitted.

The current values detected by the shunt resistors 50U, 50V, 50W, 60U, 60V, 60W are input to the control device 40. Hereinafter, the current values IU1, IV1, and IW1 detected by the first U, V, and W phase shunt resistors 50U, 50V, and 50W are referred to as the first phase current values Iph1, and the second U, V, and W phase shunt resistors 60U and 60V. , Each current value IU2, IV2, IW2 detected by 60W is referred to as a second phase current value Iph2.

In the present embodiment, as shown in FIG. 9, the currents of the voltage vectors Vct1 and Vct2 of the inverters 20A and 20B are within the period of the active voltage vectors V1, V3 and V5 in which the lower arm switches for the two phases are turned on. The detection timing is set. This is because the current can be detected based on Kirchhoff's law only when the voltage vectors are the effective voltage vectors V1, V3, and V5 in the configuration in which the shunt resistor is provided and the square wave control is performed. In FIG. 9, the first detection timing, which is the detection timing of the first phase current value Iph1, is shown by Td1, and the second detection timing, which is the detection timing of the second phase current value Iph2, is shown by Td2. In the present embodiment, the first detection timing Td1 and the second detection timing Td2 are alternately set every 60 °.

FIG. 10 shows a procedure of torque control processing according to the present embodiment. This process is repeatedly executed by the control device 40 at predetermined process cycles. In FIG. 10, the same processes as those shown in FIG. 7 above are designated by the same reference numerals for convenience.

In this series of processes, first, in step S20, it is determined whether or not the first detection timing Td1 of the first phase current value Iph1 is set.

If it is determined in step S20 that the first detection timing is Td1, the process proceeds to step S21, and the first phase current value Iph1 detected in the first detection timing Td1 is acquired. Then, based on the first estimated angle θγ1 calculated last time, the acquired first phase current value Iph1 is converted into the first γ, δ-axis current detection values Iγ1d (θe) and Iδ1d (θe).

In the following step S22, the first γ-axis current detection value Iγ1d (θe) converted in the current processing cycle and the second γ-axis current detection value Iγ2d (θe-60) converted in the processing of step S24 in the previous second detection timing Td2. °) and the following equation (eq19), the first γ-axis current value Iγ1r is calculated.

Further, in step S22, the first δ-axis current detection value Iδ1d (θe) converted in the current processing cycle and the second δ-axis current detection value Iδ2d (θe-60) converted in the processing of step S24 in the previous second detection timing Td2. °) and the following equation (eq20), the first δ-axis current value Iδ1r is calculated.

The first γ and δ axis current values Iγ1r and Iδ1r calculated in step S22 are output from the first current conversion unit 41a.

In step S13 after the processing of step S22 is completed, the first γ-axis current value Iγ1r calculated in the current processing cycle in step S22, the second γ-axis current value Iγ2r calculated last time in step S24, and the above equation (eq8) are used. Based on this, the γ-axis current value Iγm is calculated. Further, the δ-axis current value Iδm is calculated based on the first δ-axis current value Iδ1r calculated in the current processing cycle in step S22, the second δ-axis current value Iδ2r calculated last time in step S25, and the above equation (eq9). .. Further, the γ-axis voltage value Vγm is calculated based on the first and second γ-axis command voltage values Vγ1 * and Vγ2 * calculated last time in step S15 and the above equation (eq6). Then, the γ-axis induced voltage value eγm is calculated based on the calculated γ-axis voltage value Vγm, γ-axis current value Iγm, δ-axis current value Iδm, electric angular frequency ωe, and the above equation (eq12).

Further, in step S13 after the processing of step S22 is completed, the δ-axis voltage is based on the first and second δ-axis command voltage values Vδ1 * and Vδ2 * calculated last time in step S15 and the above equation (eq7). Calculate the value Vδm. Then, the δ-axis induced voltage value eδm is calculated based on the calculated δ-axis voltage value Vδm, γ-axis current value Iγm, δ-axis current value Iδm, electric angular frequency ωe, and the above equation (eq12). Then, the first estimated angle θγ1 is calculated based on the calculated γ-axis induced voltage value eγm and the δ-axis induced voltage value eδm.

