CONTROLLER FOR A ROTATING HELICOPTER DRIVE UNIT

Abstract

Controller (400) for a rotary electric motor drive apparatus (1000), with a minimum loss calculation unit (802) which, in the case (ST110) where the required minimum voltage (VHL) is greater than or equal to the power source voltage (Vb) and less than or equal to twice the power source voltage (Vb) (ST110), calculates a total loss coefficient (ST120) whose maximum order is the second order, in the case where the total second-order loss coefficient is positive, calculates an extreme voltage (Vpl) at which the total power loss of the converter (15) and the inverter (IN) becomes a minimum, and then, in the case (ST140) where the extreme voltage (Vpl) is greater than twice the power source voltage (Vb), sets twice the power source voltage (Vb) as the low-loss voltage (VHLL) (ST171), or in the case,where the extreme voltage (Vpl) is less than the required minimum voltage (VHL), sets the required minimum voltage (VHL) as the low-loss voltage (VHLL) (ST172), or in the case where the extreme voltage (Vpl) is within a range between the required minimum voltage (VHL) and twice the values ​​of the power source voltage (Vb), sets the extreme voltage (Vpl) as the low-loss voltage (VHLL) (ST173); in the case (ST130) that the total second-order loss coefficient (A2) is not a positive value, sets the low-loss voltage (VHLL) to twice the value of the power source voltage (Vb) (ST174), if (ST160) the extreme voltage (Vp1) from which the total power loss becomes a maximum is less than the average voltage (Vave) of the required minimum voltage (VHL) and twice the value of the power source voltage (Vb), and sets the required minimum voltage (VHL) to the low-loss voltage (VHLL) (ST175),if (ST160) the extreme voltage (Vp1) is not less than the average voltage (Vave) of the required minimum voltage (VHL) and twice the value of the power source voltage (Vb); in the case (ST110, ST165) where the required minimum voltage (VHL) is greater than twice the value of the power source voltage (Vb), the required minimum voltage (VHL) is set as the low-loss voltage (VHLL) (ST176); and in the case (ST165) where the required minimum voltage (VHL) is greater than the output maximum voltage (Vmax) of the converter (15), the output maximum voltage (Vmax) of the converter (15) is set as the low-loss voltage (VHLL).

Description

BACKGROUND

[0001] The present invention relates to a controller for a rotary electric motor drive apparatus, which is provided with a converter that can increase the energy source voltage of a DC power source for output to a system voltage line, and an inverter that is provided between the converter and a rotary electric motor and performs a power conversion between a DC power of the system voltage line and an AC power that drives the rotary electric motor.

[0002] Regarding the controller for the rotary electric motor drive apparatus described above, the technology is already known, as described in JP 2007-325351 A. JP 2007-325351 A discloses the rotary electric motor drive apparatus that shares an output voltage of the converter with a plurality of inverters and controls a plurality of rotary electric motors. The technology of JP 2007-325351 A is designed, for each of a plurality of candidates for a voltage command value of the converter, to calculate the power loss of the DC power source, the power loss of the converter, the power losses of a plurality of inverters, and the total power loss of these, and to search for a voltage from a plurality of candidate voltages whose total power loss is minimized, and to set the sought voltage as a voltage command value of the converter.And in the technology of JP 2007 - 325 351 A it is designed to store characteristic map data of each power loss, whose arguments are a DC voltage, a speed and a torque of the rotary electric machine and the like, and to calculate each power loss using the characteristic map data.

[0003] From EP 2 023 482 A1, a controller for a rotary electric motor drive unit is known, wherein the total power loss is calculated as the sum of the power losses of individual components for the DC link voltage. From EP 2 728 739 A1, it is known to determine inverter loss coefficients as a function of speed, torque, or battery current. From US 2009 / 0237019 A1, it is known that if a motor's target operating point lies within a resonance range specified by the motor's operating point, the boost converter increases the voltage on the inverter side to a preset target voltage. From DE 10 2007 005 138 A1, it is known to set the target value for the boosted DC link voltage to the value of the required minimum voltage above a certain converter output power.From DE 10 2015 102 163 A1 it is known to control the upward conversion of the battery voltage on the basis of a driving mode or a driving condition. SUMMARY

[0004] However, the technology of JP 2007-325351 A requires calculating each power loss from a large number of voltage candidates using the characteristic map data for each power loss, resulting in a large data processing load. To improve search accuracy, it is necessary to increase the number of voltage candidates; however, as the number of voltage candidates increases, so does the data processing load.

[0005] In the technology described in JP 2007-325351 A, it is necessary to provide the characteristic map data for each power loss, which led to a problem: the amount of storage required by the memory apparatus becomes large. Specifically, depending on the converter's circuit design, the converter's loss characteristic cannot simply be a monotonic increase with the increase in the converter's output power. Since it is necessary to finely define the characteristic map data for each DC voltage, converter output voltage, and converter output power, and to calculate the power loss at a multitude of operating points, such a problem arose: the amount of storage and data processing load increased significantly, a high-performance computing unit was required, and the controller cost increased.

[0006] Therefore, it is desirable to provide a controller for a rotary electric motor drive apparatus that is capable of reducing the data processing load and the amount of memory required for data processing that calculates a voltage command value of the converter, thereby reducing power loss of the rotary electric motor drive apparatus.

[0007] This problem is solved according to the invention as specified in the main claim. Exemplary embodiments of the invention are set forth in the dependent claims.

[0008] According to the controller for the rotary electric motor drive apparatus of the present invention, the data processing load can be reduced without calculating each power loss using characteristic map data over a multitude of voltage candidates and without searching for the voltage command value for which the total power loss is minimized. This is because it is possible to set the voltage command value based on the coefficient of the polynomial that represents each power loss characteristic. Since each power loss characteristic is approximated by the polynomial, the amount of memory required by a storage device can be significantly reduced compared to directly converting the power loss characteristic into characteristic map data. Therefore, while the data processing load is reduced, the total power loss of the rotary electric motor drive apparatus can also be reduced. BRIEF DESCRIPTION OF THE DRAWINGS Fig. Figure 1 is a schematic diagram of a rotary electric motor drive apparatus and a controller according to embodiment 1 of the present invention. Fig. Figure 2 is a schematic block diagram of a controller according to embodiment 1 of the present invention. Fig. Figure 3 is a block diagram of an inverter control unit according to embodiment 1 of the present invention. Fig. Figure 4 is a block diagram of a voltage command calculation unit according to embodiment 1 of the present invention. Fig. Figure 5 is a torque-speed characteristic figure of the electric lathe according to embodiment 1 of the present invention. Fig. Figure 6 is a power loss characteristic figure of the inverter and the rotary electric machine according to embodiment 1 of the present invention. Fig. Figure 7 is a block diagram of a required minimum voltage calculation unit according to embodiment 1 of the present invention. Fig. Figure 8 is a power loss characteristic figure of the converter according to embodiment 1 of the present invention. Fig. Figure 9 is a flowchart representing the processing of a loss minimum command computation unit according to embodiment 1 of the present invention. Fig. Figure 10 is a figure to illustrate a polynomial representing a power loss characteristic of the converter and a calculation of a coefficient according to embodiment 1 of the present invention. Fig. Figure 11 is a figure to illustrate a polynomial representing a power loss characteristic of the first inverter and a calculation of a coefficient according to embodiment 1 of the present invention. Fig. Figure 12 is a figure to illustrate a polynomial representing a power loss characteristic of the second inverter and a calculation of a coefficient according to embodiment 1 of the present invention. Fig. 13A and Fig. Figures 13B illustrate a change in the total power loss characteristic by the positive / negative of the second-order total loss coefficient according to embodiment 1 of the present invention. Fig. Figure 14 is a figure to illustrate a setting of the voltage command value according to a position of the extreme voltage according to embodiment 1 of the present invention. Fig. Figure 15 is a figure to illustrate a setting of the voltage command value according to a position of the extreme voltage according to embodiment 1 of the present invention. Fig. Figure 16 is a figure to illustrate a setting of the voltage command value according to a position of the extreme voltage according to embodiment 1 of the present invention. Fig. 17A and Fig. Figures 17B illustrate setting the voltage command value according to a position of the extreme voltage according to embodiment 1 of the present invention. Fig. Figure 18 is a block diagram of an algorithm selection unit according to embodiment 1 of the present invention. Fig. Figure 19 is a block diagram of a voltage command calculation unit according to embodiment 2 of the present invention. Fig. Figure 20 is a hardware design diagram of the controller according to embodiment 1 of the present invention. DETAILED DESCRIPTION OF THE EXECUTION FORMS Execution form 1

[0009] A controller 400 for a rotary electric motor drive apparatus 1000 (hereinafter referred to simply as the controller 400) according to embodiment 1 will be explained with reference to the drawings. Fig. Figure 1 is a schematic diagram of the rotary electric motor drive apparatus 1000 and the controller 400 according to embodiment 1 of the present invention.

[0010] The rotary electric motor drive assembly 1000 is equipped with a converter 15, which can boost the power source voltage Vb of a DC power source B for output to system voltage lines 7, 8; and an inverter IN, which is provided between the converter 15 and a rotary electric motor MG and performs a power conversion between the DC power of the system voltage lines 7, 8 and the AC power that drives the rotary electric motor MG. In the present embodiment, the rotary electric motor MG is used as a drive power source for wheels, and the rotary electric motor drive assembly 1000 and the controller 400 are mounted in a vehicle (in this example, a hybrid vehicle). A plurality of sets (in this example, 2 sets) of the rotary electric motor MG and the inverter IN are provided.

[0011] Each of the first and second rotary electric machines MG1, MG2 is provided with a stator fixed to a non-rotating component and a rotor arranged and rotatably mounted on the radially inward side of the stator. In the present embodiment, the rotary electric machine MG is a permanent magnet synchronous rotary electric machine; the stator is provided with three-phase windings, and the rotor is provided with permanent magnets. Each of the first and second rotary electric machines MG1, MG2 functions as both an electric motor and an electric generator.

