Inverter control method and inverter control device

Abstract

To protect a smoothing capacitor when upper side short circuit control and lower side short circuit control are switched in performing ASC control by a system using an open winding type rotary electric machine.SOLUTION: Upper side short circuit control and lower side short circuit control are alternately executed by a first inverter 18 and a second inverter 19 connected to an open winding type rotary electric machine (10). When the upper side short circuit control and the lower side short circuit control are switched, for a start state, the state of one of the first inverter 18 and the second inverter 19 is maintained in the same state as the start state. A shift state in which switching of the other of the first inverter 18 and the second inverter 19 is started previously is formed, and an end state is formed by starting switching of the one inverter after a predetermined time (Tdelay) has elapsed after starting switching of the other inverter previously.SELECTED DRAWING: Figure 9

Description

Technical Field

[0001] The present invention relates to a control method and a control device for an inverter that connects a DC power supply and a rotating electrical machine via a smoothing capacitor.

Background Art

[0002] Patent Document 1 discloses an inverter device that alternately performs first switching control and second switching control. The first switching control is control that turns on all the switching elements connected to the positive electrode side of the DC power supply and turns off all the switching elements connected to the negative electrode side of the DC power supply. The second switching control is control that turns off all the switching elements connected to the positive electrode side of the DC power supply and turns on all the switching elements connected to the negative electrode side of the DC power supply.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The inverter has a pair of switching elements that constitute an upper arm and a lower arm for each phase of the rotating electrical machine. Further, the inverter connects the DC power supply and the rotating electrical machine via a smoothing capacitor. When an abnormality occurs in the control of the rotating electrical machine, such as when the rotating electrical machine reaches overspeed, a short-circuit circuit may be formed between the inverter and the rotating electrical machine by turning on all of either the upper arm or the lower arm of each phase and turning off the other. Such short-circuit control is referred to as ASC (active short circuit) control, three-phase short-circuit control, or the like.

[0005] The ASC control can be performed by a short - circuit control that turns on all the upper arms and turns off all the lower arms (hereinafter referred to as upper - side short - circuit control), or a short - circuit control that turns off all the upper arms and turns on all the lower arms (hereinafter referred to as lower - side short - circuit control). However, when the rotating electrical machine is rotating, heat and current may be biased to a specific arm, and that specific arm may be prone to deterioration. For this reason, when the rotating electrical machine is rotating, it is preferable that the ASC control be performed while alternately switching between the upper - side short - circuit control and the lower - side short - circuit control.

[0006] In order to switch between the upper - side short - circuit control and the lower - side short - circuit control, it is necessary to switch the upper and lower arms of each phase. At this time, if the upper arm and the lower arm are turned on simultaneously and a through - current flows, there is a risk of damaging the switching element or the like. For this reason, a dead - time for surely turning off both of them is provided for the switching of the paired upper and lower arms. However, if a dead - time is simply provided in the switching between the upper - side short - circuit control and the lower - side short - circuit control, depending on the current (induced voltage of the rotating electrical machine) flowing into the rotating electrical machine before the dead - time, a large current may flow into the smoothing capacitor during the dead - time, and there is a risk of damaging the smoothing capacitor.

[0007] An object of the present invention is to provide a control method and a control device for an inverter that can protect a smoothing capacitor when switching between an upper - side short - circuit control and a lower - side short - circuit control in a system using an open - winding type rotating electrical machine for ASC control.

Means for Solving the Problem

[0008] One aspect of the present invention is an inverter control method for controlling a first inverter that connects one end of a stator coil of an open-wound rotating electrical machine to a DC power supply via a smoothing capacitor, and a second inverter that connects the other end of the stator coil to the DC power supply via the smoothing capacitor. In this inverter control method, an upper short-circuit control for turning on the upper arms of the first inverter and the second inverter and turning off the lower arms of the first inverter and the second inverter, and an upper arm of the first inverter and the second inverter are turned off, and a lower short-circuit control for turning off the lower arms of the first inverter and the second inverter are alternately executed. Then, when switching between the upper short-circuit control and the lower short-circuit control, while maintaining the state of one of the first inverter or the second inverter in the same state as the starting state with respect to the starting state in which the upper short-circuit control or the lower short-circuit control is being performed, a transition state is formed in which the switching of the other inverter among the first inverter or the second inverter is started in advance, and after a predetermined time has elapsed since the switching of the other inverter was started in advance, by starting the switching of one inverter, the states of the upper arm and the lower arm of one inverter are reversed with respect to the starting state, and a final state is formed in which the states of the upper arm and the lower arm of the other inverter are reversed with respect to the starting state.

Effect of the Invention

[0009] According to the present invention, when performing ASC control in a system using an open-wound rotating electrical machine, it is possible to provide an inverter control method and a control device that can protect a smoothing capacitor when switching between upper short-circuit control and lower short-circuit control.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

[0011] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0012] [First Embodiment] FIG. 1 is a block diagram showing the configuration of the electric vehicle 100 according to the first embodiment. The electric vehicle 100 is a vehicle driven by electric power, specifically, an electric vehicle or a hybrid vehicle, etc. As shown in FIG. 1, the electric vehicle 100 includes a rotating electric machine 10, a battery 11, an inverter 12, and a controller 13.

[0013] The rotating electrical machine 10 is a motor or a generator. In this embodiment, the rotating electrical machine 10 is a motor and functions as a driving power source of the electric vehicle 100. More specifically, the rotating electrical machine 10 of this embodiment is a three-phase AC synchronous motor having three phases of U-phase, V-phase, and W-phase. The currents flowing through each of the U, V, and W phases (hereinafter collectively referred to as phase currents i u , i v , i w ) are appropriately acquired as necessary by the current sensor 14. Also, the electrical angle θ of the rotor (not shown) is appropriately acquired as necessary by the rotation sensor 15.

[0014] In this embodiment, in particular, the rotating electrical machine 10 is of a so-called open-winding type. That is, the stator coils (U, V, W) of each of the U, V, and W phases of the rotating electrical machine 10 are not connected, and the stator coils (U, V, W) of each phase are connected to the inverter 12 respectively.