If a negative determination is made in step S20, the process proceeds to step S23 to determine whether or not the second detection timing Td2 of the second phase current value Iph2 is reached.

If it is determined in step S23 that the second detection timing is Td2, the process proceeds to step S24 to acquire the second phase current value Iph2 detected in the second detection timing Td2. Then, based on the second estimated angle θγ2 calculated last time, the acquired second phase current value Iph2 is converted into the second γ and δ-axis current detection values Iγ2d (θe) and Iδ2d (θe).

In the following step S25, the second γ-axis current detection value Iγ2d (θe) converted in the current processing cycle and the first γ-axis current detection value Iγ1d (θe-60) converted in the processing of step S21 in the previous first detection timing Td1. °) and the following equation (eq21), the second γ-axis current value Iγ2r is calculated.

Further, in step S25, the second δ-axis current detection value Iδ2d (θe) converted in the current processing cycle and the first δ-axis current detection value Iδ1d (θe-60) converted in the processing of step S21 in the previous first detection timing Td1. °) and the following equation (eq22), the second δ-axis current value Iδ2r is calculated.

The second γ and δ axis current values Iγ2r and Iδ2r calculated in step S25 are output from the second current conversion unit 42a.

In step S13 after the processing of step S25 is completed, the second γ-axis current value Iγ2r calculated in the current processing cycle in step S25, the first γ-axis current value Iγ1r calculated last time in step S22, and the above equation (eq8) are used. Based on this, the γ-axis current value Iγm is calculated. Further, the δ-axis current value Iδm is calculated based on the second δ-axis current value Iδ2r calculated in the current processing cycle in step S25, the first δ-axis current value Iδ1r calculated last time in step S22, and the above equation (eq9). .. Further, the γ-axis voltage value Vγm is calculated based on the first and second γ-axis command voltage values Vγ1 * and Vγ2 * calculated last time in step S15 and the above equation (eq6). Then, the γ-axis induced voltage value eγm is calculated based on the calculated γ-axis voltage value Vγm, γ-axis current value Iγm, δ-axis current value Iδm, electric angular frequency ωe, and the above equation (eq12).

Further, in step S13 after the processing of step S25 is completed, the δ-axis voltage is based on the first and second δ-axis command voltage values Vδ1 * and Vδ2 * calculated last time in step S15 and the above equation (eq7). Calculate the value Vδm. Then, the δ-axis induced voltage value eδm is calculated based on the calculated δ-axis voltage value Vδm, γ-axis current value Iγm, δ-axis current value Iδm, electric angular frequency ωe, and the above equation (eq12). Then, the first estimated angle θγ1 is calculated based on the calculated γ-axis induced voltage value eγm and the δ-axis induced voltage value eδm.

According to the present embodiment described above, even when the voltage vector capable of detecting the current is limited, it is possible to suppress a decrease in the calculation accuracy of the first and second estimated angles θγ1 and θγ2.

<Third Embodiment>
Hereinafter, the third embodiment will be described with reference to the drawings, focusing on the differences from the second embodiment. In the second embodiment, the current detection timing is fixed every 60 °. On the other hand, in the present embodiment, the timing at which the first and second γ-axis current values Iγ1 and Iγ2 are assumed to be the γ-axis current average value Iγave and the first and second δ-axis current values Iδ1 and Iδ2 are the δ-axis current average values. The timing expected to be Iδave is calculated, and the calculated timing is set as the current detection timing. This setting is for dealing with the fact that the amplitude and phase of each current value Iγ1, Iγ2, Iδ1, Iδ2 change according to the power supply voltage value VDC, the electric angular frequency ωe, and the like. That is, when the current detection timing is set to a predetermined timing regardless of the magnitude of the power supply voltage value VDC and the electric angular frequency ωe, the respective γ-axis current values Iγ1r, Iγ2r and the γ-axis current average value Iγave The deviation between the δ-axis current values Iδ1r and Iδ2r and the δ-axis current average value Iδave can be large. When the deviation becomes large, the harmonic components contained in the respective current values Iγ1, Iγ2, Iδ1, Iδ2 become large, and the calculation accuracy of the first and second estimated angles θγ1, θγ2 may decrease.