[0012] In the present embodiment, the first rotary electric machine MG1 operates as the electric generator driven by an internal combustion engine (not shown) and as the electric motor that starts the internal combustion engine. The second rotary electric machine MG2 is connected to the wheels via an output shaft and a reduction gear (not shown), operates as the electric motor that drives the wheels, and as the electric generator that performs regenerative power generation through a driving force from the wheels.

[0013] Next, the setup for driving the first and second rotary electric machines MG1 and MG2 is explained. A secondary battery, such as nickel-hydrogen or lithium-ion, is used for the DC power source B. An electric double-layer capacitor or similar device can also be used for the DC power source B. A positive electrode terminal of the DC power source B is connected to a positive electrical line 6 of the converter 15 on the power source side, and a negative electrode terminal of the DC power source B is connected to a negative electrical line 5 of the converter 15 on the power source side. A power source voltage sensor 10 is provided for detecting the power source voltage Vb of the DC power source B. An output signal from the power source voltage sensor 10 is fed to the controller 400.

[0014] A converter 15 is connected between the DC power source B and the system voltage lines 7, 8 and is a DC-DC converter that converts DC power. In the present embodiment, the converter is a voltage booster / downshifter with the function of a voltage booster chopper, which boosts or amplifies a power source voltage Vb of the DC power source B and outputs it to the system voltage lines 7, 8, and the function of a voltage downshift chopper, which reduces a system voltage VH, which is a DC voltage of the system voltage lines 7, 8, and outputs it to the DC power source B. The converter 15 is provided with at least one inductor, a switching device, and a freewheeling diode.

[0015] The converter 15 is equipped with a smoothing capacitor C1, which is connected between the positive electrical line 6 and the negative electrical line 5 on the power source side. A relay (not shown), which is switched on when the vehicle is in operation and switched off when the vehicle is stopped, is provided between the positive electrode terminal of the DC power source B and the positive electrical lines 6 on the power source side, and between the negative electrode terminal of the DC power source B and the negative electrical lines 5 on the power source side.

[0016] In the present embodiment, the converter 15 is provided with one inductor L1 for the common use of the voltage boost chopper and the voltage drop chopper, two switching devices Q3, Q4 for the voltage boost chopper, two freewheeling diodes D1, D2 for the voltage boost chopper, two switching devices Q1, Q2 for the voltage drop chopper, two freewheeling diodes D3, D4 for the voltage drop chopper, and one smoothing capacitor C0 for the common use of the voltage boost chopper and the voltage drop chopper. The four switching devices Q1, Q2, Q3, Q4 are connected in series from the positive electrode side in the order Q1, Q2, Q3 and Q4 between the positive electrode-side system voltage line 7 and the negative electrode-side system voltage line 8. Each of the four freewheeling diodes D1, D2, D3, D4 is inverted orThe circuit is connected in parallel to each of the four switching devices Q1, Q2, Q3, Q4. Each of the four switching devices Q1, Q2, Q3, Q4 is controlled on / off by a respective converter control signal S1, S2, S3 or S4, which has been output by the controller 400.

[0017] The inductor L1 is connected between a junction of switching device Q2 and switching device Q3 and the positive electrical line 6 on the power source side. The capacitor C2 is connected between the junction of switching device Q1 and switching device Q2 and between the junction of switching device Q3 and switching device Q4. The smoothing capacitor C0 is connected between the positive electrode-side system voltage line 7 and the negative electrode-side system voltage line 8. A system voltage sensor 13 is provided between the positive electrode-side system voltage line 7 and the negative electrode-side system voltage line 8 to detect the system voltage VH of the system voltage lines 7 and 8. An output signal from the system voltage sensor 13 is fed to the controller 400.

[0018] The DC voltage side of the first inverter IN1 and the second inverter IN2 are connected to the converter 15 via the common system voltage lines 7 and 8.

[0019] The first inverter IN1 is equipped with three sets of a series connection (branch), where a positive-electrode-side switching device Q11 (upper branch), connected to the positive-electrode-side system voltage line 7, and a negative-electrode-side switching device Q12 (lower branch), connected to the negative-electrode-side system voltage line 8, are connected in series, corresponding to the respective phase of the three phase windings. Thus, the first inverter IN1 is equipped with a total of six switching devices: the three positive-electrode-side switching devices Q11U, Q11V, Q11W and the three negative-electrode-side switching devices Q12U, Q12V, Q12W. Each of the freewheeling diodes D11U, D11V, D11W, D12U, D12V, D12W is connected in reverse or inverted parallel to the respective switching device Q11U, Q11V, Q11W, Q12U, Q12V, Q12W.Then, a connection node of the positive-electrode-side switching device Q11 and the negative-electrode-side switching device Q12 of each phase is connected to the winding of the corresponding phase in the first rotary electric machine MG1. A current sensor 27 for detecting a current flowing into the winding of a respective phase is provided on a wire of the respective phase, which connects the connection node of the switching devices and the winding. An output signal from the current sensor 27 is input to the controller 400. Each of the switching devices Q11U, Q11V, Q11W, Q12U, Q12V, Q12W is controlled on / off by one of the control signals from the first inverter S11, S12, S13, S14, S15, S16 output by the controller 400.

[0020] Similarly, the second inverter IN2 is equipped with three sets of a series connection (branch), where a positive-electrode-side switching device Q21 (upper branch), connected to the positive-electrode-side system voltage line 7, and a negative-electrode-side switching device Q22 (lower branch), connected to the negative-electrode-side system voltage line 8, are connected in series, corresponding to the respective phase of the three phase windings. Thus, the second inverter IN2 is equipped with a total of six switching devices: the three positive-electrode-side switching devices Q21U, Q21V, Q21W and the three negative-electrode-side switching devices Q22U, Q22V, Q22W. Each of the freewheeling diodes D21U, D21V, D21W, D22U, D22V, D22W is connected or switched in reverse or inverted parallel to the respective switching device Q21U, Q21V, Q21W, Q22U, Q22V, Q22W.Then, a connection node of the positive-electrode-side switching device Q21 and the negative-electrode-side switching device Q22 of each phase is connected to the winding of the corresponding phase in the second rotary electric machine MG2. A current sensor 27 for detecting a current flowing into the winding of each phase is provided on a wire of each phase, connecting the connection node of a switching device and the winding. An output signal from the current sensor 27 is input to the controller 400. Each of the switching devices Q21U, Q21V, Q21W, Q22U, Q22V, Q22W is controlled on / off by one of the control signals from the second inverter S21, S22, S23, S24, S25, S26 output by the controller 400.

[0021] The switching control of the controller 400 enables inverters IN1 and IN2 to convert the DC voltage from system voltage lines 7 and 8 into a three-phase AC voltage, output it to the rotary electric motors MG1 and MG2, and operate them as electric motors. The switching control of the controller 400 also enables inverters IN1 and IN2 to convert the three-phase AC voltage generated by the rotary electric motors MG1 and MG2 into a DC voltage and output it to system voltage lines 7 and 8.

[0022] The switching devices of the converter 15 and the inverter IN1, IN2 are IGBTs (Insulated Gate Bipolar Transistor), MOS (Metal Oxide Semiconductor) power transistors, bipolar power transistors, SiC, GaN or the like.

[0023] The rotary electric machines MG1 and MG2 are each equipped with a rotation angle sensor 28 (in this example, a rotary encoder) for detecting the respective rotation angle θ of the rotor. An output signal from the respective rotation angle sensor 28 is input to the controller 400. The controller 400 detects the respective rotation angle θ1 and θ2 of the rotary electric machines MG1 and MG2 based on the output signal from the respective rotation angle sensor 28 and calculates the respective rotational speed ω1 and ω2 (in this example, an angular velocity) of the rotary electric machines MG1 and MG2 based on the respective rotation angle θ1 and θ2.

[0024] The controller 400 is equipped with functional parts of a converter control unit 750, a voltage command calculation unit 700, an inverter control unit 600, and the like, which are described below. Each function of the controller 400 is implemented by processing circuits provided in the controller 400. In the present embodiment, as described in Fig. As shown in Figure 20, the controller 400 is equipped with the following processing circuits: a computer 90, such as a CPU (Central Processing Unit); memory devices 91, which exchange data with the computer 90; an input circuit 92, which inputs external signals to the computer 90; an output circuit 93, which outputs signals from the computer 90 to the outside; and the like. The memory devices 91 include a RAM (Random Access Memory), which can read and write data from the computer 90; a ROM (Read Only Memory), which can read data from the computer 90; and the like.The input circuit 92 is connected to various types of sensors and switches, such as the voltage sensors 10 and 13, and is equipped with an analog-to-digital converter (ADC) and the like for inputting output signals from the sensors and switches to the processing unit 90. The output circuit 93 is connected to electrical loads, such as a gate driver circuit, which drives the switching devices on and off, and is equipped with a control circuit and the like for outputting a control signal from the processing unit 90. In the present embodiment, the energy source voltage sensor 10, the system voltage sensor 13, the current sensor 27, and the rotary angle sensor 28 are connected to the input circuit 92. The switching devices (gate driver circuit) of the converter 15, the switching devices (gate driver circuit) of the inverters IN1 and IN2, and the like are connected to the output circuit 93.

[0025] The processing unit 90 then executes software elements (programs) stored in the memory device 91, as well as in ROM, and works together with other hardware devices in the controller 400, such as the memory device 91, the input circuit 92, and the output circuit 93, so that each function of the control units 750, 700, and 600 provided in the controller 400 is realized. Setting data elements, as well as characteristic map data to be used in the control units 750, 600, and 700, are stored as part of software elements (programs) in the memory device 91, as well as in ROM. Each function of the controller 400 will be described in detail below. <Wechselrichtersteuerabschnitt 600>

[0026] First, the inverter control unit 600 controls the operation of the rotary electric machine MG by switching the inverter IN's switching devices on and off. The inverter control unit 600 controls the switching devices of the inverter IN on and off such that the rotary electric machine MG delivers or outputs a torque of a torque command value Tqcom. The torque command value Tqcom is transmitted by an external controller of the controller 400 or other control units within the controller 400. In the present embodiment, the inverter control unit 600 is configured to perform current feedback control using a vector control method.In the present embodiment, the inverter control unit 600 is provided with the first inverter control unit 601, which controls the first inverter IN1 and the first rotary electric machine MG1, and the second inverter control unit 602, which controls the second inverter IN2 and the second rotary electric machine MG2.