[0015] The battery 11 is a DC power source that supplies power to the rotating electrical machine 10. The battery 11 is constituted by, for example, a lithium-ion battery or the like. In this embodiment, the battery 11 is rechargeable.

[0016] The inverter 12 converts the DC power input from the battery 11 into AC power by PWM (Pulse Width Modulation) control and supplies it to the rotating electrical machine 10. The inverter 12 is connected to the battery 11 via the relay 16. The DC voltage V dc on the input side (battery 11 side) of the inverter 12 is appropriately acquired as necessary by the voltage sensor 17.

[0017] In this embodiment, since the rotating electrical machine 10 is of the open-winding type, the inverter 12 includes a first inverter 18 (INV1) and a second inverter 19 (INV2). The first inverter 18 (INV1) connects one end of each phase stator coil (U, V, W) of the rotating electrical machine 10 to the battery 11. The second inverter 19 (INV2) connects the other end of each phase stator coil (U, V, W) to the battery 11 which is a DC power source. Then, the first inverter 18 and the second inverter 19 cooperate to convert the DC power output by the battery 11 into AC power and supply it to the rotating electrical machine 10. Thereby, the inverter 12 (the first inverter 18 and the second inverter 19) controls the phase currents i u , i v , i w .

[0018] The controller 13 is a control device that comprehensively controls the operations of each part of the electric vehicle 100. In this embodiment, the controller 13 functions particularly as a control device for the inverter 12. The controller 13 is constituted by, for example, one or more computers or circuits, etc. Also, the controller 13 is programmed to control the operations of each part at a predetermined control period τ.

[0019] Specifically, the controller 13 includes a torque control unit 21, a voltage distribution unit 22, a coordinate conversion unit 23, a PWM control unit 24, a PWM signal generator 25, a coordinate conversion unit 26, a rotational speed calculation unit 27, and an abnormality detection unit 28, etc.

[0020] The torque control unit 21 calculates the dq-axis voltage command values v * so that the output torque of the rotating electrical machine 10 coincides with or follows the torque command value T d * , v q * .

[0021] The torque command value T *is a command value (target value) for the torque to be output by the rotating electrical machine 10, and represents the required torque from the driver or the like to the electric vehicle 100. The controller 13 calculates (determines) the torque command value T * for example, according to the operation amount of the accelerator pedal or the like. Further, instead of calculating the torque command value T * itself, the controller 13 can obtain the torque command value T * from a higher-level controller. In this embodiment, for simplicity, it is assumed that the torque command value T * is known.

[0022] The dq-axis voltage command values v d * , v q * are command values (target values) for the voltages in the dq-axis coordinate system that rotate with the rotor. Note that the dq-axis voltage command values v d * , v q * consist of the d-axis voltage command value v d which is the command value for the d-axis voltage v d * and the q-axis voltage command value v q which is the command value for the q-axis voltage v q * .

[0023] More specifically, the torque control unit 21 calculates the dq-axis voltage command values v * , v dc based on the torque command value T d , the direct current voltage V q , the dq-axis currents i d * , i q * , and the rotational speed N by so-called current vector control (CVC) or voltage phase control (VPC). The dq-axis currents i d , i q are the detected values or estimated values of the d-axis current i d and the q-axis current i q . In this embodiment, the dq-axis voltages v d , v q are detected values, and in the coordinate conversion unit 26, the phase current iu , i v , i w and is calculated based on the detected value of the electrical angle θ. The rotational speed N [rpm] is a detected value or an estimated value. In the present embodiment, the rotational speed N is calculated based on the electrical angle θ in the rotational speed calculation unit 27.

[0024] The voltage distribution unit 22 distributes the dq-axis voltage command values v d * , v q * calculated by the torque control unit 21 to the dq-axis voltage command values for the first inverter 18 (hereinafter referred to as the first dq-axis voltage command values v d1 * , v q1 * and the dq-axis voltage command values for the second inverter 19 (hereinafter referred to as the second dq-axis voltage command values v d2 * , v q2 * ). Specifically, the voltage distribution unit 22 distributes the dq-axis voltage command values v d * , v q * calculated by the torque control unit 21 to the first dq-axis voltage command values v d1 * , v q1 * and the second dq-axis voltage command values v d2 * , v q2 * . The first dq-axis voltage command values v d1 * , v q1 * are the dq-axis voltage command values for the first inverter 18, and the second dq-axis voltage command values v d2 * , v q2 * are the dq-axis voltage command values for the second inverter 19.

[0025]

Equation

[0026] The coordinate conversion unit 23 calculates a three-phase voltage command value based on the dq-axis voltage command value and the electrical angle θ by performing a coordinate conversion from the dq-axis coordinate system to the UVW coordinate system. The three-phase voltage command value is a command value for the voltage to be applied to each of the U, V, and W phases of the rotating electrical machine 10 via the inverter 12.

[0027] The coordinate conversion unit 23 includes a first conversion unit 31 and a second conversion unit 32.

[0028] The first conversion unit 31 calculates a first three-phase voltage command value v d1 * , v q1 * based on the first dq-axis voltage command value v u1 * , v v1 * , v w1 * The first three-phase voltage command value v u1 * , v v1 * , v w1 * is a command value for the voltage to be applied to each phase of the rotating electrical machine 10 via the first inverter 18. The first three-phase voltage command value v u1 * , v v1 * , v w1 * consists of a first U-phase voltage command value v u1 * , a first V-phase voltage command value v v1 * , and a first W-phase voltage command value v w1 * .

[0029] The second conversion unit 32 calculates a second three-phase voltage command value v d2 * , v q2 * based on the second dq-axis voltage command value v u2 * , v v2 * , v w2 * The second three-phase voltage command value vu2 * , v v2 * , v w2 * is the command value for the voltage to be applied to each phase via the second inverter 19. The second three-phase voltage command value v u2 * , v v2 * , v w2 * is the second U-phase voltage command value v u2 * , the second V-phase voltage command value v v2 * , and the second W-phase voltage command value v w2 * consists of.

