JP7436988B2 - Multiphase rotating machine control device - Google Patents

Description

The present invention relates to a control device for a multiphase rotating machine.

BACKGROUND ART Conventionally, a control device is known that controls the current flowing from a power converter to two systems of polyphase windings.

For example, Patent Document 1 discloses a configuration for controlling the sum and difference of currents in two systems in a control device for a three-phase rotating machine having two sets of three-phase windings that are magnetically coupled to each other. In this configuration, the sum and difference of the two systems of dq-axis actual currents Id and Iq are fed back to the sum and difference of current command values Id ^* and Iq ^* , respectively. Based on the deviation, the sum and difference of the dq-axis voltages Vd and Vq are calculated through the motor inverse model and the controller.

Patent No. 5556845

In the control device of Patent Document 1, one arithmetic unit common to the two systems adds and subtracts the dq-axis actual currents (ie, feedback currents) based on the three-phase current detection values of each system. On the other hand, in a configuration in which two systems of arithmetic devices corresponding to power converters of each system are provided separately, it is necessary to acquire current information of other systems through communication. When controlling the currents of two systems of polyphase windings, the communication load between the arithmetic units is high and the control period cannot be shortened, so there is a problem that control performance deteriorates. Furthermore, if communication between arithmetic devices is made faster in order to shorten the calculation period, noise caused by the communication to other circuits increases, and communication becomes more susceptible to noise.

The present invention was created in view of the above points, and its purpose is to reduce the communication load associated with acquiring current information of other systems in a configuration having two systems of arithmetic units. The purpose is to provide a control device.

A unit of a group of components related to energization of polyphase windings is defined as a "system." A control device for a polyphase rotating machine according to the present invention is a control device that controls the drive of a polyphase rotating machine (80) having two systems of polyphase windings (80A, 80B). This control device for a multiphase rotating machine includes two systems of power converters (60A, 60A) and two systems of arithmetic units (40A, 40B).

The two-system power converter can individually energize the two-system polyphase windings. The two systems of arithmetic units are provided independently from each other and control the current flowing from the power converter to the polyphase winding based on the current flowing to the polyphase winding detected by the current detector (70A, 70B). Perform calculations. "Two systems of arithmetic units are provided independently of each other" means that the hardware configurations are physically separate.

The arithmetic unit of each system estimates the current of the other system based on the current flowing through the polyphase winding of the system and the output voltage of the system and other systems, and calculates the current between systems based on the estimated current of the other system. Suppress the effects of interference. "Suppressing the influence of interference between systems" is not limited to control that directly calculates a non-interference control amount, but includes any control that acts in the direction of suppressing the influence of interference.

In the present invention, by estimating the actual current of the other system using the arithmetic unit of the own system instead of acquiring it through intersystem communication, it is possible to reduce the communication load related to acquiring the current information of the other system. In addition, since only the current information of the own system is used, it has the effect of reducing the failure rate due to the same cause in the two systems.

Preferably, the arithmetic unit of each system estimates the current of the other system using an observer (570A, 570B). Control is stabilized by performing convergence calculations using the observer. Preferably, the arithmetic device of each system performs non-interference control using the estimated current of the other system. More preferably, the arithmetic device of each system controls the current flowing through the polyphase winding of the own system by controlling the sum and difference between the current of the own system and the estimated current of the other system.

1 is a schematic configuration diagram of an electric power steering device to which a control device (ECU) for a multiphase rotating machine according to each embodiment is applied. An axial cross-sectional view of a two-system electromechanical integrated motor. FIG. 3 is a sectional view taken along line III-III in FIG. 2. FIG. 2 is a schematic diagram showing the configuration of a two-system multi-phase coaxial motor. An overall configuration diagram of a motor drive system. Control block diagram of two systems of arithmetic units (microcomputers). FIG. 3 is a block diagram of a voltage command calculation section according to the first embodiment. Block diagrams of observers for (a) system A and (b) system B. FIGS. 3A and 3B are diagrams illustrating an example and another example of the gain characteristics of an observer according to the rotation speed. FIGS. Flowcharts of (a) one example and (b) another example of switching execution or non-implementation of other system current estimation. A time chart showing the relationship between the intersystem communication cycle and the current control calculation cycle. FIG. 3 is a block diagram of a voltage command calculation section according to a second embodiment. FIG. 7 is a block diagram of a voltage command calculation section according to a third embodiment.

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a plurality of embodiments of a control device for a multiphase rotating machine according to the present invention will be described based on the drawings. In each embodiment, the ECU as a "control device for a multiphase rotating machine" is applied to an electric power steering device of a vehicle, and controls energization of a motor (that is, a multiphase rotating machine) that generates a steering assist torque. Substantially the same configurations in the plurality of embodiments are designated by the same reference numerals, and description thereof will be omitted. Furthermore, the following first to third embodiments are collectively referred to as "this embodiment". First, matters common to each embodiment will be described with reference to FIGS. 1 to 6.