[0027] Each of the first and second torque command values, Tqcom1 and Tqcom2, is set to a positive or negative value according to a drive condition. Specifically, during regenerative braking of the hybrid vehicle, the second torque command value, Tqcom2, is set to a negative value (Tqcom2 < 0). In this case, through a switching operation triggered by the control signal of the second inverter S21 to S26, the second inverter IN2 converts the AC voltage generated by the second rotary electric machine MG2 into a DC voltage and supplies the DC voltage (the system voltage VH) to converter 15.

[0028] Since the first inverter control unit 601 and the second inverter control unit 602 are similar designs, the first inverter control unit 601 will be described as a representative example in the following description.

[0029] As in Fig. As shown in Figure 3, the first inverter control unit 601 is equipped with a current command calculation unit 610, a current control unit 640, a voltage coordinate conversion unit 650, a PWM signal generation unit 660, a current coordinate conversion unit 620, and a speed detection unit 630. The speed detection unit 630 detects a rotation angle θ1 (a magnetic pole position) and a rotational angular velocity ω1 of the rotor of the first rotary electric machine MG1 based on the output signal of the rotation angle sensor 28 of the first rotary electric machine MG1.

[0030] The current command calculation unit 610 calculates a d-axis current command value Idcom and a q-axis current command value Iqcom, which express command values ​​of the current flowing into the three phase windings of the first rotary electric machine MG1, using the dq-axis rotational coordinate system of the first rotary electric machine MG1. The dq-axis rotational coordinate system consists of a d-axis defined in the direction of the N-pole (magnetic pole position) of the permanent magnet provided in the rotor of the first rotary electric machine MG1, and a q-axis defined in the direction leading the d-axis by 90 degrees (π / 2) at an electrical angle, and which is the two-axis rotational coordinate system that rotates synchronously with the rotation of the rotor at that electrical angle.

[0031] The current command calculation unit 610 calculates the d-axis current command value Idcom and the q-axis current command value Iqcom, which causes the first rotary electric machine MG1 to deliver a torque corresponding to the first torque command value Tqcom1. The current command calculation unit 610 calculates the dq-axis current command values ​​Idcom and Iqcom in accordance with the current vector control method of maximum torque / current control, field weakening control, and the like. In maximum torque / current control, the dq-axis current command values ​​Idcom and Iqcom are calculated to maximize the generated torque for the same current. In field weakening control, the d-axis current command value Idcom is made to increase more in the negative direction than the dq-axis current command values ​​Idcom and Iqcom calculated in maximum torque / current control.In field weakening control, the dq-axis current command values ​​Idcom and Iqcom are moved along an ellipse of a constant induced voltage (a voltage limiting ellipse) according to the first torque command value Tqcom1. In field weakening control, torque control is performed such that the amplitude of a fundamental frequency component is nearly fixed.

[0032] The current command calculation unit 610 calculates the dq-axis current command values ​​Idcom, Iqcom according to the first torque command value Tqcom1 by using characteristic map data in which the relationship between the first torque command value Tqcom1 and the dq-axis current command values ​​Idcom, Iqcom is set in preparation for each control procedure.

[0033] The current command calculation unit 610 calculates the dq-axis current command values ​​by the maximum torque / current control under the operating condition that the maximum torque / current control can perform, and calculates the dq-axis current command values ​​by the field weakening control under the operating condition that cannot perform the calculation of the dq-axis current command values ​​by the maximum torque / current control due to a limitation of the voltage limiting ellipse.

[0034] The current coordinate conversion unit 620 converts three-phase currents Iu, Iv, Iw, which flow into the winding of a respective phase and have been detected on the basis of the output signal of the current sensor 27 of the first rotary electric machine MG1, into a d-axis current Id and a q-axis current Iq, represented in the dq-axis rotational coordinate system, by performing a three-phase / two-phase conversion and a rotational coordinate conversion based on the magnetic pole position θ1. The current control unit 640 performs a current feedback control, which, by means of PI control and the like, changes a d-axis voltage command value Vd# and a q-axis voltage command value Vq#, which represent a command signal of the voltage applied to the first rotary electric machine MG1 in the dq-axis rotational coordinate system, so that the dq-axis currents Id, Iq approach the dq-axis current command values ​​Idcom, Iqcom.The voltage coordinate conversion unit 650 then converts the dq-axis voltage command values ​​Vd#, Vq# into three-phase AC voltage command values ​​Vu, Vv, Vw, which are the AC voltage command values ​​for each phase of the three-phase windings, by performing a fixed coordinate conversion and a two-phase / three-phase conversion based on the magnetic pole position θ1. The system voltage VH is also represented or reflected in the voltage coordinate conversions.

[0035] The PWM signal generation unit 660 compares each of the three-phase AC voltage command values ​​Vu, Vv, Vw with a carrier wave (a triangular wave) that has an amplitude equal to the system voltage VH and oscillates at a carrier frequency. It activates a square wave pulse when the AC voltage command value exceeds the carrier wave and deactivates it when the AC voltage command value falls below the carrier wave. Based on each phase of the three-phase square wave pulses, the PWM signal generation unit 660 generates the control signals for the first inverters S11 to S16 and outputs them to the first inverter IN1. <Wandlersteuereinheit 750>

[0036] The converter control unit 750 controls the converter 15 such that the system voltage VH, which is the DC voltage of the system voltage lines 7, 8, approaches a voltage command value VH# when the voltage command value VH# is greater than a power source voltage Vb. In the present embodiment, the converter control unit 750 detects the power source voltage Vb based on an output signal from the power source voltage sensor 10 and detects the system voltage VH based on an output signal from the system voltage sensor 13. The converter control unit 750 changes the duty cycle of the converter control signals S1 to S4 according to a PWM control method based on the system voltage VH and the voltage command value VH#.

[0037] In the case of performing a voltage boost operation on the converter 15, for example, the converter control unit 750 alternately sets the ON period of switching devices Q1, Q2 and the ON period of switching devices Q3, Q4, and changes the ratio of the two ON periods, thus changing a voltage boost ratio. In the case of performing a voltage drop operation on the converter 15, for example, the converter control unit 750 alternately sets the ON period of switching devices Q1, Q2 and the OFF period of switching devices Q1, Q2, Q3, Q4, and changes the ratio of the ON period to the OFF period, thus changing a voltage drop ratio. In the case where the voltage command value VH# is less than or equal to the power source voltage Vb, the converter control unit 750 switches all switching devices Q1, Q2, Q3, Q4 OFF and changes the DC power source B and the system voltage lines 7, 8 to a direct connection state.

[0038] During the voltage boost operation, converter 15 supplies the system voltage VH, which boosts the power source voltage Vb supplied by the DC power source B, to inverters IN1 and IN2. During the voltage reduction operation, converter 15 reduces the system voltage VH, which is supplied by inverters IN1 and IN2 via the smoothing capacitor C0, and supplies it to DC power source B. <Spannungsbefehl-Berechnungseinheit 700>

[0039] The voltage command calculation unit 700 calculates the voltage command value VH# within a range that is greater than or equal to the power source voltage Vb and less than or equal to the output maximum voltage Vmax of the converter 15.

[0040] Since the counter-electromotive force of the rotary electric machine MG will increase and the induced voltage will rise as the rotational speed ω and torque increase, the required minimum voltage VHL will be high, which is the minimum required system voltage VH in the case of maximum torque / current control of the rotary electric machine MG. To perform maximum torque / current control, it is necessary to make the system voltage VH higher than the required minimum voltage VHL. On the other hand, there is a limit to the voltage increase of the converter 15, and there is an upper limit (the output voltage limit Vmax) in the output voltage (system voltage VH) of the converter 15.

[0041] As in Fig. 4 and Fig. As shown in Figure 7, the voltage command calculation unit 700 is provided with a required minimum voltage calculation unit 800, which calculates the required minimum voltage VHL, which is the minimum system voltage VH required in the case of performing the maximum torque / current control of the rotary electromachine MG, under conditions of the current torque command value Tqcom and the current speed ω of the rotary electromachine MG.

[0042] In the case where the required minimum voltage VHL is less than or equal to the output maximum voltage Vmax, the voltage command calculation unit 700 sets the voltage command value VH# within a candidate voltage range that is greater than or equal to the required minimum voltage VHL or less than or equal to the output maximum voltage Vmax; and the inverter control unit 600 performs the maximum torque / current control. Conversely, in the case where the required minimum voltage VHL is greater than the output maximum voltage Vmax, the voltage command calculation unit 700 sets the output maximum voltage Vmax as the voltage command value VH# and performs the field weakening control.

[0043] Fig. Figure 5 shows a torque-speed characteristic to illustrate a maximum torque region that performs maximum torque / current control. A vertical axis represents the torque of the rotary electric machine MG, a horizontal axis represents the rotational speed ω of the rotary electric machine MG, and the solid line in the figure shows a maximum torque curve at a given rotational speed ω when maximum torque / current control is performed. In the case where the rotational speed ω is less than or equal to a base rotational speed, the maximum output torque of the rotary electric machine MG is determined by limiting the current of the rotary electric machine MG to a rated current, and a constant current is applied to a change in rotational speed ω.In the case where the rotational speed ω is greater than the base rotational speed, the maximum output torque of the rotary electric machine MG is determined by limiting a line voltage of the rotary electric machine MG to the system voltage VH and decreases as the rotational speed ω increases.

[0044] A plurality of solid line curves of Fig. Figure 5 shows a change in the maximum torque curve of the maximum torque / current control when the system voltage VH is changed. When the system voltage VH is increased from the power source voltage Vb to the output limit voltage Vmax, as shown in Fig. As shown in Figure 5, the maximum torque line and the base speed are shifted towards the high-speed side, and the maximum torque region can be extended. In the case where the system voltage VH is the output maximum voltage Vmax, the base speed becomes the highest, and the maximum torque region becomes the widest. The system voltage VH is modified within a range from the required minimum voltage VHL to the output maximum voltage Vmax so that the maximum torque / current control is performed in the maximum torque region according to this output maximum voltage Vmax.