First, it will be described that the amplitude and phase of the current value depend on the power supply voltage value VDC and the electric angular frequency ωe. In this embodiment, the d-axis corresponding to the γ-axis will be described as an example.

In the present embodiment, for the sake of simplicity, the condition "ωe = 0" is imposed in the above equation (eq1). Then, when the above equation (eq1) is solved for each current value, the following equation (eq23) is derived. The matrix Q on the right side of the following equation (eq23) is represented by the following equation (eq24).

In the present embodiment, for convenience of explanation, only the first d-axis current value Id1 is considered among the current values Id1, Iq1, Id2, and Iq2 of the above equation (eq23). The lower equation (eq25) is derived from the upper equation (eq23).

The first d-axis voltage value Vd1 and the second d-axis voltage value Vd2 are expressed by the following equation (eq26). In the following equation (eq26), Va indicates the magnitude of the combined vector of the two voltage vectors Vct1 and Vct2, and is a value determined based on the power supply voltage value VDC. Further, θ1 indicates the first phase which is the phase of the first d-axis voltage value Vd1, and θ2 indicates the second phase which is the phase of the second d-axis voltage value Vd2.

In this embodiment, as shown in FIG. 11, the voltage vector is switched every 30 ° of the electric angle. Therefore, the first phase θ1 is expressed by the following equation (eq27), and the second phase θ2 is expressed by the following equation (eq28).

Δ in the above equation (eq27) indicates the phase of the voltage vector Vct1 of the first inverter 20A in the γδ coordinate system, and is a value determined from the first γ and δ axis command voltage values Vγ1 * and Vδ1 *. Based on the above equations (eq25) to (eq28) and the Laplace inverse transformation, the following equations (eq29) to (eq31) are derived from the above equation (eq25) as stationary solutions.

Subsequently, a transient solution corresponding to the initial value response of the first d-axis current value Id1 will be examined based on the following equation (eq32).

The Laplace transform of the above equation (eq32) leads to the following equation (eq33).

The second and third terms on the right side of the above equation (eq33) correspond to the transient solution. Laplace inverse conversion of the second term on the right side of the above equation (eq33) leads to the lower equation (eq34), and inverse transformation of the right side third term of the above equation (eq33) leads to the lower equation (eq35). In addition, Id1 (0) of the following formula (eq34) and Id2 (0) of the following formula (eq35) may be calculated by numerical calculation, for example.

From the above equations (eq29) to (eq31), (eq34), and (eq35), the first d-axis current value Id1 is expressed as the following equation (eq36).

Since the Va included in the right side of the above equations (eq29) to (eq31) depends on the power supply voltage value VDC and the electric angular frequency ωe is included in the right side of the above equations (eq29) to (eq31), the first d The amplitude and phase of the shaft current value Id1 depend on the power supply voltage value VDC and the electric angular frequency ωe. Similarly, the first q-axis current value Iq1 corresponding to the first δ-axis current value Iδ1, the second d-axis current value Id2 corresponding to the second γ-axis current value Iγ2, and the second q-axis current value corresponding to the second δ-axis current value Iδ2. In each of Iq2, the amplitude and phase depend on the power supply voltage value VDC and the electric angular frequency ωe.

Since each motor parameter is included on the right side of the above equations (eq29) to (eq31), (eq34), and (eq35), the amplitude and phase of each current value Id1, Iq1, Id2, Iq2 are those of the winding group. It also depends on the temperature.

FIG. 12 shows a procedure of torque control processing according to the present embodiment. This process is repeatedly executed by the control device 40 at predetermined process cycles. Note that, in FIG. 12, the same processing as that shown in FIG. 10 above is designated by the same reference numerals for convenience.