[Configuration of electric power steering device]
FIG. 1 shows the overall configuration of a steering system 99 including an electric power steering device 90. Although FIG. 1 shows a "mechanically and electrically integrated" motor 800 in which the ECU 10 is integrally configured on one side of the motor 80 in the axial direction, a "mechanically and electrically integrated" motor 800 in which the ECU 10 and the motor 80 are connected by a harness is illustrated. This embodiment is similarly applicable to the "separate mechanical and electrical system". Further, although the electric power steering device 90 shown in FIG. 1 is a column assist type electric power steering device, the present embodiment is similarly applicable to a rack assist type electric power steering device.

The steering system 99 includes a handle 91, a steering shaft 92, a pinion gear 96, a rack shaft 97, wheels 98, an electric power steering device 90, and the like. A steering shaft 92 is connected to the handle 91. A pinion gear 96 provided at the tip of the steering shaft 92 meshes with a rack shaft 97. A pair of wheels 98 are provided at both ends of the rack shaft 97 via tie rods or the like. When the driver rotates the handle 91, a steering shaft 92 connected to the handle 91 rotates. The rotational motion of the steering shaft 92 is converted into linear motion of the rack shaft 97 by the pinion gear 96, and the pair of wheels 98 are steered to an angle corresponding to the amount of displacement of the rack shaft 97.

The electric power steering device 90 includes steering torque sensors 93A, 93B, an ECU 10, a motor 80, a reduction gear 94, and the like. The steering torque sensors 93A and 93B are redundantly provided in the middle of the steering shaft 92 and detect the driver's steering torques TsA and TsB. The ECU 10 controls the drive of the motor 80 so that the motor 80 generates a desired assist torque based on the steering torques TsA and TsB. The assist torque output by the motor 80 is transmitted to the steering shaft 92 via a reduction gear 94.

The configuration of an electromechanical integrated motor 800 in which an ECU 10 is integrally configured on one side of the motor 80 in the axial direction will be described with reference to FIGS. 2 and 3. In the form shown in FIG. 2, the ECU 10 is arranged coaxially with the axis Ax of the shaft 87 on the side opposite to the output side of the motor 80. Note that in other embodiments, the ECU 10 may be configured integrally with the motor 80 on the output side of the motor 80. The motor 80 is a three-phase brushless motor and includes a stator 840, a rotor 860, and a housing 830 that accommodates them.

Stator 840 has a stator core 844 fixed to housing 830 and two sets of three-phase windings 80A and 80B wound around each slot 848 of stator core 844. Hereinafter, a unit of a group of components related to energization of the three-phase windings 80A and 80B will be defined as a "system." This embodiment is a control device that controls the drive of a motor 80 having two systems of three-phase windings 80A and 80B. Lead wires 851, 853, and 855 extend from the three-phase winding 80A of the A system. Lead wires 852, 854, and 856 extend from the three-phase winding 80B of the B system.

The rotor 860 includes a shaft 87 supported by a rear bearing 835 and a front bearing 836, and a rotor core 864 into which the shaft 87 is fitted. The rotor 860 is provided inside the stator 840 and is rotatable relative to the stator 840. The motor 80 of this embodiment is an embedded magnet type synchronous rotating machine (so-called IPMSM) in which a plurality of magnets 865 are embedded in the outer periphery of a rotor core 864. A permanent magnet 88 for rotation angle detection is provided at one end of the shaft 87.

The housing 830 has a bottomed cylindrical case 834 including a rear frame end 837 and a front frame end 838 provided at one end of the case 834. The case 834 and the front frame end 838 are fastened to each other with bolts or the like. The lead wires 851, 852, etc. of the three-phase windings 80A, 80B are inserted through the lead wire insertion hole 839 of the rear frame end 837, extend toward the ECU 10, and are connected to the board 230.

The ECU 10 includes a cover 21, a heat sink 22 fixed to the cover 21, a board 230 fixed to the heat sink 22, and various electronic components mounted on the board 230. The cover 21 protects electronic components from external impacts and prevents dust, water, etc. from entering the ECU 10. The cover 21 includes a connector section 214 for external connection of a power supply cable or a signal cable from the outside, and a cover section 213. Power supply terminals 215 and 216 of the external connection connector section 214 are connected to the board 230 via a path not shown.