[0045] A region on a high-speed side and a high-torque side, rather than the maximum torque region corresponding to this output voltage limit Vmax, is set as a field weakening region on which field weakening control is performed. In this field weakening region, the system voltage VH is controlled or regulated to the output voltage limit Vmax.

[0046] The maximum torque region is a region where the required minimum voltage VHL to perform maximum torque / current control becomes less than or equal to the system voltage VH (the output upper limit voltage Vmax); the field weakening region is a region where the required minimum voltage VHL becomes greater than the system voltage VH (output upper limit voltage Vmax).

[0047] Fig. Figure 6 shows a figure plotting equal-torque curves for each torque at a given rotational speed, where a vertical axis represents the total power loss of the inverter IN and the rotary motor MG, and a horizontal axis represents the system voltage VH. Since, in general, a large motor current is required to produce a large torque, the loss will be correspondingly large. A dashed line in Fig. Figure 6 shows the line where the maximum torque region and the field weakening region switch; and the dashed line is a line where the system voltage VH becomes the required minimum voltage VHL. Fig. Figure 6 shows that the total power loss of the inverter IN and the rotary electric machine MG becomes small near the line where the maximum torque region and the field weakening region switch.

[0048] In the field weakening region, the total power loss of the inverter IN and the rotary engine MG becomes large when the system voltage VH becomes low. Therefore, if the first and second rotary engines MG1 and MG2 are present, and the required minimum voltage VHL of the first rotary engine MG1 (hereinafter referred to as the first required voltage Vmg1) differs from the required minimum voltage VHL of the second rotary engine MG2 (hereinafter referred to as the second required voltage Vmg2), then by setting the voltage command value VH# so that the system voltage VH is close to a higher value than Vmg1 and Vmg2, both the first and second rotary engines MG1 and MG2 can be operated in the maximum torque region, and losses can be reduced.In the present embodiment, as described later, the required minimum voltage calculation unit 800 sets a voltage higher from the first required voltage Vmg1 and the second required voltage Vmg2 than the required minimum voltage VHL.

[0049] In the case of maximum torque / current control, the voltage command value VH# can be set within a voltage range from the required minimum voltage VHL to the output maximum voltage Vmax, and there is flexibility in setting this value. The voltage command calculation unit 700 is then equipped with a minimum loss command calculation unit 802, which calculates a low-loss voltage VHLL. This low-loss voltage is the system voltage VH for which the total power loss of the converter 15 and the inverter IN is minimized, within a candidate voltage range that is greater than or equal to the required minimum voltage VHL and less than or equal to the output maximum voltage Vmax of the converter 15. The low-loss voltage VHLL is then set as the voltage command value VH#.

[0050] However, calculating the system voltage VH for which the total power loss is minimized presented a problem: the data processing load increases. For example, it is necessary to calculate the power loss using power loss characteristic map data, as described in Fig. 6 shown how to calculate over each of a plurality of candidate voltages and to determine the voltages for which the total power loss is minimized from a plurality of candidate voltages.

[0051] Thus, the loss minimum command calculation unit 802 calculates a converter loss coefficient, which represents a power loss characteristic of the converter 15, which is a coefficient of a polynomial in which the system voltage VH is a variable, in the case where the required minimum voltage VHL is greater than or equal to the energy source voltage Vb and less than or equal to the output maximum voltage Vmax of the converter 15; and calculates an inverter loss coefficient, which represents a power loss characteristic of the inverter IN, which is a coefficient of the polynomial in which the system voltage VH is a variable.The 802 minimum loss command calculation unit then calculates a total sum of the converter loss coefficient and the inverter loss coefficient for each order of the polynomials; calculates, based on the calculated total loss coefficient for each order, a low loss voltage VHLL within the candidate voltage range; and sets the low loss voltage VHLL as the voltage command value VH.

[0052] Since, according to this training, the low-loss voltage VHLL is calculated based on the total loss coefficient, which is the sum of the converter loss coefficient and the inverter loss coefficient for each order of the polynomials, it is not necessary to perform a calculation that uses characteristic map data for each of a plurality of candidate voltages, and this can reduce the data processing load. Because each power loss characteristic is approximated by the polynomial, the amount of memory required by a storage device can be significantly reduced compared to directly converting the power loss characteristic into characteristic map data.

[0053] The design of each unit of the voltage command calculation unit 700 is explained in detail below. <Erforderliche-Minimumspannung-Berechnungseinheit 800>

[0054] First, the detailed design of the Required Minimum Voltage Calculation Unit 800 is explained. The Required Minimum Voltage Calculation Unit 800 calculates a first required voltage Vmg1, which is the minimum system voltage VH required when performing maximum torque / current control of the first rotary electric machine MG1, under conditions of the current torque command value Tqcom1 and the current speed ω1 of the first rotary electric machine MG1. The Required Minimum Voltage Calculation Unit 800 also calculates a second required voltage Vmg2, which is the minimum system voltage VH required when performing maximum torque / current control of the second rotary electric machine MG2, under conditions of the current torque command value Tqcom2 and the current speed ω2 of the second rotary electric machine MG2.Then the Required Minimum Voltage calculation unit 800 sets a maximum value from the first required voltage Vmg1 and the second required voltage Vmg2 as the required minimum voltage VHL.

[0055] According to this training, the minimum system voltage VH, at which both the first and second rotary motors MG1 and MG2 can perform maximum torque / current control, can be set as the required minimum voltage VHL. Therefore, field weakening control, which increases power loss, can be avoided, and the power loss of the first and second inverters and rotary motors can be reduced.

[0056] In the present embodiment, as in Fig. As shown in Figure 7, the Required Minimum System Voltage Calculation Unit 800 is provided with first and second Maximum Torque Current Command Calculation Units 1110, first and second Inductance Calculation Units 1120, first and second Required Voltage Calculation Units 1130 and a Maximum Voltage Calculation Unit 1140.

[0057] The first maximum torque current command calculation unit 1110 calculates the first maximum torque dq-axis current command values ​​Idcom_loss1, Iqcom_loss1, which cause the first rotary electric machine MG1 to output the first torque command value Tqcom1, respectively, by performing maximum torque / current control. The first maximum torque current command calculation unit 1110 calculates the dq-axis current command values ​​using a similar procedure to the current command calculation unit 610 of the first inverter control unit 601.

[0058] In the present embodiment, the first maximum torque current command calculation unit 1110 calculates the first maximum torque dq-axis current command values ​​Idcom_loss1, Iqcom_loss1 on a fixed operating condition of a pre-set rotational speed (for example, ω1 = 0) less than or equal to the base rotational speed. According to this embodiment, even if the current operating condition is the field weakening region, the current command values ​​for the maximum torque / current control can be reliably calculated.

[0059] The first maximum torque current command calculation unit 1110 calculates the first maximum torque dq-axis current command values ​​Idcom_loss1, Iqcom_loss1 corresponding to the first torque command value Tqcom1 by using characteristic map data in which the relationship between the first torque command value Tqcom1 and the first maximum torque dq-axis current command values ​​Idcom_loss1, Iqcom_loss1 is predefined. The characteristic map data is predefined based on measured values ​​or a magnetic field analysis. The q-axis current command value and the d-axis current command value corresponding to the respective current command value are set to the characteristic map data at the interval (unit) of a predetermined torque command value.

[0060] The first inductance calculation unit 1120 calculates the first dq-axis inductances Ld_loss1, Lq_loss1 of the first rotary electric machine MG1 according to the first maximum torque dq-axis current command values ​​Idcom_loss1, Iqcom_loss1. The first inductance calculation unit 1120 calculates the first dq-axis inductances Ld_loss1, Lq_loss1 according to the first maximum torque dq-axis current command values ​​Idcom_loss1, Iqcom_loss1 by using characteristic map data in which the relationship between the first maximum torque dq-axis current command values ​​Idcom_loss1, Iqcom_loss1 and the first dq-axis inductance Ld_loss1, Lq_loss1 is predefined.

[0061] The first required voltage calculation unit 1130 calculates the first required voltage Vmg1 using a subsequent equation based on the first maximum torque dq axis current command values ​​Idcom_loss1, Iqcom_loss1, the first dq axis inductances Ld_loss1, Lq_loss1 and the rotational speed ω1 of the first rotary electromachine MG1. Vmg1=1η1(R1⋅Idcom_loss1−ω1⋅Lq_loss1⋅Iqcom_loss1)2+(R1⋅Idcom_loss1+ω1⋅Ld_loss1⋅Idcom_loss1+ω1⋅ϕmag1)2

[0062] Here, η1 is a voltage utilization factor and expresses the ratio that converts the system voltage VH into the line voltage or mains voltage of the first rotary electric machine MG1. Therefore, equation (1) calculates the minimum line voltage of the first rotary electric machine MG1, required when performing maximum torque / current control of the first rotary electric machine MG1, by calculating a square root and converting the line voltage into the system voltage using η1. R1 is a resistance of the winding of the stator of the first rotary electric machine MG1; and φmag1 is a magnetic flux of the permanent magnet of the rotor of the first rotary electric machine MG1.

[0063] By means of a similar procedure to the first maximum torque current command calculation unit 1110, the second maximum torque current command calculation unit 1110 calculates the second maximum torque dq axis current command values ​​Idcom loss2, Iqcom_losss2, which cause the second rotary electric machine MG2 to output the second torque command value Tqcom2 by performing the maximum torque / current control.

[0064] By a similar procedure to the first inductance calculation unit 1120, the second inductance calculation unit 1120 calculates second dq-axis inductances Ld_loss2, Lq_loss2 of the second rotary electric machine MG2 according to the second maximum torque dq-axis current command values ​​Idcom_loss2, Iqcom_loss2.