In this series of processes, first, in step S20, it is determined whether or not the current timing is the first detection timing Td1 calculated in step S41. If an affirmative determination is made in step S20, the process proceeds to step S30 via steps S21 and S22. In steps S30 and S31, the same processing as in steps S13 and S14 after step S22 in FIG. 10 is performed.

In the following step S32, based on the first γ and δ axis current values Iγ1r and Iδ1r calculated in step S22, the first γ axis deviation calculation unit 41c, the first δ axis deviation calculation unit 41d and the first command voltage setting unit 41e 1γ, δ-axis command voltage values Vγ1 *, Vδ1 * are calculated.

In the following step S33, the velocity calculation unit 43 calculates the electric angular frequency ωe. In the present embodiment, the processing of step S33 and step S39 described later corresponds to the speed calculation unit.

In the following step S34, the power supply voltage value VDC, the first winding temperature TH1 and the second winding temperature TH2 are acquired. In this embodiment, the process of step S34 corresponds to the voltage acquisition unit.

In the following step S35, the second detection timing Td2 of the next second phase current value Iph2 is calculated. In the present embodiment, the timing at which the actual second γ-axis current value Iγ2 is expected to become the next γ-axis current average value Iγave is calculated as the second detection timing Td2. The next second detection timing Td2 can be calculated by using the following equation (eq37) based on the above equation (eq36), and the second γ-axis current value Iγ2 corresponding to the second d-axis current value Id2 is the γ-axis current average value. It is based on the fact that the electric angle θe that becomes Iγave can be calculated, and that the time interval between the calculated electric angle θe and the current electric angle can be grasped based on the electric angular frequency ωe.

In the present embodiment, since the second d-axis current value Id2 depends on the electric angle frequency ωe, the power supply voltage value VDC, the first winding temperature TH1 and the second winding temperature TH2, the calculated electric angle frequency ωe is acquired. The next second detection timing Td2 is calculated based on the power supply voltage value VDC, the first winding temperature TH1 and the second winding temperature TH2. In step S35, specifically, for example, based on the map information in which the electric angular frequency ωe, the power supply voltage value VDC, the first winding temperature TH1 and the second winding temperature TH2, and the second detection timing Td2 are related. Then, the second detection timing Td2 may be calculated.

Incidentally, in the present embodiment, the processing of step S35 and step S41 described later corresponds to the timing calculation unit.

If a negative determination is made in step S20, the process proceeds to step S23, and it is determined whether or not the current timing is the first detection timing Td1 calculated in step S35. If an affirmative determination is made in step S23, the process proceeds to step S36 via steps S24 and S25. In steps S36 and S37, the same processing as in steps S13 and S14 after step S25 in FIG. 10 is performed.

In the following step S38, the second γ-axis deviation calculation unit 42c, the second δ-axis deviation calculation unit 42d, and the second command voltage setting unit 42e are used based on the second γ and δ-axis current values Iγ2r and Iδ2r calculated in step S25. 2γ, δ-axis command voltage values Vγ2 *, Vδ2 * are calculated. Then, in step S39, the velocity calculation unit 43 calculates the electric angular frequency ωe. Then, in step S40, the power supply voltage value VDC, the first winding temperature TH1 and the second winding temperature TH2 are acquired.

In the following step S41, the first detection timing Td1 of the next first phase current value Iph1 is calculated. In the present embodiment, the timing at which the actual first γ-axis current value Iγ1 is expected to become the next γ-axis current average value Iγave is calculated as the first detection timing Td1. In the present embodiment, the next first detection timing Td1 is calculated based on the calculated electric angular frequency ωe, the acquired power supply voltage value VDC, the first winding temperature TH1, and the second winding temperature TH2. In step S41, specifically, for example, based on the map information in which the electric angular frequency ωe, the power supply voltage value VDC, the first winding temperature TH1 and the second winding temperature TH2, and the first detection timing Td1 are related. Then, the first detection timing Td1 may be calculated.