The board 230 is, for example, a printed circuit board, is provided at a position facing the rear frame end 837, and is fixed to the heat sink 22. On the board 230, electronic components for two systems are provided independently for each system. In this embodiment, there is one substrate 230, but in other embodiments, two or more substrates may be provided. Of the two main surfaces of the board 230, the surface facing the rear frame end 837 is defined as a motor surface 237, and the surface opposite thereto, that is, the surface facing the heat sink 22 is defined as a cover surface 238.

A plurality of switching elements 24A, 24B, rotation angle sensors 25A, 25B, custom ICs 26A, 26B, etc. are mounted on the motor surface 237. In this embodiment, the plurality of switching elements 24A and 24B are six for each system, and constitute three-phase upper and lower arms of the motor drive circuit. The redundantly provided rotation angle sensors 25A and 25B are arranged to face a permanent magnet 88 provided at the tip of the shaft 87. The custom ICs 26A, 26B and the microcomputers 40A, 40B have a control circuit for the ECU 10.

On the cover surface 238, microcomputers 40A, 40B, capacitors 28A, 28B, inductors 27A, 27B, etc. are mounted. In particular, the A-system microcomputer 40A and the B-system microcomputer 40B are arranged at a predetermined interval on a cover surface 238 that is a surface on the same side of the same board 230. The capacitors 28A and 28B smooth the power input from the power supply, and also prevent noise from flowing out due to switching operations of the switching elements 24A and 24B. Inductors 27A and 27B constitute a filter circuit together with capacitors 28A and 28B.

As shown in FIG. 4, the motor 80 that is controlled by the ECU 10 is a double-winding brushless motor in which two systems of three-phase windings 80A and 80B are coaxially provided. The three-phase windings 80A and 80B have the same electrical characteristics, and have an electrical angle of 30[deg] to each other on the common stator 840.Generally speaking, (30±60×n)[deg] (n is an integer) They are staggered.

The three-phase windings 80A and 80B of the two systems are magnetically coupled to each other, and mutual inductance occurs between the systems. Therefore, there is a need for "non-interference control" that eliminates interference from the voltage generated in the three-phase windings of the own system due to the current flowing in the three-phase windings of other systems. "Non-interference" in this specification means non-interference with inter-system interference, rather than non-interference with dq interference within the same system.

[Configuration of ECU and microcomputer]
The overall configuration of the motor drive system will be described with reference to FIG. 5. The ECU 10 includes an A-system microcomputer 40A and a B-system microcomputer 40B as a "two-system arithmetic unit," and an A-system inverter 60A and a B-system inverter 60B as a "two-system power converter." In symbols representing physical quantities such as current, voltage, and electrical angle, the suffix "A" indicates the physical quantity of the A system, and the suffix "B" indicates the physical quantity of the B system.

In the system example of FIG. 5, inverters 60A and 60B of each system are connected to two batteries 11A and 11B provided for each system. Note that one battery common to the two systems may be provided. In each inverter 60A, 60B, six switching elements such as MOSFETs are bridge-connected between a power supply line and a ground line. Smoothing capacitors 16A and 16B are provided at the input section of each inverter 60A and 60B.

Each inverter 60A, 60B performs a switching operation by driving signals DrA, DrB from the microcomputer 40A, 40B of its own system, converts the DC power of the batteries 11A, 11B, and supplies it to the three-phase windings 80A, 80B. In this way, the inverters 60A and 60B can individually energize the two three-phase windings 80A and 80B.

Current detectors 70A, 70B detect three-phase currents iuvwA, iuvwB flowing through three-phase windings 80A, 80B of each system, and output them to microcomputers 40A, 40B. The rotation angle sensors 25A, 25B redundantly detect the electrical angles θA, θB of the motor 80 and output them to the microcomputers 40A, 40B. Here, the electrical angles θA and θB have a phase difference of 30[deg]. Furthermore, the steering torques TsA and TsB detected by the steering torque sensors 93A and 93B are input to the microcomputers 40A and 40B.

The two microcomputers 40A and 40B have physically separate hardware configurations and are provided independently from each other. In other words, the configuration is not such that multiple cores provided within one microcomputer function together. Each of the microcomputers 40A and 40B includes a CPU (not shown), a ROM, a RAM, an I/O, a bus line connecting these components, and the like. The microcomputers 40A and 40B perform software processing by executing a program stored in advance in a physical memory device such as a ROM (that is, a readable non-temporary tangible recording medium) on a CPU, and hardware processing using a dedicated electronic circuit. Execute control by processing.