[0030] More specifically, the first conversion unit 31 calculates the first three-phase voltage command value v u1 * , v v1 * , v w1 * according to the following formula (2). Also, the second conversion unit 32 calculates the second three-phase voltage command value v u2 * , v v2 * , v w2 * according to the following formula (3).

[0031]

Equation

[0032] The PWM control unit 24 calculates a duty command value that determines the duty ratio in PWM control.

[0033] Specifically, the PWM control unit 24 includes a first PWM control unit 33 and a second PWM control unit 34. The first PWM control unit 33 calculates a first duty command value used for controlling the first inverter 18. The first duty command value consists of first duty command values Duty_U1, Duty_V1, and Duty_W1 for each of the U, V, and W phases. Hereinafter, the first duty command values Duty_U1, Duty_V1, and Duty_W1 may be abbreviated as Duty_UVW1, etc. The second PWM control unit 34 calculates a second duty command value used for controlling the second inverter 19. The second duty command value consists of second duty command values Duty_U2, Duty_V2, and Duty_W2 for each of the U, V, and W phases. Hereinafter, the second duty command values Duty_U2, Duty_V2, and Duty_W2 may be abbreviated as Duty_UVW2, etc.

[0034] Also, the PWM control unit 24 includes an output determination unit 35. The output determination unit 35 sets or changes the gate permission signal S GP and the generation mode of the duty command values (Duty_UVW1, Duty_UVW2) (hereinafter referred to as the duty command value generation mode S mode ).

[0035] The gate permission signal S GP is a signal that uniformly permits or prohibits the gate voltage control of the switching element. When the gate permission signal S GP is "permit", the inverter 12 becomes operable. On the other hand, when the gate permission signal S GP is "prohibit", the inverter 12 becomes substantially inoperable.

[0036] In this embodiment, the output determination unit 35 is based on the status signal S output by the abnormality detection unit 28 state , the absolute value |N| of the rotational speed N, and the DC voltage V dc to set the gate permission signal S GP and the generation mode of the duty command values (Duty_UVW1, Duty_UVW2) (hereinafter referred to as the duty command value generation mode S modeSet or change (hereinafter referred to as). Specifically, the PWM control unit 24 sets or changes the gate permission signal S GP and the duty command value generation mode S mode as follows.

[0037]

Table 1

[0038] As shown in Table 1, when the state signal S sate indicates "normal" and the rotating electrical machine 10 can be driven without any problems, the PWM control unit 24 sets the gate permission signal S dc to "permit" and sets the duty command value generation mode S GP regardless of the absolute value |N| of the rotational speed N and the DC voltage V mode to the normal mode. The normal mode is a mode for generating the duty command value Duty_UVW for driving the rotating electrical machine 10. In the normal mode, the first PWM control unit 33 calculates the first duty command value Duty_UVW1 [%] according to the following formula (4) based on the first three-phase voltage command values v u1 * , v v1 * , v w1 * . Similarly, in the normal mode, the second PWM control unit 34 calculates the second duty command value Duty_UVW2 [%] according to the following formula (5) based on the second three-phase voltage command values v u2 * , v v2 * , v w2 * .

[0039]

Equation

[0040] The state signal S stateindicates "abnormal", and when there is an obstacle to the drive of the rotating electrical machine 10, the output determination unit 35 determines the gate permission signal S dc and the duty command value generation mode S GP based on the absolute value |N| of the rotational speed N and the DC voltage V mode .

[0041] Specifically, when the state signal S state is "abnormal" and the absolute value |N| of the rotational speed N is greater than the rotational speed upper limit value N MAX (|N| > N MAX ), the output determination unit 35 sets the gate permission signal S GP to "permit", and sets the duty command value generation mode S mode to the ASC mode. The rotational speed upper limit value N MAX is determined in advance by adaptation based on experiments or simulations. The ASC mode is a mode for generating duty command values (Duty_UVW1, Duty_UVW2) for forming a short-circuit circuit between the inverter 12 and the rotating electrical machine 10.

[0042] In particular, in the present embodiment, the output determination unit 35 sets the duty command value generation mode S dc to either the first ASC mode (ASC1) or the second ASC mode (ASC2) according to the DC voltage V mode .

[0043] As shown in Table 1, the first ASC mode (ASC1) is the ASC mode selected when the DC voltage V dc exceeds the voltage threshold V th . The second ASC mode (ASC2) is the ASC mode selected when the DC voltage V dc is less than or equal to the voltage threshold V th . Note that the voltage threshold V dc for the DC voltage V th is determined in advance by adaptation based on experiments or simulations.

[0044] Also, as shown in Table 1, the state signal S stateis "abnormal", and when the absolute value |N| of the rotational speed N is less than or equal to the rotational speed upper limit value N MAX (|N| ≦ N MAX ), the output determination unit 35 sets the gate permission signal S dc to "prohibited" regardless of the DC voltage V GP . In this case, the first PWM control unit 33 and the second PWM control unit 34 do not generate duty command values (Duty_UVW1, Duty_UVW2). Therefore, the rotating electrical machine 10 is stopped without performing ASC control.

[0045] When the gate permission signal S GP is "permitted", the PWM signal generator 25 generates drive signals (hereinafter referred to as PWM signals) D uu1 * ~D wl1 * , D uu2 * ~D wl2 * for each switching element of the inverter 12 by comparing the carrier signal CS with the duty command values (Duty_UVW1, Duty_UVW2). The inverter 12 controls the on / off of the switching elements according to this PWM signal D uu1 * ~D wl1 * , D uu2 * ~D wl2 * to perform drive control or ASC control of the rotating electrical machine 10. In this embodiment, for simplicity, it is assumed that the carrier signal CS is a triangular wave having a constant period (2τ).

[0046] In this embodiment, the PWM signal generator 25 includes a first PWM signal generator 36 and a second PWM signal generator 37. The first PWM signal generator 36 controls the on / off of the switching elements of the first inverter 18 by comparing the carrier signal CS with the first duty command value Duty_UVW1, and generates a PWM signal D uu1 * ~D wl1 *is generated. The second PWM signal generator 37 controls the on / off of the switching elements of the second inverter 19 by comparing and matching the carrier signal CS and the second duty command value Duty_UVW2 to generate a PWM signal D uu2 * ~D wl2 * is generated. When the gate permission signal S GP is "prohibited", the first PWM signal generator 36 and the second PWM signal generator 37 stop generating the PWM signals D uu1 * ~D wl1 * ,D uu2 * ~D wl2 * .