As described above, in the present embodiment, the next first and second detection timings Td1 and Td2 are calculated based on the power supply voltage value VDC, the electric angular frequency ωe, and the like. Then, the first and second phase current values Iph1 and Iph2 detected at the calculated first and second detection timings Td1 and Td2 were acquired. Therefore, the harmonic components contained in the respective current values Iγ1r, Iδ1r, Iγ2r, and Iδ2r calculated based on the first and second phase current values Iph1 and Iph2 can be reduced, and the first and second estimated angles θγ1, θγ2 can be reduced. It is possible to suppress a decrease in the calculation accuracy of.

<Modified example of the third embodiment>
In step S35, the timing at which the actual second δ-axis current value Iδ2 is expected to become the next δ-axis current average value Iδave may be calculated as the second detection timing Td2. Further, in step S41, the timing at which the actual first δ-axis current value Iδ1 is expected to become the next δ-axis current average value Iδave may be calculated as the first detection timing Td1.

-In steps S35 and S41, it is not necessary to use the first winding temperature TH1 and the second winding temperature TH2 for calculating the detection timings Td1 and Td2.

<Fourth Embodiment>
Hereinafter, the fourth embodiment will be described with reference to the drawings, focusing on the differences from the second embodiment. In the present embodiment, the deviation between the first γ-axis current value Iγ1 and the γ-axis current average value Iγave at the first detection timing Td1 and the deviation between the first δ-axis current value Iδ1 and the δ-axis current average value Iδave at the first detection timing Td1. , Correct the deviation between the second γ-axis current value Iγ2 and the γ-axis current average value Iγave at the second detection timing Td2, and the deviation between the second δ-axis current value Iδ2 and the δ-axis current average value Iδave at the second detection timing Td2. ..

FIG. 13 shows a procedure of torque control processing according to the present embodiment. This process is repeatedly executed by the control device 40 at predetermined process cycles. Note that, in FIG. 13, the same processes as those shown in FIG. 12 above are designated by the same reference numerals for convenience.

In this series of processes, after the process of step S22 is completed, the process proceeds to step S50 via steps S33 and S34. In step S50, the first correction process shown in FIG. 14 is performed.

In step S51, the γ-axis current average values Iγave and δ-axis are based on the electric angular frequency ωe calculated in step S33, the power supply voltage value VDC acquired in step S34, and the first and second winding temperatures TH1 and TH2. The current average value Iδave is calculated. The average values Iγave and Iδave can be calculated because the transition of one fluctuation cycle of each current value can be calculated as shown in the above equation (eq37). In this embodiment, the process of step S51 and the process of step S61 described later correspond to the average value calculation unit.

In the following step S52, the γ-axis correction amount Δγ is calculated by subtracting the γ-axis current average value Iγave calculated in step S51 from the first γ-axis current value Iγ1r calculated in the current processing cycle in step S22. Further, in step S52, the δ-axis correction amount Δδ is calculated by subtracting the δ-axis current average value Iδave calculated in step S51 from the first δ-axis current value Iδ1r calculated in the current processing cycle in step S22. In this embodiment, the process of step S52 and the process of step S62 described later correspond to the correction amount calculation unit.

In the following step S53, the first γ-axis current value Iγ1r calculated in the current processing cycle in step S22 is reduced and corrected by the γ-axis correction amount Δγ calculated in step S52. Further, in step S53, the first δ-axis current value Iδ1r calculated in the current processing cycle in step S22 is reduced and corrected by the δ-axis correction amount Δδ calculated in step S52. The first γ and δ axis current values Iγ1r and Iδ1r corrected in step S53 are output from the first current conversion unit 41a. In this embodiment, the process of step S53 and the process of step S63 described later correspond to the correction unit.

Returning to the above description of FIG. 13, after the processing of step S50 is completed, the process proceeds to step S13.

On the other hand, after the processing of step S25 is completed, the process proceeds to step S60 via steps S39 and S40. In step S60, the second correction process shown in FIG. 15 is performed.

In step S61, the γ-axis current average values Iγave and δ-axis are based on the electric angular frequency ωe calculated in step S39, the power supply voltage value VDC acquired in step S40, and the first and second winding temperatures TH1 and TH2. The current average value Iδave is calculated.