Steering torques TsA and TsB are input to each microcomputer 40A and 40B. The A-system microcomputer 40A performs calculations to control the current flowing from the first inverter 60A to the three-phase winding 80A by feedback control of the current iuvwA flowing through the A-system three-phase winding 80A. The B-system microcomputer 40B performs calculations to control the current flowing from the second inverter 60B to the three-phase winding 80B by feedback control of the current iuvwB flowing through the three-phase winding 80B of the B-system. In other words, each microcomputer 40A, 40B causes the current to flow from the inverters 60A, 60B to the three-phase windings 80A, 80B based on the currents iuvwA, iuvwB flowing to the three-phase windings 80A, 80B detected by the current detectors 70A, 70B. perform calculations to control the

The A-system microcomputer 40A and the B-system microcomputer 40B communicate information such as a current command value, a current limit value, and abnormality information indicating the presence or absence of an abnormality in at least one direction, preferably in both directions, through intersystem communication. Inter-system communication between the microcomputers 40A and 40B may be performed via a vehicle network such as CAN, or may be performed within the ECU 10 by serial communication or CAN communication. In particular, in this embodiment, it is assumed that communication using a vehicle network with a relatively long communication cycle is used.

Refer now to FIG. Components of the A-system microcomputer 40A are suffixed with "A", and components of the B-system microcomputer 40B are suffixed with "B". The A system microcomputer 40A includes a torque command calculation section 41A, a current command value calculation section 42A, a current limit value calculation section 43A, a failsafe section 44A, a voltage command value calculation section 50A, a dq/3 phase conversion section 46A, and a 3 phase/dq It includes a conversion section 47A and a rotation speed calculation section 48A. The B system microcomputer 40B includes a torque command calculation section 41B, a current command value calculation section 42B, a current limit value calculation section 43B, a failsafe section 44B, a voltage command value calculation section 50B, a dq/3 phase conversion section 46B, and a 3 phase/dq It includes a conversion section 47B and a rotation speed calculation section 48B. In addition, in FIG. 6, the inverter is written as "INV".

Since the configurations of the A-system microcomputer 40A and the B-system microcomputer 40B are substantially the same, parts that can be explained by one of them will be described using the reference numerals of the components of the A-system microcomputer 40A as a representative. The B-system microcomputer 40B is interpreted in the same manner by replacing the "A" at the end of the code and symbol with "B". The torque command calculation unit 41A calculates a torque command trqA ^* based on the steering torque TsA.

Hereinafter, the d-axis component and the q-axis component of the dq-axis current and voltage vectors will not be described individually, but will be expressed as vectors as shown in equations (1.1) to (1.6). iA ^* and iB ^* are current command values for the A system and B system, respectively. iA and iB are currents flowing through the three-phase windings 80A and 80B of the A system and B system, respectively. vA and vB are voltage command values (in other words, "output voltages") of the A system and the B system, respectively. Regarding current, add " ^* " to the command value to distinguish it from the actual current. As for voltage, there is only one type of voltage command value, so " ^* " is omitted. Current "i" and voltage "v" are shown in italics in figures and formulas, and in normal typeface in the specification.

The current command value calculation unit 42A receives the torque command trqA ^* of the own system and various information of the own system and other systems, and calculates the current command value iA ^* based on the information. As one of the various pieces of information, the current limit value calculation unit 43A calculates a current limit value i ^* _limA, which is the upper limit of the current command value, based on, for example, the temperature of the inverter 60A and the temperature of the three-phase winding 80A. The current limit value i ^* _limA may be used as information in place of the current command value.

Further, as one of various types of information, the failsafe unit 44A generates an abnormality signal diagA indicating the presence or absence of an abnormality. In this embodiment, "abnormality" means an abnormality to the extent that current information for non-interference control cannot be normally acquired. The failsafe unit 44A determines the control state in which an abnormality in that sense may occur. Furthermore, abnormalities such as communication interruption may occur in intersystem communication. Even in the event of such a communication abnormality, the failsafe unit 44A of the system in which the abnormality has been detected generates an abnormality signal diagA.

The torque command trqA ^* , the current limit value i ^* _limA, and the abnormal signal diagA of the A system are input to the current command value calculation unit 42A of the own system, and the current command value of the B system, which is another system, is calculated by intersystem communication. 42B. Similarly, the torque command trqB ^* , current limit value i ^* _limB, and abnormal signal diagB of the B system are input to the current command value calculation unit 42B of the own system, and the current of the A system, which is another system, is It is transmitted to the command value calculation section 42A.

Thereby, the current command value calculation unit 42A calculates the current command value iA ^* based on the input signal, and outputs it to the voltage command value calculation unit 50A of its own system. Furthermore, the current command value iA ^* of the A system is transmitted to the voltage command value calculation unit 50B of the B system through intersystem communication. Note that, as will be described later with reference to FIG. 11, the current command value is not limited to bidirectional communication, but the current command value is transmitted only in one direction from one master system to the other slave system, and the current command value is transmitted in the slave system. May be updated.