[0065] The second required voltage calculation unit 1130 calculates the second required voltage Vmg2 using a subsequent equation based on the second maximum torque dq axis current command values ​​Idcom loss2, Iqcom_loss2, the second dq axis inductances Ld_loss2, Lq_loss2 and the rotational speed ω2 of the second rotary electric machine MG2. Vmg2=1η2(R2⋅Idcom_loss2−ω2⋅Lq_loss2⋅Iqcom_loss2)2+(R2⋅Idcom_loss2+ω2⋅Ld_loss2⋅Idcom_loss2+ω2⋅ϕmag2)2

[0066] Here, η2 is a voltage utilization factor and expresses the ratio that converts the system voltage VH into the line voltage of the second rotary electric machine MG2. R2 is a resistance of the winding of the stator of the second rotary electric machine MG2; and φmag2 is a magnetic flux of the permanent magnet of the rotor of the second rotary electric machine MG2.

[0067] The maximum voltage calculation unit 1140 calculates a maximum value from the first required voltage Vmg1, the second required voltage Vmg2 and the energy source voltage Vb using a subsequent equation and sets the maximum value as the required minimum voltage VHL. VHL=MAX(Vmg1, Vmg2, Vb)

[0068] Here, MAX(A, B, C) is a function that returns the largest value among A, B, and C. VHL = MAX(Vmg1, Vmg2) can be used. <Verlustminimumbefehl-Berechnungseinheit 802>

[0069] As mentioned above, the loss minimum command calculation unit 802 calculates the converter loss coefficient, which represents the power loss characteristic of converter 15, and which is the coefficient of the polynomial in which the system voltage VH is a variable, in the case that the required minimum voltage VHL is greater than or equal to the energy source voltage Vb and less than or equal to the output maximum voltage Vmax of converter 15; calculates the inverter loss coefficient, which represents the power loss characteristic of inverter IN, and which is the coefficient of the polynomial in which the system voltage VH is a variable; and calculates the sum of the converter loss coefficient and the inverter loss coefficient for each order of the polynomials.Based on the calculated total loss coefficient for each order, the loss minimum command calculation unit 802 then calculates the low loss voltage VHLL, which is the system voltage VH, from which the total power loss of the converter 15 and the inverter IN will be minimized, within the candidate voltage range that is greater than or equal to the required minimum voltage VHL and less than or equal to the output maximum voltage Vmax of the converter 15; and sets the low loss voltage VHLL as the voltage command value VH#.

[0070] First, the power loss characteristics of converter 15 will be explained. Fig. Figure 8 shows the power loss characteristic in the case where the output (output power) of converter 15 is low (in this case, the output is zero). A horizontal axis is the system voltage VH, and a vertical axis is the power loss of converter 15. The power loss of converter 15 becomes an upward-projected parabolic characteristic in a region where the system voltage VH ranges from the power source voltage Vb up to twice the value 2×Vb of the power source voltage Vb. In a region where the system voltage VH is greater than or equal to twice the value 2×Vb of the power source voltage, the power loss of converter 15 becomes a monotonic rising characteristic, with the power loss of converter 15 increasing as the system voltage VH increases.If the required minimum voltage VHL is within the range of Vb to 2×Vb, the total power loss of the rotary electric motor drive apparatus 1000 will therefore not necessarily be minimized simply by setting the required minimum voltage VHL as the voltage command value VH#.

[0071] Therefore, the loss minimum command calculation unit 802 determines whether or not the required minimum voltage VHL is within the range from the energy source voltage Vb up to twice the value 2×Vb of the energy source voltage; and in the case of within the range, it performs a processing that sets the voltage command value VH# taking into account the upwardly projected parabolic power loss characteristic of the converter 15.

[0072] Fig. Figure 9 is a flowchart illustrating the processing of the Loss Minimum Instruction Unit 802. In step ST100, the Loss Minimum Instruction Unit 802 first reads information about the power source voltage Vb, the first and second required voltages Vmg1 and Vmg2, and the required minimum voltage VHL. In step ST110, the Loss Minimum Instruction Unit 802 then determines whether the required minimum voltage VHL is within the range that is greater than or equal to the power source voltage Vb and less than or equal to twice the value 2×Vb of the power source voltage. If it is determined that the required minimum voltage VHL is within the range, it proceeds to step ST120. If it is determined that the required minimum voltage VHL is outside the range, it proceeds to step ST165.

[0073] In step ST120, the loss minimum instruction calculation unit 802 calculates the converter loss coefficient, which represents the power loss characteristic of converter 15, which is the coefficient of the polynomial in which the system voltage VH is a variable; calculates the inverter loss coefficient, which represents the power loss characteristic of inverter IN, which is the coefficient of the polynomial in which the system voltage VH is a variable; and calculates the total sum of the converter loss coefficient and the inverter loss coefficient for each order of the polynomials (the total sum loss coefficient).

[0074] In the present embodiment, the polynomial representing the power loss characteristic of the converter 15 is a polynomial whose order is less than or equal to the second order (in this example, the second order). Ploss_dcdc(VH)=A0dcdc+A1dcdc⋅VH+A2dcdc⋅VH2

[0075] Here, Ploss_dcdc is the power loss of converter 15, A0dcdc is a zeroth-order converter loss coefficient, A1dcdc is a first-order converter loss coefficient, and A2dcdc is a second-order converter loss coefficient.

[0076] As in Fig. As shown in Figure 10, the loss minimum instruction calculation unit 802 calculates the converter loss coefficient according to the current power source voltage Vb and the current output of the converter 15 over each order by using a relationship characteristic in which a relationship between the power source voltage Vb, the output (output power) of the converter 15, and the converter loss coefficient is pre-established. In this example, the loss minimum instruction calculation unit 802 calculates the zeroth-order converter loss coefficient A0dcdc according to the current power source voltage Vb and the current output of the converter 15 by using characteristic map data in which a relationship between the power source voltage Vb, the output of the converter 15, and the zeroth-order converter loss coefficient A0dcdc is pre-established.Similarly, the Loss Minimum Instruction Unit 802 calculates the first-order converter loss coefficient A1dcdc according to the current power source voltage Vb and the current output of converter 15 by using characteristic map data in which a relationship between the power source voltage Vb, the output of converter 15, and the first-order converter loss coefficient A1dcdc is predefined. Furthermore, the Loss Minimum Instruction Unit 802 calculates the second-order converter loss coefficient A2dcdc according to the current power source voltage Vb and the current output of converter 15 by using characteristic map data in which a relationship between the power source voltage Vb, the output of converter 15, and the second-order converter loss coefficient A2dcdc is predefined.

[0077] Each of the converter loss coefficients is pre-set based on power loss data of converter 15 for a system voltage range VH between the energy source voltage Vb and twice the value 2×Vb of the energy source voltage. Each of the converter loss coefficients is pre-set using the least squares method based on a measured value and loss data calculated by a loss analysis of the converter.

[0078] In the present embodiment, the loss minimization unit 802 calculates the inverter loss coefficients for each of the first and second inverters IN1 and IN2. The loss minimization unit 802 calculates the inverter loss coefficient, which is a coefficient of a polynomial in which the system voltage VH is a variable and which represents a power loss characteristic of the rotary engine MG in addition to the inverter IN. That is, the loss minimization unit 802 calculates the first inverter loss coefficient, which is a coefficient of a polynomial in which the system voltage VH is a variable and which represents a sum power loss characteristic of the first inverter IN1 and the first rotary engine MG1.The loss minimum command calculation unit 802 calculates the second inverter loss coefficient, which is a coefficient of a polynomial in which the system voltage VH is a variable, and which represents a total power loss characteristic of the second inverter IN2 and the second rotary electric machine MG2.

[0079] The polynomial representing the total power loss characteristic of the first inverter IN1 and the first rotating electric machine MG1 is a polynomial whose order is less than or equal to the second order (in this example the second order), as shown in the following equation. Ploss_mg1(VH)=A0mg1+A1mg1⋅VH+A2mg1⋅VH2

[0080] Here, Ploss_mg1 is the total power loss of the first inverter IN1 and the first rotary electric machine MG1, A0mg1 is a zero-order loss coefficient of the first inverter, A1mg1 is a first-order loss coefficient of the first inverter, and A2gm1 is a second-order loss coefficient of the first inverter.

[0081] As in Fig. As shown in Figure 11, the loss minimum instruction calculation unit 802 calculates the loss coefficient of the first inverter or the first inverter loss coefficient according to the current torque instruction value Tqcom1 and the current speed ω1 of the first rotating electric machine MG1 over each order by using a relationship characteristic in which a relationship between the output torque of the first rotating electric machine MG1, the speed ω1 of the first rotating electric machine MG1 and the loss coefficient of the first inverter is pre-established.In this example, the loss minimum instruction calculation unit 802 calculates a zero-order loss coefficient of the first inverter A0mg1 according to the current torque instruction value Tqcom1 and the current speed ω1 of the first rotary electric machine MG1 by using characteristic map data in which a relationship between the output torque of the first rotary electric machine MG1, the speed ω1 of the first rotary electric machine MG1 and the zero-order loss coefficient of the first inverter A0mg1 is pre-established.Similarly, the loss minimum instruction calculation unit 802 calculates a first-order loss coefficient of the first inverter A1mg1 according to the current torque instruction value Tqcom1 and the current speed ω1 of the first rotary electric machine MG1 by using characteristic map data in which a relationship between the output torque of the first rotary electric machine MG1, the speed ω1 of the first rotary electric machine MG1 and the first-order loss coefficient of the first inverter A1mg1 is pre-established.The loss minimum command calculation unit 802 calculates a second-order loss coefficient of the first inverter A2mg1 according to the current torque command value Tqcom1 and the current speed ω1 of the first rotary electric machine MG1 by using characteristic map data in which a relationship between the output torque of the first rotary electric machine MG1, the speed ω1 of the first rotary electric machine MG1 and the second-order loss coefficient of the first inverter A2mg1 is pre-established.

[0082] Each of the loss coefficients of the first inverter is pre-set based on the total power loss data of the first inverter IN1 and the first rotary electric machine MG1 for a system voltage range VH between the energy source voltage Vb and twice the value 2×Vb of the energy source voltage. Each of the loss coefficients of the first inverter is pre-set using the least squares method based on a measured value, a magnetic field analysis, and loss data calculated by an inverter loss analysis.