[0047] The coordinate conversion unit 26 calculates the dq-axis currents i u ,i v ,i w of the rotating electrical machine 10 based on the phase currents i d ,i q and the electrical angle θ. Specifically, the coordinate conversion unit 26 calculates the dq-axis currents i d ,i q according to the following formula (6).

[0048]

Equation

[0049] The rotational speed calculation unit 27 calculates the rotational speed N based on the electrical angle θ. Specifically, the rotational speed calculation unit 27 calculates the rotational speed N by unit-converting the change amount (dθ / dt) of the electrical angle θ per unit time using the number of pole pairs p of the rotating electrical machine 10 according to the following formula (7).

[0050]

Equation

[0051] In addition, in the present embodiment, the rotation speed N is used as a parameter representing the rotation state of the rotating electrical machine 10 in the drive control of the rotating electrical machine 10 and the ASC control, but it is not limited thereto. For example, as a parameter representing the rotation state of the rotating electrical machine 10, instead of the rotation speed N, the electrical angle θ, the electrical angular velocity, the mechanical angle, or the mechanical angular velocity, etc. can be used. Also in this case, the drive control of the rotating electrical machine 10 and the ASC control can be performed.

[0052] The abnormality detection unit 28 detects an abnormality in the system that drives the rotating electrical machine 10, that is, whether there is an obstacle to driving the rotating electrical machine 10. In the present embodiment, the abnormality detection unit 28 monitors the rotation speed N of the rotating electrical machine 10, and when the absolute value |N| of the rotation speed N exceeds the rotation speed upper limit value N MAX and the rotating electrical machine 10 reaches overspeed, it is determined that an abnormality has occurred in the system. Further, the abnormality detection unit 28 monitors the phase currents i u , i v , i w , and when any of these phase currents i u , i v , i w exceeds a predetermined threshold and reaches an overcurrent state, it is determined that an abnormality has occurred in the system. Note that the abnormality detection unit 28 can monitor parameters other than the above, such as the voltage and temperature of the rotating electrical machine 10 and the inverter 12 (switching element), and determine that an abnormality has occurred in the system. Also, when the abnormality detection unit 28 detects a failure of the current sensor 14, the voltage sensor 17, or other detectors (not shown), it can determine that there is an abnormality in the system. In the present embodiment, for simplicity, as described above, the abnormality detection unit 28 is assumed to detect overspeed and overcurrent as abnormalities in the system.

[0053] The abnormality detection unit 28 inputs the result of the abnormality detection to the PWM control unit 24 as a state signal S state . In the present embodiment, the state signal S state is a signal indicating either "normal" or "abnormal".

[0054] In addition, when the abnormality detection unit 28 detects an abnormality in the system, if necessary according to the cause of the abnormality or the like, it can further output a relay cut-off signal S RC When this happens, the relay 16 turns off, and the connection between the battery 11 and the inverter 12 is forcibly disconnected. In the present embodiment, when the abnormality detection unit 28 detects over-rotation or overcurrent, it outputs the relay cut-off signal S RC . Therefore, when over-rotation or overcurrent occurs, the relay 16 is forcibly turned off, and the connection between the battery 11 and the inverter 12 is disconnected.

[0055] FIG. 2 is a circuit diagram showing the configuration of the inverter 12. As shown in FIG. 2, the inverter 12 includes a smoothing capacitor 39, a first inverter 18 which is a first bridge circuit formed by a plurality of switching elements Q1 to Q6, and a plurality of switching elements Q7 to Q 12 which is a second bridge circuit formed by, and a second inverter 19.

[0056] The smoothing capacitor 39 is provided at the input terminal of the inverter 12 to smooth the input voltage from the battery 11. The voltage across the smoothing capacitor 39 is the DC voltage V dc in the present embodiment. The switching elements Q1 to Q 12 are configured using power semiconductor elements such as IGBTs (insulated gate bipolar transistors) and MOSFETs (metal oxide semiconductor field effect transistors), and each includes a freewheeling diode. At the timing of forward bias, the switching elements Q1 to Q 12 become conductive in the forward direction (from the positive electrode side to the negative electrode side of the battery 11) when they are controlled to be on, and are non-conductive when they are controlled to be off. Also, at the timing of reverse bias, the switching elements Q1 to Q6 can conduct in the reverse direction (from the negative electrode side to the positive electrode side of the battery 11) through the freewheeling diodes.

[0057] The switching elements Q1 to Q6 that make up the first inverter 18 configure the legs of each phase of UVW as follows.

[0058] The switching elements Q1 and Q2 constitute the U-phase leg of the first inverter 18 (hereinafter referred to as the first U-phase leg). That is, one end of the U-phase stator coil (U) is connected between the serially connected switching elements Q1 and Q2. The switching element Q1 is the upper arm of the first U-phase leg and is connected to the positive electrode side (high side) of the battery 11. Also, the switching element Q2 is the lower arm of the first U-phase leg and is connected to the negative electrode side (low side) of the battery 11. The PWM signal for controlling the on / off of the switching element Q1 is D uu * and the PWM signal for controlling the on / off of the switching element Q2 is D ul * is.

[0059] The switching elements Q3 and Q4 constitute the V-phase leg of the first inverter 18 (hereinafter referred to as the first V-phase leg). That is, one end of the V-phase stator coil (V) is connected between the serially connected switching elements Q3 and Q4. The switching element Q3 is the upper arm of the first V-phase leg and the switching element Q4 is the lower arm of the first V-phase leg. The PWM signal for controlling the on / off of the switching element Q3 is D vu * and the PWM signal for controlling the on / off of the switching element Q4 is D vl * is.

[0060] Similarly, the switching elements Q5 and Q6 constitute the W-phase leg of the first inverter 18 (hereinafter referred to as the first W-phase leg). That is, one end of the W-phase stator coil (W) is connected between the serially connected switching elements Q5 and Q6. The switching element Q5 is the upper arm of the first W-phase leg and the switching element Q6 is the lower arm of the first W-phase leg. The PWM signal for controlling the on / off of the switching element Q5 is D wu * and the PWM signal for controlling the on / off of the switching element Q6 is D wl * is.