In the following step S62, the γ-axis correction amount Δγ is calculated by subtracting the γ-axis current average value Iγave calculated in step S61 from the second γ-axis current value Iγ2r calculated in the current processing cycle in step S25. Further, in step S62, the δ-axis correction amount Δδ is calculated by subtracting the δ-axis current average value Iδave calculated in step S61 from the second δ-axis current value Iδ2r calculated in the current processing cycle in step S25.

In the following step S63, the second γ-axis current value Iγ2r calculated in the current processing cycle in step S25 is reduced and corrected by the γ-axis correction amount Δγ calculated in step S62. Further, in step S63, the second δ-axis current value Iδ2r calculated in the current processing cycle in step S25 is reduced and corrected by the δ-axis correction amount Δδ calculated in step S62. The second γ and δ axis current values Iγ2r and Iδ2r corrected in step S63 are output from the second current conversion unit 42a.

According to the present embodiment described above, the same effect as that of the third embodiment can be obtained.

<Fifth Embodiment>
Hereinafter, the fifth embodiment will be described with reference to the drawings, focusing on the differences from the third and fourth embodiments. In the present embodiment, after the detection timing calculation process described in the third embodiment is performed, the correction process described in the fourth embodiment is performed as necessary.

FIG. 16 shows a procedure of torque control processing according to the present embodiment. This process is repeatedly executed by the control device 40 at predetermined process cycles. Note that, in FIG. 16, the same processing as that shown in FIG. 13 is designated by the same reference numerals for convenience. In the process shown in FIG. 16, the initial values of the first and second flags F1 and F2, which will be described later, are set to 0.

In this series of processes, after the process of step S34 is completed, the process proceeds to step S70, and it is determined whether or not the first flag F1 is 1. The first flag F1 is set to 1 by the process of step S83 described later, and is set to 0 by the process of step S82.

If it is determined in step S70 that the first flag F1 is 1, the process proceeds to step S50 and the first correction process is performed.

When the process of step S50 is completed, or when it is determined in step S70 that the first flag F1 is 0, the process proceeds to step S71 via steps S30 to S32 and S35. In step S71, it is determined whether or not the second detection timing Td2 calculated in step S35 is included in the detection prohibition period Tnd illustrated in FIG. In the present embodiment, the processing of step S71 and step S81 described later corresponds to the determination unit. That is, even if the timing at which the second γ-axis current value Iγ2 becomes the γ-axis current average value Iγave is set to the second detection timing Td2, the second detection timing Td2 may be included in the detection prohibition period Tnd. In this case, if the second phase current value Iph2 detected according to the second detection timing Td2 calculated in step S35 is used for the calculation of the first estimated angle θγ1, the calculation accuracy of the first estimated angle θγ1 may decrease.

Therefore, if an affirmative determination is made in step S71, the process proceeds to step S73, and the second detection timing Td2 calculated in step S35 is shifted to a period Tag other than the detection prohibition period Tnd. Specifically, the second detection timing Td2 calculated in step S35 is shifted to the earliest timing among the period tags other than the detection prohibition period Tnd. Then, the second flag F1 is set to 1. In this embodiment, the process of step S73 and the process of step S83 described later correspond to the shift unit.

If a negative determination is made in step S71, the process proceeds to step S72 and the second flag F2 is set to 0.

On the other hand, after the processing of step S40 is completed, the process proceeds to step S80, and it is determined whether or not the second flag F2 is 1. If it is determined in step S80 that the second flag F2 is 1, the process proceeds to step S60 and the second correction process is performed.

When the process of step S60 is completed, or when it is determined in step S80 that the second flag F2 is 0, the process proceeds to step S81 via steps S36 to S38 and S41. In step S81, it is determined whether or not the first detection timing Td1 calculated in step S41 is included in the detection prohibition period Tnd.