The voltage command value calculation unit 50A calculates the dq-axis voltage command value vA by current feedback (“current FB” in the figure) control that causes the actual current iA to follow the current command value iA ^* . Further, the voltage command value calculation unit 50A performs "non-interference control" in which the voltage generated in the three-phase winding of the own system is made non-interfering with the current flowing in the three-phase winding of the other system. The non-interference control will be described later.

The dq/3-phase conversion unit 46A performs coordinate conversion from the dq-axis voltage command value vA to the 3-phase voltage command value vuvwA. Inverter 60A is driven by drive signal DrA generated based on three-phase voltage command value vuvwA. The 3-phase/dq conversion unit 47A coordinate-converts the 3-phase current iuvwA into a dq-axis current iA, and feeds it back to the voltage command value calculation unit 50A. The rotation speed calculation section 48A calculates the rotation speed ωA which is proportional to the angular velocity obtained by time-differentiating the electrical angle θA, and outputs it to the voltage command value calculation section 50A.

By the way, the control device for a double-winding motor disclosed in Patent Document 1 (Japanese Patent No. 5,556,845) controls the sum and difference of currents in each system using one arithmetic unit common to the two systems. Further, the control device for a double winding motor disclosed in Japanese Patent No. 6497106 also performs non-interference control using one arithmetic unit common to the two systems. Therefore, there is no need to consider the communication load associated with communication of current information between two arithmetic units for each system. On the other hand, in the configuration in which information is communicated between two systems of microcomputers 40A and 40B as in this embodiment, a problem arises in that the communication load associated with acquiring current information of other systems becomes high.

Therefore, the ECU 10 of this embodiment is configured to have two microcomputers 40A and 40B, and aims to reduce the communication load associated with acquiring current information of other systems. The specific configuration of the voltage command value calculation units 50A and 50B for this purpose will be explained for each embodiment. As for the code of the voltage command value calculation unit of each embodiment, the number of the embodiment is attached to the third digit following "50".

(First embodiment)
FIG. 7 shows the configuration of the A system voltage command value calculation unit 501A in the first embodiment. Since the configurations of the A system and the B system are basically the same, illustration of the voltage command value calculation unit of the B system is omitted. In the voltage command value calculation section of the B system, the suffix "A" of the symbols of blocks, currents, etc. in the A system voltage command value calculation section 501A is replaced with "B".

The voltage command value calculation unit 501A of the first embodiment includes a dq/sum-difference conversion unit 51A, a dq/sum-difference conversion unit 52A, a current controller 53A, a sum-difference/dq conversion unit 54A, and an observer (state estimator). 570A, and the sum and difference of the two currents are controlled by a current controller 53A. Here, symbols of vectors indicating the sum and difference of currents of the two systems and the sum and difference of voltages are defined by equations (2.1) to (2.6). Add the letter "S" for sum and "D" for difference at the end of the symbol.

The dq/sum-difference conversion unit 51A converts the A-system current command value iA ^* and the B-system current command value iB ^* into a sum iS ^* and a difference iD ^* of current command values. The dq/sum-difference conversion unit 52A converts the A-system actual current iA and the B-system current estimated value iB_estA estimated and calculated by the observer 570A into a sum iS and a difference iD of currents.

The current controller 53A receives the deviation between the sum of current command values iS ^* and the sum of currents iS, and the deviation between the difference between current command values iD ^* and the current difference iD, and brings these deviations closer to 0. The sum uS and difference uD of voltage command values are calculated as follows. The sum-difference/dq converter 54A converts the sum uS and difference uD of voltage command values into an A-system voltage command value vA and outputs it.

The sum uS and difference uD of the voltage command values output by the current controller 53A are input to the mathematical model 57A of the double-wound motor of the observer 570A. Furthermore, the A system current iA is input to the observer 570A. Based on these inputs, the observer 570A estimates the B-system current estimated value iB_estA, that is, the current of other systems from the viewpoint of the A-system. The detailed configuration of the observer 570A will be described later.

In this way, the A-system microcomputer 40A estimates the current iB of the B-system, which is another system, based on the current iA flowing through the three-phase winding 80A of the own system, and the output voltages of the own system and the other systems. The microcomputer 40A then suppresses the influence of interference between systems based on the estimated current iB_estA of other systems. Typically, the microcomputer 40A of this embodiment performs non-interference control using the estimated current iB_estA of the other system.

Next, the theory of other system current estimation will be explained. In general, voltage equations for double-wound motors are expressed by equations (3.1) to (3.4). In the formula, R is the resistance of the three-phase windings 80A and 80B, Ld and Lq are the dq-axis self-inductance, Md and Mq are the dq-axis mutual inductance, ω is the rotation speed, φ is the back electromotive force constant, and s is the Laplace operator. It is. The term including the mutual inductances Md and Mq is an interference term that occurs in the three-phase winding of the own system due to the current flowing in the three-phase winding of the other system.