[0061] The switching elements Q7 to Q that make up the second inverter 19 12 constitute the legs of each of the U, V, and W phases in the same way as the switching elements Q1 to Q6 of the first inverter 18.

[0062] The switching elements Q7 and Q8 constitute the U-phase leg of the second inverter 19 (hereinafter referred to as the second U-phase leg). That is, the other end of the U-phase stator coil (U) is connected between the serially connected switching elements Q7 and Q8. The switching element Q7 is the upper arm of the second U-phase leg, and the switching element Q8 is the lower arm of the second U-phase leg. The PWM signals for controlling the on / off of the switching elements Q7 and Q8 are D uu2 * , D ul2 * respectively.

[0063] The switching elements Q9 and Q 10 constitute the V-phase leg of the second inverter 19 (hereinafter referred to as the second V-phase leg). That is, the other end of the V-phase stator coil (V) is connected between the serially connected switching elements Q9 and Q 10 . The switching element Q9 is the upper arm of the second V-phase leg, and the switching element Q 10 is the lower arm of the second V-phase leg. The PWM signals for controlling the on / off of the switching elements Q9 and Q 10 are D vu2 * , D vl2 * respectively.

[0064] Similarly, the switching elements Q 11 , Q 12 constitute the W-phase leg of the second inverter 19 (hereinafter referred to as the second W-phase leg). That is, the other end of the W-phase stator coil (W) is connected between the serially connected switching elements Q 11 , Q 12 . The switching element Q 11 is the upper arm of the second W-phase leg, and the switching element Q12 is the lower arm of the second W-phase leg. The PWM signals for controlling the on / off of the switching elements Q 11 , Q 12The PWM signals for controlling the on / off are respectively D wu2 * , D wl2 * .

[0065] FIG. 3 is an explanatory diagram showing the PWM signals in the normal mode. In FIG. 3, for the first U-phase leg of the first inverter 18, its first duty command value Duty_U1 and the PWM signal D uu1 * , D ul1 * are shown. The same applies to the first duty command value Duty_VW1 for the first V-phase leg and the first W-phase leg, and the PWM signals D vu1 * ~D wl1 * . Also, the same applies to the second duty command value Duty_UVW2 for the second inverter 19 and the PWM signals D uu2 * ~D wl2 * .

[0066] As shown in FIG. 3, the first PWM control unit 33 calculates and updates the first duty command value Duty_U1 every predetermined control period (τ). However, the first duty command value Duty_U1 calculated in a certain control period is reflected in the next control period. For example, the first duty command value Duty_U1 calculated in the control period P1 is used in the control period P2. Similarly, the first duty command value Duty_U1 calculated in the control period P2 is used in the control period P3, and the first duty command value Duty_U1 calculated in the control period P3 is used in the control period P4.

[0067] The first PWM signal generator 36, in principle, generates the PWM signal D uu1 * , D ul1 * when the carrier signal CS matches the first duty command value Duty_U1 calculated and updated as described above.Switch the ON / OFF state. Also, the PWM signal D for the upper arm uu1 * and the PWM signal D for the lower arm ul1 * are complementary. That is, when the PWM signal D uu1 * for the upper arm is switched from ON to OFF, the PWM signal D ul1 * for the lower arm is switched from OFF to ON. Also, when the PWM signal D ul1 * for the lower arm is switched from ON to OFF, the PWM signal D uu1 * for the upper arm is switched from OFF to ON.

[0068] However, in order to prevent the upper arm and the lower arm from being ON simultaneously and a through current from flowing through the first U-phase leg, the first PWM signal generator 36 provides a dead time DT. The dead time DT is the period during which both the PWM signal D uu1 * for the upper arm and the PWM signal D ul1 * for the lower arm are both OFF. In this embodiment, the first PWM signal generator 36 inserts the dead time DT into the PWM signal of the arm being switched from OFF to ON. That is, when switching the upper arm to ON, the timing of turning on the PWM signal D uu1 * for the upper arm is delayed by the amount of the dead time DT. Similarly, when switching the lower arm to ON, the timing of turning on the PWM signal D ul1 * for the lower arm is delayed by the amount of the dead time DT.

[0069] Figure 4 is a block diagram showing the configuration of the part of the PWM control unit 24 related to the ASC control. As shown in Figure 4, the first PWM control unit 33 and the second PWM control unit 34 of the PWM control unit 24 include an up / down switching signal generation unit 41, a three-phase sum processing unit 42, a timing adjustment processing unit 43, and an ASC switching processing unit 44.

[0070] The up / down switching signal generation unit 41 generates an ASC switching signal S at a timing for switching between upper short-circuit control and lower short-circuit control. The upper short-circuit control is an ASC control for forming a short-circuit circuit by turning on the upper arms of the first inverter 18 and the second inverter 19 and turning off the lower arms of the first inverter 18 and the second inverter 19. In the upper short-circuit control, all the upper arms of the first U-phase leg, the first V-phase leg, the first W-phase leg, the second U-phase leg, the second V-phase leg, and the second W-phase leg are turned on, and all the lower arms are turned off. The lower short-circuit control is an ASC control for forming a short-circuit circuit by turning off the upper arms of the first inverter 18 and the second inverter 19 and turning on the lower arms of the first inverter 18 and the second inverter 19. In the lower short-circuit control, all the upper arms of the first U-phase leg, the first V-phase leg, the first W-phase leg, the second U-phase leg, the second V-phase leg, and the second W-phase leg are turned off, and all the lower arms are turned on. UL1 The ASC switching signal S is represented by "ON" designating the upper short-circuit control or "OFF" designating the lower short-circuit control. The execution times t (see FIG. 5) of the upper short-circuit control and the lower short-circuit control are generally common. The execution times t of the upper short-circuit control and the lower short-circuit control are set to be at least sufficiently longer than the control period τ. The execution times t of the upper short-circuit control and the lower short-circuit control are determined by adaptation based on experiments or simulations, etc.