If an affirmative determination is made in step S81, the process proceeds to step S83, and the first detection timing Td1 calculated in step S41 is shifted to a period Tag other than the detection prohibition period Tnd. Specifically, the first detection timing Td1 calculated in step S41 is shifted to the earliest timing among the period tags other than the detection prohibition period Tnd. Then, the first flag F1 is set to 1. On the other hand, if a negative determination is made in step S81, the process proceeds to step S82 and the first flag F1 is set to 0.

According to the present embodiment described above, it is possible to accurately suppress a decrease in calculation accuracy of the first and second estimated angles θγ1 and θγ2.

<Other Embodiments>
In addition, each of the above-described embodiments may be modified as follows.

-In the second embodiment, the current detection timing is set every 60 °, but the present invention is not limited to this. For example, as shown in FIG. 18, a pair of timings separated by 30 ° may be set as the current detection timing in the period in which the voltage vectors are the effective voltage vectors V1, V3, and V5. In FIG. 18, the first detection timing Td1 is shown at time t2, t3, t6, t7, t10, t11, and the second detection timing Td2 is shown at time t1, t2, t5, t6, t9, t10. In this case, the first phase current value Iph1 detected at the first detection timing Td1 is the first of the set of the first γ and δ axis current values Iγr1 and Iδr1 and the set of the second γ and δ axis current values Iγr2 and Iδr2. It may be used for calculation of only a set of 1γ and δ-axis current values Iγr1 and Iδr1. The second phase current value Iph2 detected at the second detection timing Td2 is the second γ, among the set of the first γ, δ axis current values Iγr1, Iδr1 and the second γ, δ axis current values Iγr2, Iδr2. It may be used for the calculation of only the set of the δ-axis current values Iγr2 and Iδr2.

Further, in the current detection method shown in FIG. 18, at times t2, t6, and t10 where the first detection timing Td1 and the second detection timing Td2 overlap, the average values of the first phase current value Iph1 and the second phase current value Iph2 are It may be used for calculating each current value Iγr1, Iδr1, Iγr2, Iδr2.

The timings indicated by the times t4, t8, and t12 are not the current detection timings because the voltage vectors do not become the effective voltage vectors V1, V3, and V5. In this case, at time t4, the first γ, δ-axis current values Iγr1, Iδr1 calculated at time t3 are used, and at time t8, the first γ, δ-axis current values Iγr1, Iδr1 calculated at time t7 are used. At t12, the first γ and δ axis current values Iγr1 and Iδr1 calculated at time t11 may be used.

Further, in the current detection method shown in FIG. 18, the correction process described in the fourth embodiment may be used.

The spatial phase difference Δα is not limited to 30 °, and may be any value satisfying “30 + 60 × N” ° (N is an integer) such as −30 °, 90 °, and 150 °.

-The current detection timing is not limited to that illustrated in each of the above embodiments. For example, in the second embodiment, the current detection timing may be set for each electric angle “60 × M” ° (M is an integer of 2 to 5). Further, the current detection timing is not limited to the one set every "60 × M" °, and may be set every other interval. Further, the current detection timing may be set at unequal intervals.

-For example, in the first and second embodiments, the current detection timing may be included in the detection prohibition period Tnd.

-In each of the above embodiments, the average value of the first γ-axis current value Iγ1 in one fluctuation cycle and the average value of the second γ-axis current value Iγ2 in one fluctuation cycle may be different. Further, the average value of the first δ-axis current value Iδ1 in one fluctuation cycle and the average value of the second δ-axis current value Iδ2 in one fluctuation cycle may be different.

-The motor is not limited to a motor having double windings, and may be, for example, a motor having only one winding group. Even in this case, for example, as shown at time t2, t3, t6, t7, t10, t11 in FIG. 18, the timing at which the first γ-axis current value Iγ1 is expected to become the γ-axis current average value Iγave is detected. The electric angle can be estimated by setting the timing. Further, even in this case, the configurations described in the third to fifth embodiments can be used.

Further, the motor may have an L heavy winding (L is an integer of 3 or more). Here, when the first to third winding groups are provided, for example, among these winding groups, the current detection timings of the first winding group and the second winding group are set to be the same, and the second winding is set. The current detection timings of the group and the third winding group may be set to be the same.