Equations (3.1) to (3.4) are expressed in state space using the state equation of equation (4). The state vector x has elements of the d-axis and q-axis currents (idA, iqA, idB, iqB) of each system according to equation (5.1). The input vector u has elements of the sum and difference (vdS, vqS, vdD, vqD) of the d-axis and q-axis voltage command values of the two systems according to equation (5.2).

The A matrix and the B matrix are expressed as square matrices with 4 rows and 4 columns according to equations (6.1) and (6.2). Furthermore, the A matrix is represented by dividing into blocks into A11, A12, A21, and A22 matrices each having 2 rows and 2 columns, and the B matrix is represented by dividing into blocks into B11, B12, B21, and B22 matrices each having 2 rows and 2 columns. expressed.

The observer 570A calculates the estimated B-system current value iB_estA by finding a solution to the differential equation (7) for the estimated B-system current value iB_estA. In the formula, the differential value is indicated by the symbol (d/dt).

The first term (underlined by a dashed-dotted line) of equation (7) is the sum and difference uS, uD of the A system current iA, the B system current estimated value iB_estA, and the voltage command value in the mathematical model 57A of the double winding motor. is calculated using state equation (4).

The second term (double line underlined part) of Equation (7) is a term that feeds back the error of the estimated A-system current value. This is to bring the estimated value closer to the true value by performing a convergence calculation on the deviation between the differential value of the estimated current value iA_estA and the differential value of the actual current iA. This convergence operation is why the block "570A" is called the "observer (state estimator)".

Here, when the differential value of the current estimated value iA_estA is expressed using the A system current iA, the B system current estimated value iB_estA, and the sum and difference uS, uD of the voltage command value, equation (7) becomes equation (8 ) can be rewritten as The underlined double-dashed line in the second term of equation (8) corresponds to the differential value of the current estimated value iA_estA.

FIG. 8(a) shows the configuration of an observer 570A that realizes the estimation calculation of the B-system current. Motor constants R, L, and M of equations (3.1) to (3.4) are stored inside the mathematical model 57A of the double-wound motor. The mathematical model 57A of the double winding motor receives the sum and difference uS, uD of the A system current iA and voltage command value, and outputs the estimated B system current value iB_estA. The differentiator 58 calculates the differential value of the deviation between the A system current iA and the estimated current value iA_estA. A gain multiplier 59A multiplies the output of the differentiator 58 by a gain G and feeds it back to the mathematical model 57A of the double winding motor.

For reference, FIG. 8(b) shows the configuration of the B-system observer 570B. Like the A-system observer 570A, the B-system observer 570B estimates the estimated calculated value iA_estB of the A-system using the double-winding motor mathematical model 57B, the differentiator 58B, and the gain multiplier 59B. In this way, the microcomputers 40A and 40B of the first embodiment simultaneously perform non-interference control and current control using the observers 570A and 570B.

Referring to FIG. 9, the gain G of the observer 570A will be supplemented. The gain G of the observer 570A may be defined by a constant a or a unit matrix as shown by the diagonal matrix in equation (9.1). Alternatively, as shown in equation (9.2), the gain G of the observer 570A may be variably set according to the rotation speed ω of the motor 80. k is a proportionality coefficient.

In the case of the characteristic expressed by equation (9.2), the relationship between the rotational speed ω and the gain G is shown in FIG. 9(a). Furthermore, as shown in FIG. 9(b), a dead zone is provided in a region where the rotational speed ω is around 0, and an upper limit of the absolute value of the gain G is set in a region where the absolute value of the rotational speed ω is a predetermined value or more. It's okay.

Next, with reference to the flowchart of FIG. 10, switching between execution and non-implementation of other system current estimation will be described. In the flowchart description, the symbol "S" means step. As described above with reference to FIG. 6, the microcomputers 40A and 40B of each system bidirectionally communicate abnormality information of each system through intersystem communication.

In S11 of FIG. 10A, the microcomputers 40A and 40B determine whether another system is abnormal or is currently detecting an abnormality. If YES in S11, the microcomputers 40A and 40B stop estimating other system currents in S13. The A-system microcomputer 40A sets the B-system current estimated value iB_estA=0, and the B-system microcomputer 40B sets the A-system current estimated value iA_estB=0. If NO in S11, the microcomputers 40A and 40B perform estimation calculation in S14. That is, the microcomputers 40A and 40B of each system perform estimation calculation of the other system current only when the other system is normal. Note that when restarting the estimation calculation after stopping the estimation, the estimation calculation may be performed using the previous value as the initial value, or the estimation calculation may be performed using 0 as the initial value.