[0071] The ASC switching signal S UL1 is represented by "ON" designating the upper short-circuit control or "OFF" designating the lower short-circuit control. The execution times t ASC (see FIG. 5) of the upper short-circuit control and the lower short-circuit control are generally common. The execution times t ASC of the upper short-circuit control and the lower short-circuit control are set to be at least sufficiently longer than the control period τ. The execution times t ASC of the upper short-circuit control and the lower short-circuit control are determined by adaptation based on experiments or simulations, etc.

[0072] The three-phase sum processing unit 42 calculates the sum of the phase currents i u , i v , i w . In the open-winding type rotating electrical machine 10, the phase currents i u , i v , i wThe sum represents the zero-phase current i0, which is the DC current component flowing through each phase in common. That is, the three-phase sum processing unit 42 calculates the zero-phase current i0 by summing the phase currents i u , i v , i w according to the following formula (8).

[0073]

Equation

[0074] The timing adjustment processing unit 43 generates the final ASC switching signal S UL1 by correcting the ASC switching signal S UL2 so that the timing for switching between the upper short-circuit control and the lower short-circuit control is adjusted according to the zero-phase current i0.

[0075] FIG. 5 is a time chart showing the phase currents i u , i v , i w , the zero-phase current i0, and the ASC switching signals S UL1 , S UL2 . As shown in FIG. 5, the ASC switching signal S UL1 generated by the up / down switching signal generation unit 41 is set such that the upper short-circuit control and the lower short-circuit control are alternately executed every elapse of a predetermined execution time t u , i v , i w regardless of the phase currents i and the zero-phase current i0. When the zero-phase current i0 is negative and the basic ASC switching signal S ASC attempts to switch between the upper short-circuit control and the lower short-circuit control, the timing adjustment processing unit 43 corrects the basic ASC switching signal S UL1 to delay the switching timing between the upper short-circuit control and the lower short-circuit control until the zero-phase current i0 becomes a value greater than zero (i0 > 0), thereby generating the final ASC switching signal S UL1 UL2 .

[0076] Specifically, in FIG. 5, the basic ASC switching signal S UL1 ​The first switching timing specified is the timing at which the zero-phase current i0 becomes negative. Therefore, the final ASC switching signal S UL2 is modified so that this switching is delayed until the zero-phase current i0 becomes greater than zero.

[0077] Here, the ASC switching signal S UL1 , S UL2 illustrates the timing at which it changes from off to on (the timing of switching from lower short-circuit control to upper short-circuit control), but it is not limited to this. The ASC switching signal S UL1 , S UL2 is the same for the timing at which it changes from off to on (the timing of switching from upper short-circuit control to lower short-circuit control). That is, when the basic ASC switching signal S UL1 tries to switch the ASC switching signal S UL1 , S UL2 from off to on at the timing when the zero-phase current i0 becomes negative, in the final switching signal S UL2 this switching is delayed until the zero-phase current i0 becomes greater than zero.

[0078] In this embodiment, the current flowing from the first inverter 18 through the stator coils (U, V, W) to the second inverter 19 is defined as positive. That is, for example, the U-phase current i u flowing from the first U-phase leg through the U-phase stator coil (U) to the second U-phase leg is the positive U-phase current i u , and the U-phase current i u flowing from the second U-phase leg through the U-phase stator coil (U) to the first U-phase leg is the negative U-phase current i u . The same applies to the V-phase current i v , the W-phase current i w , and the zero-phase current i0.

[0079] The ASC switching processing unit 44 (see FIG. 4) determines the carrier signal CS, the final ASC switching signal S UL2 , and the duty command value generation mode S modeGenerate duty command values (Duty_UVW1, Duty_UVW2) for the ASC mode according to

[0080]

Table 2

[0081] Specifically, when the duty command value generation mode S mode is the first ASC mode (ASC1), when the ASC switching processing unit 44 switches the upper short-circuit control and the lower short-circuit control, the first duty command value Duty_UVW1 for the first inverter 18 and the second duty command value Duty_UVW2 for the second inverter 19 are set to different values. However, the first duty command value Duty_UVW1 is common for each of the UVW phases. Therefore, when comparing between the phases of the first inverter 18, the switching timings of the arms of all phases are substantially simultaneous. Similarly, the second duty command value Duty_UVW2 is common for each of the UVW phases. Therefore, when comparing between the phases of the second inverter 19, the switching timings of the arms of all phases are substantially simultaneous.

[0082] That is, when the ASC switching processing unit 44 performs the ASC control in the first ASC mode, the switching timings of the arms of the first inverter 18 and the second inverter 19 are made different. However, the ASC switching processing unit 44 simultaneously switches the arms of each phase of the first inverter 18 and simultaneously switches the arms of each phase of the second inverter 19.

[0083] As a result, when switching between the upper short-circuit control and the lower short-circuit control, the ASC switching processing unit 44 maintains the state of one of the first inverter 18 or the second inverter to the same state as the starting state while forming a transition state in which the switching of the other one of the first inverter 18 or the second inverter 19 is started in advance. Further, after a predetermined time elapses since the switching of the other inverter (the inverter to be switched in advance) is started in advance, the ASC switching processing unit 44 starts the switching of one inverter (the inverter to be switched with a delay). After that, when the switching of one inverter is completed, a final state is formed in which the states of the upper arm and the lower arm of one inverter are reversed with respect to the starting state, and the states of the upper arm and the lower arm of the other inverter are reversed with respect to the starting state.

[0084] In particular, the ASC switching processing unit 44 forms a first transition state, a second transition state, and a third transition state as the above-described transition state.

[0085] The first transition state is a state in which the upper arm and the lower arm of one of the first inverter 18 or the second inverter 19 are both turned off while maintaining the states of the upper arm and the lower arm of one of the first inverter 18 or the second inverter 19 in the same state as the starting state.

[0086] The second transition state is a state in which the states of the upper arm and the lower arm of one inverter are maintained in the same state as the starting state while the states of the upper arm and the lower arm of the other inverter are reversed with respect to the starting state.