[0083] The polynomial representing the total power loss characteristic of the second inverter IN2 and the second rotary electric machine MG2 is a polynomial whose order is less than or equal to the second order (in this example the second order), as shown in the following equation. Ploss_mg2(VH)=A0mg2+A1mg2⋅VH+A2mg2⋅VH2

[0084] Here, Ploss_mg2 is the total power loss of the second inverter IN2 and the second rotary motor MG2, A0mg2 is a zero-order loss coefficient of the second inverter, A1mg2 is a first-order loss coefficient of the second inverter, and A2gm2 is a second-order loss coefficient of the second inverter.

[0085] As in Fig. As shown in Figure 12, the loss minimum instruction calculation unit 802 calculates the loss coefficient of the second inverter or the second inverter loss coefficient according to the current torque instruction value Tqcom2 and the current speed ω2 of the second rotary electric machine MG2 over each order by using a relationship characteristic in which a relationship between the output torque of the second rotary electric machine MG2, the speed ω2 of the second rotary electric machine MG2 and the loss coefficient of the second inverter is pre-established.In this example, the loss minimum instruction calculation unit 802 calculates a zero-order loss coefficient of the second inverter A0mg2 according to the current torque instruction value Tqcom2 and the current speed ω2 of the second rotary electric machine MG2 by using characteristic map data in which a relationship between the output torque of the second rotary electric machine MG2, the speed ω2 of the second rotary electric machine MG2 and the zero-order loss coefficient of the second inverter A0mg2 is pre-established.Similarly, the loss minimum instruction calculation unit 802 calculates a first-order loss coefficient of the second inverter A1mg2 according to the current torque instruction value Tqcom2 and the current speed ω2 of the second rotary electric machine MG2 by using characteristic map data in which a relationship between the output torque of the second rotary electric machine MG2, the speed ω2 of the second rotary electric machine MG2 and the first-order loss coefficient of the second inverter A1mg2 is pre-established.The loss minimum command calculation unit 802 calculates a second-order loss coefficient of the second inverter A2mg2 according to the current torque command value Tqcom2 and the current speed ω2 of the second rotary electric machine MG2 by using characteristic map data in which a relationship between the output torque of the second rotary electric machine MG2, the speed ω2 of the second rotary electric machine MG2 and the second-order loss coefficient of the second inverter A2mg2 is pre-established.

[0086] Each of the loss coefficients of the second inverter is pre-set based on the total power loss data of the second inverter IN2 and the second rotary motor MG2 for a system voltage range VH between the energy source voltage Vb and twice the value 2×Vb of the energy source voltage. Each of the loss coefficients of the second inverter is pre-set using the least squares method based on the measured value, a magnetic field analysis, and loss data calculated by an inverter loss analysis.

[0087] The Loss Minimum Command Calculation Unit 802 calculates the total sum of the converter loss coefficients, the loss coefficients of the first inverter and the loss coefficients of the second inverter for each order of the polynomial (the total loss coefficient), as shown in the following equation. A0=A0dcdc+A0mg1+A0mg2A1=A1dcdc+A1mg1+A1mg2A2=A2dcdc+A2mg1+A2mg2

[0088] Here, A0 is a zeroth-order total loss coefficient, A1 is a first-order total loss coefficient, and A2 is a second-order total loss coefficient.

[0089] A total power loss Ploss, summing the power loss Ploss_dcdc of converter 15, the total power loss Ploss_mgl of the first inverter IN1 and the first rotary motor MG1, and the total power loss Ploss_mg2 of the second inverter IN2 and the second rotary motor MG2, becomes a polynomial using the total loss coefficients A0, A1, A2, as shown in the following equation. Ploss(VH)=A0+A1⋅VH+A2⋅VH2

[0090] Next, in step ST130, the loss minimum command calculation unit 802 determines whether the total second-order loss coefficient A2 is positive; if A2 is positive, it proceeds to step ST140; if A2 is 0 or negative, it proceeds to step ST160. As in Fig. 13A and Fig. As shown in Figure 13B, in the case where the total second-order loss coefficient A2 is a positive value, the total power loss is projected downwards, and in the case where the total second-order loss coefficient A2 is a negative value, the total power loss is projected upwards.

[0091] In step ST140, the Loss Minimum Instruction Unit 802 calculates an extreme voltage Vpl, which is the system voltage VH at which the total power loss is minimized, based on the total second-order and first-order loss coefficients A2 and A1. The Loss Minimum Instruction Unit 802 calculates the extreme voltage Vpl using the following equation, a theoretical formula that calculates the extremum of a quadratic equation. Therefore, by simply calculating this extreme voltage using the total second-order and first-order loss coefficients A2 and A1, the extreme voltage Vpl can be determined, and the processing load can be significantly reduced. Vpl=−A1 / (2×A2)

[0092] Then the loss minimum command calculation unit 802 determines whether or not the extreme voltage Vpl is greater than twice the value 2×Vb of the energy source voltage; in the case where the extreme voltage Vpl is greater, it proceeds to step ST171; in the case where the extreme voltage Vpl is not greater, it proceeds to step ST150.

[0093] In step ST171, as in Fig. As shown in Figure 14, in the case where the extreme voltage Vpl is greater than twice the value 2×Vb of the power source voltage, the minimum-loss command calculation unit 802 sets twice the value 2×Vb of the power source voltage as the low-loss voltage VHLL. This is because the power loss of the converter 15 will become large if the system voltage VH becomes greater than twice the value 2×Vb of the power source voltage, as with the use of Fig. 8 explained.

[0094] In step ST150, the loss minimum command calculation unit 802 determines whether or not the extreme voltage Vpl is less than the required minimum voltage VHL; if the extreme voltage Vpl is less, it proceeds to step ST172; if the extreme voltage Vpl is not less, it proceeds to step ST173.

[0095] In step ST172, as in Fig. As shown in Figure 15, in the case where the extreme voltage Vpl is less than the required minimum voltage VHL, the loss minimum command calculation unit 802 sets the required minimum voltage VHL as the low-loss voltage VHLL. This is because the total power loss becomes a minimum at the required minimum voltage VHL within a settable range of the voltage command value VH# that is greater than the required minimum voltage VHL.

[0096] In step ST173, as in Fig. As shown in Figure 16, in the case where the extreme voltage Vpl is within the range from the required minimum voltage VHL to twice the value 2×Vb of the power source voltage, the loss minimum command calculation unit 802 sets the extreme voltage Vpl as the low loss voltage VHLL. This is because the total power loss becomes a minimum at the extreme voltage Vpl.

[0097] In the case where the total second-order loss coefficient A2 is determined to be 0 or a negative value in step ST130, the loss minimum command calculation unit 802 determines a smaller of the total power loss at the required minimum voltage VHL and the total power loss at twice the value 2×Vb of the power source voltage, based on the total loss coefficient, in step ST160. In the present embodiment, as shown in Fig. 17A and Fig. As shown in Figure 17B, the Loss Minimum Control Unit 802 determines whether the extreme voltage Vpl, which is the system voltage VH from which the total power loss is maximized, is less than the average voltage Vave of the required minimum voltage VHL and twice the value 2×Vb of the power source voltage. If the extreme voltage Vpl is less than this, it proceeds to step ST174; if the extreme voltage Vpl is not less than this, it proceeds to step ST175. The Loss Minimum Control Unit 802 calculates the average voltage Vave using the following equation. Vave=(VHL+2×Vb) / 2

[0098] As in Fig. 17A and Fig. As shown in Figure 17B, in the case where the extreme voltage Vpl is less than the average voltage Vave, it can be estimated that the total power loss will be at a minimum at twice the value 2×Vb of the energy source voltage; and in the case where the extreme voltage Vpl is greater than the average voltage Vave, it can be estimated that the total power loss will be at a minimum at the required minimum voltage VHL.

[0099] Therefore, in step ST174, if the extreme voltage Vpl is less than the average voltage Vave, the loss minimum command calculation unit 802 sets the low loss voltage VHLL to twice the value 2×Vb of the power source voltage. Conversely, in step ST175, if the extreme voltage Vpl is greater than or equal to the average voltage Vave, the loss minimum command calculation unit 802 sets the required minimum voltage VHLL to the low loss voltage VHLL.

[0100] In step ST160, the loss minimum command calculation unit 802 can calculate the total power loss Ploss (VHL) in the case where the system voltage VH is the required minimum voltage VHL, and the total power loss Ploss (2×Vb) in the case where the system voltage VH is twice the value 2×Vb of the power source voltage, using the calculation equation for the total power loss Ploss of equation (8); and determine which is smaller.

[0101] In the case where it is determined that the required minimum voltage VHL is outside the range from the power source voltage Vb up to twice the value 2×Vb of the power source voltage, and the required minimum voltage VHL is greater than twice the value 2×Vb of the power source voltage, in step ST110, the loss minimum command calculation unit 802 determines whether or not the required minimum voltage VHL is less than the output maximum voltage Vmax, in step ST165; and in the case where the required minimum voltage VHL is less, it proceeds to step ST176, in the case where the required minimum voltage VHL is not less, it proceeds to step ST177.

[0102] In step ST176, where the required minimum voltage VHL is within the range of twice the value 2×Vb of the power source voltage up to the output maximum voltage Vmax, the loss minimum instruction calculation unit 802 sets the required minimum voltage VHL as the low-loss voltage VHLL. This is because the total power loss is minimized when the required minimum voltage VHL is within the settable range of the voltage instruction value VH#, which is greater than the required minimum voltage VHL.

[0103] On the other hand, in step ST177, in the case where the required minimum voltage VHL is greater than or equal to the output maximum voltage Vmax, the minimum loss instruction calculation unit 802 sets the output maximum voltage Vmax as the low loss voltage VHLL. <Algorithmusauswahleinheit 803>

[0104] When the output (output power) of converter 15 becomes large, the power loss characteristic of converter 15 approaches the monotonic rising characteristic, with which the power loss increases monotonically with the increase in the system voltage VH, from which in Fig. The characteristic shown in Figure 8 is relevant here. If the output of converter 15 exceeds a predetermined level, the total power loss will therefore be at a minimum at the required minimum voltage VHL.