-The motor is not limited to the permanent magnet field type, and for example, a winding field type motor may be used.

-The control amount of the motor is not limited to torque, but may be, for example, rotational speed.

-The rotation position information to be estimated is not limited to the electric angle, but may be, for example, the mechanical angle of the motor.

10 ... Motor, 10A, 10B ... First and second winding groups, 20A, 20B ... First and second inverters, 40 ... Control device.

Claims (5)

A rotary electric machine (10) having a winding group (10A, 10B) wound around the stator (13), and
In a rotary electric machine control device (40) including a power conversion circuit (20A, 20B) for applying a voltage to the winding group and a position estimation unit for estimating the rotation position information of the rotary electric machine. ,
A current acquisition unit that acquires the current values (Iph1, Iph2) of the fixed coordinate system flowing in the winding group, and
A current for calculating a converted current value (Iγ1r, Iδ1r, Iγ2r, Iδ2r) which is a current value of a rotating coordinate system flowing in the winding group based on the current value acquired by the current acquisition unit and the rotation position information. Calculation part and
Based on the rotation position information, an operation unit that operates the power conversion circuit to drive the rotary electric machine in a rectangular wave , and an operation unit .
A voltage acquisition unit that acquires the input voltage value (VDC) of the power conversion circuit, and
A speed calculation unit that calculates the electric angular frequency (ωe) of the rotary electric machine based on the rotation position information,
The timing at which the actual conversion current value is assumed to be the average value in one fluctuation cycle based on the input voltage value acquired by the voltage acquisition unit and the electric angular frequency calculated by the speed calculation unit. With a timing calculation unit that calculates as the detection timing ,
The position estimation unit estimates the rotation position information based on the conversion current value calculated by the current calculation unit.
The current acquisition unit acquires the current value of the fixed coordinate system at the detection timing calculated by the timing calculation unit.
The current calculation unit is a control device for a rotating electric machine that converts a current value acquired by the current acquisition unit into the conversion current value used by the position estimation unit based on the rotation position information . The detection prohibition period (Tnd) is a period shorter than the period from the first timing at which the voltage vector of the power conversion circuit is switched to the second timing at which the voltage vector is switched immediately after the first timing.
The control device for a rotary electric machine according to claim 1 , wherein the current acquisition unit acquires a current value of the fixed coordinate system in a period other than the period from the first timing to the elapse of the detection prohibition period. The winding group is plural,
Of the plurality of winding groups, any one is designated as the first target winding group (10A), and the remaining one is designated as the second target winding group (10B).
The current acquisition unit has a first current value (Iph1) which is a current value of the fixed coordinate system flowing in the first target winding group and a current value of the fixed coordinate system flowing in the second target winding group. Obtain 2 current values (Iph2) and
Based on the first current value and the rotation position information, the current calculation unit uses the first conversion current value (1st conversion current value), which is the current value of the rotation coordinate system flowing in the first target winding group, as the conversion current value. Iγ1r, Iδ1r) are calculated, and based on the second current value and the rotation position information, the second conversion current, which is the current value of the rotation coordinate system flowing in the second target winding group, is used as the conversion current value. Calculate the values (Iγ2r, Iδ2r) and
The position estimation unit estimates the rotation position information based on the first conversion current value and the second conversion current value.
In the current acquisition unit, the total value of the first conversion current value and the second conversion current value calculated by the current calculation unit is 1 of each of the actual first conversion current value and the second conversion current value. The control device for a rotary electric machine according to claim 1 or 2 , wherein the acquisition timings of the first current value and the second current value are set so as to be the total value of the average values in the fluctuation cycle. The winding group is two.
The rotation according to claim 3 , wherein when N is an integer, the angle formed by the first target winding group and the second target winding group is "30 + 60 × N" ° in terms of electrical angle. Electric control device. The control device for a rotary electric machine according to claim 4 , wherein the current acquisition unit sets the acquisition timing of the first current value and the acquisition timing of the second current value by shifting the electric angle by 60 °.