[0087] The third transition state is a state in which the upper arm and the lower arm of one inverter are both turned off while maintaining the state in which the states of the upper arm and the lower arm of the other inverter are not reversed with respect to the starting state.

[0088] In this embodiment, the ASC switching processing unit 44 sets the first duty command value Duty_UVW1 and the second duty command value Duty_UVW2 such that the arms of the second inverter 19 are switched first, and then the arms of the first inverter 18 are switched with a delay.

[0089] Therefore, when switching between the upper short-circuit control and the lower short-circuit control, the ASC switching processing unit 44 forms a transition state in which the switching of the second inverter 19 is started first while maintaining the state of the first inverter 18 in the same state as the initial state with respect to the initial state in which the upper short-circuit control and the lower short-circuit control are being performed. Further, the ASC switching processing unit 44 starts the switching of the first inverter 18 after a predetermined time has elapsed since the switching of the second inverter 19 was started. After that, when the switching of the first inverter 18 is completed, a final state is formed in which the states of the upper arm and the lower arm of the first inverter are reversed with respect to the initial state, and the states of the upper arm and the lower arm of the second inverter 19 are reversed with respect to the initial state.

[0090] For this reason, the first transition state in this embodiment is a state in which both the upper arm and the lower arm of the second inverter 19 are turned off while maintaining the states of the upper arm and the lower arm of the first inverter 18 in the same state as the initial state. The second transition state in this embodiment is a state in which the states of the upper arm and the lower arm of the second inverter 19 are reversed with respect to the initial state while maintaining the states of the upper arm and the lower arm of the first inverter 18 in the same state as the initial state. And the third transition state in this embodiment is a state in which both the upper arm and the lower arm of the first inverter 18 are turned off while maintaining the state in which the states of the upper arm and the lower arm of the second inverter 19 are not reversed with respect to the initial state.

[0091] As is clear from the calculation formula shown in Table 2, the delay time T delay (predetermined time) in the switching of the first inverter 18 and the second inverter 19 is the sum of the dead time DT and the additional time T p (T delay=DT + T p ) is longer than the dead time DT. Further, the ASC switching processing unit 44 sets the additional time T so that the balance between the charging amount and the discharging amount of the smoothing capacitor 39 in the transition state becomes "discharge" (charging amount < discharging amount). p is determined according to the zero-phase current i0.

[0092] Note that, as described above, the duty command values (Duty_UVW1, UVW2) for causing the transition from the start state through the transition state to the end state change according to the combination of the rise or fall of the carrier signal CS and the change mode of the final ASC switching signal S UL2 Therefore, as shown in Table 2 above, the ASC switching processing unit 44 calculates the duty command values (Duty_UVW1, Duty_UVW2) based on the rise or fall of the carrier signal CS and the final ASC switching signal S UL2 In addition, in Table 2 above, the duty command value (Duty_UVW2) of the second inverter 19 that is switched in advance when forming the transition state is set to 50%, but this is just an example. As long as the switching of the first inverter 18 with respect to the second inverter 19 can be delayed by a predetermined time (T

[0093] ), the duty command value (Duty_UVW2) of the second inverter 19 that is switched in advance to form the transition state can be set to any value other than 50%. delay In addition, in the second ASC mode (ASC2), the duty command values (Duty_UVW1, Duty_UVW2) of the first inverter 18 and the second inverter 19 are equal. In this embodiment, both are 50%. That is, the delay time T

[0094] (predetermined time) in the switching between the first inverter 18 and the second inverter 19 is zero (T delay (predetermined time) is zero (T delayTherefore, in the second ASC mode, the first inverter 18 and the second inverter 19 are switched at similar timings. Note that the "similar timings" mentioned here refers to the relationship between the first inverter 18 and the second inverter 19. In terms of the relationship between the upper arm and the lower arm, the dead time DT is provided in the second ASC mode as described above.

[0095] Hereinafter, the operation of the ASC control performed in the electric vehicle 100 configured as described above will be described.

[0096] FIG. 6 is a flowchart related to the ASC control. As shown in FIG. 6, in step S10, the controller 13 acquires the phase currents i u , i v , i w , the DC voltage V dc , and the electrical angle θ. Further, the controller 13 calculates the rotational speed N of the rotating electrical machine 10 based on the electrical angle θ by the rotational speed calculation unit 27. In step S11, the abnormality detection unit 28 detects the presence or absence of an abnormality in the system that drives the rotating electrical machine 10, such as over-rotation or over-current.

[0097] In step S11, if no abnormality is detected, the process proceeds to step S12, and the PWM control unit 24 calculates the duty command values Duty_UVW1 and Duty_UVW2 for the normal mode that drives the rotating electrical machine 10. Then, the PWM signal generator 25 generates PWM signals D uu1 * ~D wl1 * , D uu2 * ~D wl2 * according to the duty command values Duty_UVW1 and Duty_UVW2 for the normal mode, and drives the rotating electrical machine 10. Thereby, the rotation of the rotating electrical machine 10 is controlled so as to output torque corresponding to the torque command value T * .

[0098] On one hand, in step S11, when an abnormality such as over-rotation or overcurrent is detected, the abnormality detection unit 28 outputs a relay cut-off signal S RC to turn off the relay 16 and proceed to step S13. In step S13, the output determination unit 35 compares the absolute value |N| of the rotation speed N with the rotation speed upper limit value N MAX .

[0099] In step S13, when the absolute value |N| of the rotation speed N is less than or equal to the rotation speed upper limit value N MAX , it proceeds to step S14, and the output determination unit 35 sets the gate permission signal S GP to "prohibited". Therefore, the ASC control is not performed. Even if any abnormality occurs during the driving of the rotating electrical machine 10, when the absolute value |N| of the rotation speed N is less than or equal to the rotation speed upper limit value N MAX , the induced voltage of the rotating electrical machine 10 is not so large, so the system can be shut down without performing the ASC control.