In the example shown in FIG. 10(b), in S12, the microcomputers 40A and 40B determine whether the other systems are stopped, that is, no current flows. The processing of S13 when YES in S12 and S14 when NO in S12 is the same as that in FIG. 10(a). Note that in a configuration in which a system in which an abnormality is detected is stopped, the determination step of S12 is substantially the same as S11. In this way, the microcontrollers 40A and 40B of each system perform calculations to estimate the currents of other systems only when the other systems are normal, thereby avoiding meaningless estimations or estimations that lead to failure propagation, and improving calculation efficiency and reliability. can improve sex.

Next, with reference to the time chart of FIG. 11, the relationship between the intersystem communication cycle and the current control calculation cycle by the microcomputers 40A and 40B of each system will be described. The communication cycle, which is a "long cycle," is illustrated as a cycle five times as long as the current control calculation cycle, which is a "short cycle." That is, the current control calculation for each system is performed, for example, five times while the inter-system communication is performed once. However, the specific period ratio is not limited to this.

The microcomputers 40A and 40B of each system calculate current limit values I ^* _limA and I ^* _limB based on signals input to each microcomputer 40A and 40B at a cycle longer than the current control calculation cycle, and at least one Send the current limit value in the direction. In addition, the microcomputers 40A and 40B of each system calculate abnormality signals diag1 and diag2 indicating the presence or absence of an abnormality using signals input to each microcontroller 40A and 40B at a cycle longer than the current control calculation cycle, and communicate between the systems. transmits an abnormality signal in at least one direction.

In this embodiment, the communication load is reduced by estimating the current of other systems in the own system without acquiring it through inter-system communication, while the current limit value and abnormal signal of other systems must be acquired through inter-system communication. I can't do it. Therefore, the communication load can be reduced by performing the calculation and communication of the current limit value and the abnormal signal in a cycle longer than the cycle of the current control calculation.

Further, in calculating the current command value, for example, the A-system microcomputer 40A operates as a master, and the B-system microcomputer 40B operates as a slave. In the configuration example shown in FIG. 11, the master calculates a current command value and communicates it to the slave. The slave that receives the current command value from the master updates the current command value. Current control calculations by the master and slave are performed at a cycle shorter than the communication cycle of the current command value in inter-system communication. As another configuration example, the slave may calculate the current command correction coefficient after receiving the current command value from the master, and calculate the slave's current command value by multiplying the master's current command value by the current command correction coefficient. .

(Effects of the first embodiment)
(1) The ECU 10 of this embodiment, including the following second and third embodiments, estimates the actual current of the other system by using the calculation device of the own system instead of acquiring the actual current of the other system through intersystem communication. Communication load related to information acquisition can be reduced. In addition, since only the current information of the own system is used, it has the effect of reducing the failure rate due to the same cause in the two systems.

(2) The microcomputers 40A and 40B of each system estimate the current of other systems using the observers 570A and 570B. Control is stabilized by performing convergence calculations using the observers 570A and 570B.

(3) The microcontrollers 40A and 40B of each system control the current flowing through the three-phase windings 80A and 80B of the own system by controlling the sum and difference between the current of the own system and the estimated current of other systems. . As described in Patent Document 1, by controlling the sum and difference of currents in a two-system configuration, torque ripple due to high frequency components can be suppressed and thermal characteristics can be improved.

(Second embodiment)
FIG. 12 shows the configuration of a voltage command value calculation section 502A in the second embodiment. The voltage command value calculation unit 502A controls the current of each system individually in the current controller 55A, rather than the sum and difference of the currents of the two systems. In other words, the current controller 55A receives input of the deviation between the A system current command value iA ^* and the A system current iA, and the deviation between the B system current command value iB ^* and the B system current estimated value iB_estA. The A system voltage command value vA and the B system voltage command value vB are calculated so that the voltage command value vA and the B system voltage command value vB are made close to zero. Here, the B system current estimated value iB_estA is a value estimated and calculated by the observer 570A.

The two systems of voltage command values vA and vB output by the current controller 55A are converted into the sum uS and difference uD of voltage command values by the dq/sum-difference conversion unit 56A, and are converted into the sum uS and difference uD of the voltage command values, and the mathematical model of the double winding motor of the observer 570A is used. 57A. The configuration of the observer 570A is similar to that of the first embodiment. Even with such a configuration, the same effects as in the first embodiment can be obtained.