[0105] In the present embodiment, as in Fig. 4 and Fig. As shown in Figure 18, the voltage command calculation unit 700 is equipped with an algorithm selection unit 803, which, in the case where the output (output power) of the converter 15 becomes greater than or equal to a pre-set determination output Pmg_th, replaces it with the voltage command value VH# (the low-loss voltage VHLL) set by the minimum-loss command calculation unit 802, and sets the required minimum voltage VHL as the final voltage command value VH#.

[0106] The algorithm selection unit 803 is provided with a rotary electric machine output calculation unit 801, which calculates an output Pmg of the rotary electric machine MG. In the present embodiment, the first rotary electric machine output calculation unit 801 calculates an output Pmg1 of the first rotary electric machine MG1 based on the first torque command value Tqcom1 and the rotational speed ω1 of the first rotary electric machine MG1 (for example, Pmg1 = Tqcom1 × ω1). The second rotary electric machine output calculation unit 801 calculates an output Pmg2 of the second rotary electric machine MG2 based on the second torque command value Tqcom2 and the rotational speed ω2 of the second rotary electric machine MG2 (for example, Pmg2 = Tqcom2 × ω2).

[0107] The algorithm selection unit 803 is equipped with a maximum value selection unit 910, which selects the larger of the two absolute values ​​of the output Pmg1 of the first rotary electric machine MG1 and the output Pmg2 of the second rotary electric machine MG2, and outputs the larger value as the maximum rotary electric machine output Pmg_max. The maximum value selection unit 910 can output the absolute value of the sum of the output Pmg1 of the first rotary electric machine MG1 and the output Pmg2 of the second rotary electric machine MG2 as the maximum rotary electric machine output Pmg_max.

[0108] In the case where the maximum rotary engine output Pmg_max is greater than or equal to the determination output or determination output Pmg_th, the algorithm selection unit 803 directly outputs the low-loss voltage VHLL, set by the minimum-loss instruction calculation unit 802, as the final voltage instruction value VH#; and in the case where the maximum rotary engine output Pmg_max is less than the determination output Pmg_th, the algorithm selection unit 803 outputs the required minimum voltage VHL as the final voltage instruction value VH#. <zusammenfassung>

[0109] Without calculating each power loss using map data of a power loss characteristic and without searching for the voltage command value that minimizes the total power loss, it is possible, according to the above configuration, to calculate the power loss for each of a multitude of voltage candidates based on the coefficient of the polynomial representing each power loss characteristic using a determination logic and a simple calculation equation, thus reducing the data processing load. Although map data is used to calculate the coefficient of each order, it is not necessary to perform a calculation using map data for each of a multitude of voltage candidates, since the value corresponding to the current operating condition is read from the map data only once.Since each power loss characteristic is approximated by the simple second-order polynomial, the number of memory locations in the storage device can be significantly reduced compared to directly converting the power loss characteristic into characteristic map data. In the example case where the system voltage VH changes from 150V to 650V, if a power loss characteristic data table is prepared with 50V units, ten sets of a data table are required; however, if approximated with the quadratic equation, the number of data tables can be reduced to three sets of coefficients (minimal second sets in the case of using only coefficients A2 and A1).Since in the current embodiment the voltage command value VH# is set using the extreme voltage Vpl, which has been calculated based on the total sum of the second-order and first-order loss coefficients A2 and A1, it is not necessary to perform a power loss calculation using the polynomial over each of a multitude of voltage candidates, and the data processing load can be reduced. Therefore, by reducing the data processing load, the total sum of the power loss of the rotary electric motor drive apparatus 1000 can be reduced. Design 2

[0110] The controller 400 of the rotary electric motor drive unit 1000 according to embodiment 2 is explained. The explanation for components that are the same as in embodiment 1 is omitted. The basic design of the controller 400 according to the present embodiment is the same as that of embodiment 1; however, embodiment 2 differs from embodiment 1 in that the voltage command value VH# is set such that a resonant voltage, which occurs as a resonance of the rotary electric motor drive unit 1000, is avoided.

[0111] In the case where the voltage command value VH#, set by the aforementioned embodiment 1, is used as is, depending on the selection of a constant inductance and capacitor used for the converter 15 and the selection of an operating point, voltage and current ripple of the converter 15 may occur due to the LC resonance of the converter 15. Therefore, it is desirable to calculate the voltage command value VH# to suppress the LC resonance of the converter 15 while reducing the total power loss of the rotary electric motor drive apparatus 1000 as much as possible.

[0112] In the present embodiment, the voltage command calculation unit 700 then applies a lower limit to the voltage command value VH# set by the loss minimum command calculation unit 802 by means of a resonance avoidance voltage VHLC, which is set in preparation to a greater value than a resonance voltage command value, which is the voltage command value VH# that causes resonance of the rotary electric motor drive apparatus 1000, and sets a value on which the lower limit has been applied as the final voltage command value VH#.

[0113] As in Fig. As shown in Figure 19, the voltage command calculation unit 700 is equipped with a resonance avoidance voltage calculation unit 1220, which calculates the resonance avoidance voltage VHLC. The resonance avoidance voltage calculation unit 1220 calculates the resonance avoidance voltage VHLC according to the current torque command value Tqcom and the current rotational speed ω of the rotary electric machine MG by using characteristic map data in which a relationship between the torque command value Tqcom, the rotational speed ω of the rotary electric machine MG and the resonance avoidance voltage VHLC is predefined.

[0114] The voltage command calculation unit 700 is provided with a maximum value selection unit 1230, which selects a larger one from the voltage command value VH#, calculated by the same procedure as the embodiment 1 mentioned above, and the resonance avoidance voltage VHLC, and outputs the larger one as the final voltage command value VH#.

[0115] According to the above embodiment, the LC resonance can be suppressed while maintaining the effect of embodiment 1 as far as possible. embodiment 3

[0116] The controller 400 of the rotary electric motor drive apparatus 1000 according to embodiment 3 is described below. Explanations for components that are identical to those in embodiments 1 and 2 are omitted. The basic design of the controller 400 according to the present embodiment is the same as that of embodiments 1 and 2; however, embodiment 3 differs from embodiments 1 and 2 in that the voltage command value VH# is set in such a way as to avoid a pre-set avoidance voltage range.

[0117] In the case where the voltage command value VH# calculated in embodiments 1 and 2 is used as is, oscillations in the system voltage VH can occur, depending on the selection of constants for the inductance and capacitor used for the converter 15 and the selection of an operating point, particularly when the system voltage VH is close to the power source voltage Vb. If the voltage command value VH# set by the loss-minimization command calculation unit 802 enters a pre-set avoidance voltage range, the voltage command calculation unit 700 then sets a voltage outside the avoidance voltage range as the final voltage command value VH#.The voltage command calculation unit 700 calculates the avoidance voltage range based on the torque command value Tqcom, the rotational speed ω of the rotary electric machine MG, the energy source voltage Vb, the maximum rotary electric machine output Pmg_max, and the like, using set data as well as pre-set characteristic map data. The voltage command calculation unit 700 calculates the range from the energy source voltage Vb up to a voltage (Vb+ΔVth) that adds a pre-set ΔVth to the energy source voltage Vb as the avoidance voltage range; and sets Vb+ΔVth as the voltage command value VH# in the case where the voltage command value VH# enters the avoidance voltage range. According to this design, the oscillation of the system voltage VH can be suppressed while maintaining the effect of embodiments 1 and 2 as far as possible. [Other embodiments]

[0118] Finally, other embodiments of the present invention will be explained. Each of the embodiments to be explained below is not limited to being used separately, but can be used in combination with the embodiments of other designs, as long as no discrepancy arises. (1) In each of the embodiments mentioned above, the case has been described where the rotary electric machine MG and the inverters IN are provided as two sets, and the controller 400 is configured to adapt to the first and second sets. However, the rotary electric machine MG and the inverters IN can be provided as one set, three sets, or more than three sets. The controller 400 is appropriately configured to adapt to the number of sets. (2) In each of the embodiments mentioned above, the case has been explained where each of the polynomials is a polynomial whose order is second order. However, each of the polynomials can be a polynomial whose order is higher than or equal to third order. (3) In each of the embodiments mentioned above, the case has been explained where the loss minimum command calculation unit 802 calculates the inverter loss coefficient, which is a coefficient of a polynomial in which the system voltage VH is a variable and which represents a power loss characteristic of the inverter IN and the rotary electric machine MG. However, the loss minimum command calculation unit 802 can also calculate the inverter loss coefficient, which is a coefficient of a polynomial in which the system voltage VH is a variable and which represents a power loss characteristic of only the inverter IN.The loss minimum command calculation unit 802 can furthermore calculate the loss coefficient of the rotary electromachine, which is a coefficient of a polynomial in which the system voltage VH is a variable and which represents a power loss characteristic of the rotary electromachine MG, and calculate a total sum of the converter loss coefficient, the inverter loss coefficient and the loss coefficient of the rotary electromachine as the total sum loss coefficient for each order of the polynomials. (4) In each of the embodiments mentioned above, the case has been explained where the loss minimum command calculation unit 802 calculates the first-order, second-order, and third-order converter loss coefficients, the loss coefficients of the first inverter, the loss coefficients of the second inverter, and the total loss coefficients. Since the loss minimum command calculation unit 802 does not calculate every zero-order loss coefficient using the flowchart of Fig. The processing described in section 9 is used, but the minimum loss command calculation unit 802 does not need to calculate the zeroth order of the converter loss coefficient, the first inverter loss coefficient, the second inverter loss coefficient, and the total loss coefficient. (5) In each of the embodiments mentioned above, the case has been explained where the loss minimum command unit 802 calculates the extreme voltage Vpl based on the second-order and first-order total loss coefficients A2, A1, and sets the voltage command value VH# using the extreme voltage Vpl. However, the loss minimum command unit 802 can, for each of a plurality of voltage candidates, calculate the total power loss Ploss using the polynomial that uses the total power loss coefficient representing the total power loss characteristic, as shown in equation (8); and find a voltage for which the total power loss Ploss is minimized, and set the low-loss voltage VHLL.Since in this case the calculation uses only one polynomial, the data processing load can be reduced more easily than in the calculation using the characteristic map data, which represent every power loss. (6) In each of the embodiments mentioned above, the calculation example of the approximation equation and the coefficient for the converter 15, the inverter IN, and the rotary electric machine MG showed only the representative example and can be calculated on the basis of a different method or other variables. In the case where the number of converter 15, the number of inverters IN, and the number of rotary electric machines MG increase, their loss characteristic can be approximated in the same way, a coefficient can be calculated in the same way, and the voltage command value VH# for which the power loss becomes a minimum can be determined according to the method described in Fig. The flowchart shown in section 9 will be set. (7) Under the power loss of a converter, the power loss of an inverter and the power loss of a rotary electric machine, only over some limited losses where the rate of change for changing the system voltage VH is large, the coefficient can be calculated in the case of approximating a power loss by a polynomial, and on the basis of the calculated coefficient, the low-loss voltage VHLL for which a power loss becomes a minimum can be set. (8) In each of the embodiments mentioned above, the case has been described where the converter 15 is provided with the two switching devices Q3, Q4 for the voltage boost chopper and with the two switching devices Q1, Q2 for the voltage drop chopper. However, the circuit configuration may be modified. For example, the converter 15 may be provided with one switching device for the voltage boost chopper and may be provided with one switching device for the voltage drop chopper. (9) In each of the embodiments mentioned above, the case has been explained where the rotary electric motor drive unit 1000 is mounted in the hybrid vehicle. However, the rotary electric motor drive unit 1000 can be a drive power source of a device other than the hybrid vehicle, such as being mounted in an electric vehicle.