(Third embodiment)
FIG. 13 shows the configuration of a voltage command value calculation section 503A in the third embodiment. Unlike the first embodiment, the third embodiment does not have an observer function. That is, in the mathematical model 57A of the double-winding motor, no convergence calculation is performed on the deviation between the differential value of the A-system current iA and the differential value of the estimated current value iA_estA.

In the third embodiment, the calculation of the current estimated value iB_estA of other systems is performed in a feedforward manner, and even if the estimated value deviates from the actual value, it is used as is. Except for this point, the same effects as the above embodiment can be obtained.

(Other embodiments)
(a) In the above embodiment, the microcomputers 40A and 40B perform deinterference calculation using the estimated currents of other systems. In the well-known technique of non-interference calculation, for example, as shown in FIG. 6 of Japanese Patent No. 6,497,106, in a block diagram based on a motor model, Ms terms and Mω terms containing information of other systems are input to the own system. It is an operation. However, the arithmetic device of the present invention is not limited to the control that directly calculates the non-interference control amount on the block diagram, but may be any one that performs control that acts at least in the direction of suppressing the influence of interference.

(b) This embodiment assumes a configuration in which two systems of microcomputers 40A and 40B are provided individually and communicate information with each other. However, in addition to the two microcomputers, another arithmetic device for monitoring or the like may be provided separately. Moreover, the torque sensor and the rotation angle sensor are not limited to the configuration in which they are provided redundantly, but one may be provided in common to the two systems.

(c) The number of phases of the multiphase rotating machine is not limited to three phases, but may be four or more phases. In rotating machines other than three-phase, the "three-phase winding" in the above embodiment is generalized to "multiphase winding." The control device for a multiphase rotating machine of the present invention is not limited to a control device for a steering assist motor of an electric power steering device, but may be applied as a control device for a motor or a generator for other purposes.

(d) As a communication standard for inter-system communication, in addition to CAN and serial communication (UART), LIN, FlexRay, Ethernet (registered trademark), etc. may be used.

As described above, the present invention is not limited to these embodiments, and can be implemented in various forms without departing from the spirit thereof.

10...ECU (control unit),
40A, 40B...Microcomputer (computing device),
60A, 60B... Inverter (power converter),
80...Motor (multiphase rotating machine),
80A, 80B...3-phase winding (polyphase winding).

Claims (10)

If we define a system as a unit of a group of components related to energization of polyphase windings, then it is a control device that controls the drive of a polyphase rotating machine (80) having two systems of polyphase windings (80A, 80B). There it is,
two systems of power converters (60A, 60B) that can individually energize the two systems of polyphase windings;
Two systems are provided independently of each other and perform calculations to control the current flowing from the power converter to the polyphase winding based on the current flowing to the polyphase winding detected by the current detector (70A, 70A). Arithmetic devices (40A, 40B),
Equipped with
The arithmetic device of each system is
Estimating the current in other systems based on the current flowing through the polyphase windings of the own system and the output voltages of the own system and other systems,
A control device for multiphase rotating machines that suppresses the effects of interference between systems based on estimated currents of other systems. The control device for a polyphase rotating machine according to claim 1, wherein the arithmetic unit of each system estimates the current of other systems using an observer (570A, 570B). The control device for a polyphase rotating machine according to claim 1 or 2, wherein the arithmetic unit of each system performs non-interference control using the estimated current of other systems. 4. The arithmetic unit of each system controls the current flowing through the polyphase winding of the own system by controlling the sum and difference between the current of the own system and the estimated current of other systems. A control device for a multiphase rotating machine according to item (1). The control device for a polyphase rotating machine according to claim 2, wherein the gain of the observer is variably set according to the rotation speed of the polyphase rotating machine. Any one of claims 1 to 5, wherein the arithmetic unit of each system bidirectionally communicates at least abnormality information of each system through intersystem communication, and performs estimation calculation of other system current only when the other system is normal. A control device for a multi-phase rotating machine as described in 2. The arithmetic device of each system is
Any one of claims 1 to 6, wherein the current limit value is calculated based on a signal input to each of the calculation devices at a cycle longer than the cycle of current control calculation, and the current limit value is transmitted in at least one direction by intersystem communication. A control device for a multiphase rotating machine according to item 1. The arithmetic device of each system is
An abnormality signal indicating the presence or absence of an abnormality is calculated using a signal input to each of the calculation devices at a cycle longer than a cycle of current control calculation, and the abnormality signal is transmitted in at least one direction through intersystem communication. 8. The control device for a multiphase rotating machine according to any one of 7. A control device for a multiphase rotating machine mounted on a vehicle,
The control device for a multiphase rotating machine according to any one of claims 6 to 8, wherein the inter-system communication is performed via a vehicle network. The control device for a multiphase rotating machine according to any one of claims 6 to 8, wherein the inter-system communication is performed within the control device.

Images