[0119] Numerous modifications and alterations of this invention will be apparent to those familiar with the field, without deviating from the scope of protection of this invention, and it should be understood that it is not limited to the illustrative embodiments disclosed herein.< / zusammenfassung>

Claims

Controller (400) for a rotary electric machine drive apparatus (1000), which is provided with a converter (15) that can increase an energy source voltage (Vb) of a DC power source (B) for output to a system voltage line (7, 8), and an inverter (IN) that is provided between the converter (15) and a rotary electric machine (MG) and performs a power conversion between a DC power of the system voltage line (7, 8) and an AC power that drives the rotary electric machine (MG), wherein the controller (400) for the rotary electric machine drive apparatus (1000) comprises: a converter control unit (750) that controls the converter (15) such that a system voltage (VH), which is a DC voltage of the system voltage line (7, 8), approaches a voltage command value (VH#), in the case where the voltage command value (VH#) is greater than the Energy source voltage (Vb) is; and a voltage command calculation unit (700),which calculates the voltage command value (VH#) within a range that is greater than or equal to the energy source voltage (Vb) and less than or equal to an output maximum voltage (Vmax) of the converter (15); wherein the voltage command calculation unit (700) is provided with a required minimum voltage calculation unit (800) which calculates a required minimum voltage (VHL) that is a minimum system voltage required in the case of performing maximum torque / current control of the rotary electric machine (MG), under conditions of the current torque command value and the current speed of the rotary electric machine (MG); and a loss minimum command calculation unit (802) which, in the case where the required minimum voltage (VHL) is greater than or equal to the energy source voltage (Vb) and less than or equal to the output maximum voltage (Vmax) of the converter (15),a converter loss coefficient is calculated, which is a coefficient of a polynomial in which the system voltage (VH) is a variable and represents a power loss characteristic of the converter (15); an inverter loss coefficient is calculated, which is a coefficient of a polynomial in which the system voltage (VH) is a variable and represents a power loss characteristic of the inverter (IN); a sum of the converter loss coefficient and the inverter loss coefficient is calculated for each order of the polynomials; based on the calculated sum loss coefficient for each order, a low-loss voltage (VHLL) is calculated, which is a system voltage (VH) at which a sum power loss of the converter (15) and the inverter (IN) is minimized, within a candidate voltage range.which is greater than or equal to the required minimum voltage (VHL) and less than or equal to the output maximum voltage (Vmax) of the converter (15); and sets the low-loss voltage (VHLL) as the voltage command value (VH#); wherein the minimum loss calculation unit (802), in the case (ST110) where the required minimum voltage (VHL) is greater than or equal to the power source voltage (Vb) and less than or equal to twice the power source voltage (Vb), calculates the total loss coefficient (ST120) of which the maximum order is the second order, and determines (ST130) whether or not the total second-order loss coefficient (A2) is a positive value; in the case of determining (ST130) that the total second-order loss coefficient is a positive value, it calculates an extreme voltage (Vpl) (ST140) which is the system voltage (VH),from which the total power loss of the converter (15) and the inverter (IN) becomes a minimum, based on the total second-order and first-order loss coefficients, and then in the case (ST140) where the extreme voltage (Vpl) is greater than twice the value of the energy source voltage (Vb), sets twice the value of the energy source voltage (Vb) as the low-loss voltage (VHLL) (ST171), or in the case where the extreme voltage (Vpl) is less than the required minimum voltage (VHL), sets the required minimum voltage (VHL) as the low-loss voltage (VHLL) (ST172), or in the case where the extreme voltage (Vpl) is within a range between the required minimum voltage (VHL) and twice the values ​​of the energy source voltage (Vb), sets the extreme voltage (Vpl) as the low-loss voltage (VHLL) (ST173); in the case of determining (ST130),that the total second-order loss coefficient (A2) is not a positive value, sets twice the value of the power source voltage (Vb) as the low-loss voltage (VHLL) (ST174) if (ST160) the extreme voltage (Vp1), which is the system voltage (VH) from which the total power loss becomes a maximum, is less than the average voltage (Vave) of the required minimum voltage (VHL) and twice the value of the power source voltage (Vb), and sets the required minimum voltage (VHL) as the low-loss voltage (VHLL) (ST175) if (ST160) the extreme voltage (Vp1) is not less than the average voltage (Vave) of the required minimum voltage (VHL) and twice the value of the power source voltage (Vb); in the case (ST110, ST165) where the required minimum voltage (VHL) is greater than twice the value of the power source voltage (Vb),the required minimum voltage (VHL) is set as the low-loss voltage (VHLL) (ST176); in the case (ST165) where the required minimum voltage (VHL) is greater than the output maximum voltage (Vmax) of the converter (15), the output maximum voltage (Vmax) of the converter (15) is set as the low-loss voltage (VHLL). Controller (400) for the rotary electric machine drive apparatus (1000) according to claim 1, wherein the loss minimum command calculation unit (802) calculates the converter loss coefficient according to the current energy source voltage (Vb) and a current output of the converter (15), over each order, by using a relationship characteristic in which a relationship between the energy source voltage (Vb), the output of the converter (15) and the converter loss coefficient is pre-established; and calculates the inverter loss coefficient according to the current torque command value and the current speed of the rotary electric machine (MG), over each order, by using a relationship characteristic in which a relationship between a delivery torque of the rotary electric machine (MG), the speed of the rotary electric machine (MG) and the inverter loss coefficient is pre-established. Controller (400) for the rotary electric machine drive apparatus (1000) according to one of claims 1 to 2, wherein a plurality of sets of rotary electric machine (MG) and inverter (IN) are provided, wherein the required minimum voltage calculation unit (800) calculates an individual required voltage which is the minimum system voltage required in the case of performing a maximum torque / current control of the rotary electric machine (MG), via each of the rotary electric machines (MG), under conditions of the current torque command value and the current speed of the rotary electric machine (MG);and sets a maximum value from the individual required voltages in each of the rotary electric machines (MG) as the required minimum voltage (VHL), wherein the loss minimum command calculation unit (802) calculates the inverter loss coefficient over each of the inverters (IN) and calculates the total sum of the converter loss coefficient and the inverter loss coefficient of each of the inverters (IN) for each order of the polynomial.; Controller (400) for the rotary electric machine drive apparatus (1000) according to one of claims 1 to 3, wherein the loss minimum command calculation unit (802) calculates the inverter loss coefficient, which is a coefficient of a polynomial in which the system voltage (VH) is a variable and which represents a loss power characteristic of the rotary electric machine (MG) in addition to the inverter (IN). Controller (400) for the rotary electric machine drive apparatus (1000) according to one of claims 1 to 3, wherein the loss minimum instruction calculation unit (802) calculates a loss coefficient of the rotary electric machine (MG) which is a coefficient of a polynomial in which the system voltage (VH) is a variable and which represents a power loss characteristic of the rotary electric machine (MG), and calculates a total sum of the converter loss coefficient, the inverter loss coefficient and the loss coefficient of the rotary electric machine (MG) as the total sum loss coefficient for each order of the polynomials. Controller (400) for the rotary electric motor drive apparatus (1000) according to one of claims 1 to 5, wherein each of the polynomials is a polynomial whose order is less than or equal to the second order. Controller (400) for the rotary electric motor drive apparatus (1000) according to one of claims 1 to 6, wherein the coefficient in each of the polynomials is pre-set using the method of least squares. Controller (400) for the rotary electric machine drive apparatus (1000) according to one of claims 1 to 7, wherein the coefficient in each of the polynomials is pre-set on the basis of respective power loss data for a range of the system voltage (VH) between the energy source voltage (Vb) and twice the value of the energy source voltage (Vb). Controller (400) for the rotary electric motor drive apparatus (1000) according to one of claims 1 to 8, wherein the voltage command calculation unit (700) applies a lower limit to the voltage command value (VH#) set by the loss minimum command calculation unit (802) by means of a resonance avoidance voltage, which is preparatoryly set to a larger value than a resonance voltage command value (VH#), which is the voltage command value (VH#) that causes a resonance of the rotary electric motor drive apparatus (1000), and a value on which the lower limit has been applied, as the final voltage command value (VH#). Controller (400) for the rotary electric motor drive apparatus (1000) according to one of claims 1 to 9, wherein the voltage command calculation unit (700) in the case where the voltage command value (VH#) set by the loss minimum command calculation unit (802) enters a pre-set avoidance voltage range, sets a voltage outside the avoidance voltage range as the final voltage command value (VH#).

Images