[0100] On the other hand, in step S13, when the absolute value |N| of the rotation speed N is greater than the rotation speed upper limit value N MAX , the ASC control is executed. Specifically, it proceeds to step S15, and the up-down switching signal generation unit 41 generates an ASC switching signal S UL1 for switching between the upper short-circuit control and the lower short-circuit control. In step S16, the timing adjustment processing unit 43 corrects the ASC switching signal S UL1 so that the switching between the upper short-circuit control and the lower short-circuit control is performed at the timing when the zero-phase current i0 is greater than zero, thereby generating the final ASC switching signal S UL2 .

[0101] In step S17, the output determination unit 35 compares the DC voltage V dc with the voltage threshold value V th . In step S17, when the DC voltage V dc is greater than the voltage threshold value V th , the duty command value generation mode S modeIt becomes the first ASC mode (ASC1). Therefore, in the subsequent step S18, the ASC switching processing unit 44 sets the delay time T delay in the switching of the first inverter 18 to the sum of the dead time DT and the additional time T p .

[0102] In step S17, when the DC voltage V dc is equal to or lower than the voltage threshold V th , the duty command value generation mode S mode becomes the second ASC mode. Therefore, in the subsequent step S19, the ASC switching processing unit 44 sets the delay time T delay in the switching of the first inverter 18 to zero.

[0103] As described above, based on the DC voltage V dc , the duty command value generation mode S mode is determined, and when the corresponding delay time T delay is set, in step S20, the ASC switching processing unit 44 generates the first duty command value Duty_UVW1 and the second duty command value Duty_UVW2, which are the duty command values for ASC. Also, in step S21, the output determination unit 35 sets the gate permission signal S GP to "permit". Thereby, ASC control is executed in either the first ASC mode or the second ASC mode.

[0104] FIG. 7 is a time chart showing an example of the PWM signals D uu1 * ~D wl1 * , D uu2 * ~D wl2 * in the first ASC mode (ASC1). In FIG. 7, for four control periods P1 to P4, the PWM signals D uu1 * ~D wl1 * , D uu2 * ~D wl2 *is shown. In FIG. 7, at time t0 when switching from control period P1 to control period P2, the ASC switching signal S UL2 is assumed to change from on (upper short - circuit control) to off (lower short - circuit control). Also, in the scene shown in FIG. 7, the U - phase current i u is a positive value and flows from the first inverter 18 through the U - phase stator coil (U) to the second inverter 19, and the V - phase current i v and the W - phase current i w are negative values and flow from the second inverter 19 through the stator coils (V, W) of each phase to the first inverter 18.

[0105] As shown in FIG. 7, when the ASC switching signal S UL2 changes from on to off at time t0, the ASC switching processing unit 44 generates duty command values Duty_UVW1 and Duty_UVW2 for forming a transition state in the switching from upper short - circuit control to lower short - circuit control during control period P2.

[0106] From the rising / falling timings of the ASC switching signal SUL2 and the carrier signal CS, here, the ASC switching processing unit 44 calculates the first duty command value Duty_UVW1 as Duty_UVW1 = 50+100×(DT + T p ) / τ [%], and sets the second duty command value Duty_UVW2 as Duty_UVW2 = 50 [%]. Therefore, the first duty command value Duty_UVW1 is larger than the second duty command value Duty_UVW2 (Duty_UVW1 > Duty_UVW2). These duty command values Duty_UVW1 and Duty_UVW2 are reflected in the generation of the PWM signals D uu1 * ~D wl1 * ,D uu2 * ~D wl2 * in the next control period P3.

[0107] During the control period P3, the PWM signal generator 25 generates the PWM signals D uu1 * ~D wl1 * ,D uu2 * ~D wl2 * by comparing and matching the above duty command values Duty_UVW1, Duty_UVW2 and the carrier signal CS. However, during the control period P3, the carrier signal CS rises, and the second duty command value Duty_UVW2 is smaller than the first duty command value Duty_UVW1. Therefore, the PWM signals D vu2 * ~D wl2 * for the second inverter 19 switch first, and then, when a predetermined time (DT + T p ) has elapsed, the PWM signals D uu1 * ,D ul1 * for the first inverter 18 switch. Also, the PWM signal generator 25 inserts a dead time DT for the PWM signals D ul1 * ,D vl1 * ,D wl1 * ,D ul2 * ,D vl2 * ,D wl2 * of the arm switching from off to on.

[0108] Specifically, first, at time t a , the PWM signals D 11 ) for the upper arm (Q7, Q9, Q uu2 * ,D vu2 * ,D wu2 * for the second inverter 19 switch from on to off, and the switching of the second inverter 19 starts. Then, at time t a when the dead time DT has elapsed from time t b , the lower arm (Q8, Q10 , Q 12 PWM signal D for ul2 * , D vl2 * , D wl2 * turns from off to on, and the switching of the second inverter 19 is completed. After that, an additional time T p elapses at time t c At this time, the PWM signal D for the upper arms (Q1, Q3, Q5) of the first inverter 18 uu1 * , D vu1 * , D wu1 * turns from on to off, and the switching of the first inverter 18 is started. And at time t c when the dead time DT has elapsed from time t d At this time, the PWM signal D for the lower arms (Q2, Q4, Q6) of the first inverter 18 ul1 * , D vl1 * , D wl1 * turns from off to on, and the switching of the first inverter 18 is completed.

[0109] That is, in the first ASC mode, when switching from the upper short - circuit control to the lower short - circuit control, with respect to the initial state (the state before time t a ) in which the upper short - circuit control is being performed, a transition state is formed in which the switching of the second inverter 19 is started first while maintaining the state of the first inverter 18 the same as the initial state (the state from time t a to t d ). Also, after a predetermined time (T delay = DT + T p ) has elapsed since the switching of the second inverter 19 was started first, the switching of the first inverter 18 is started (at time t c ). And when the switching of the first inverter 18 is completed (at time t d) The upper arm and the lower arm of the first inverter 18 are reversed with respect to the initial state, and a final state (a state in which lower-side short-circuit control is being performed) is formed in which the upper arm and the lower arm of the second inverter 19 are reversed with respect to the initial state.

[0110] In particular, the transition state is at time t a ~t b in the first transition state, at time t b ~t c in the second transition state, and at time t c ~t d in the third transition state. The first transition state is a state in which all arms of the second inverter 19 are off while the upper arm of the first inverter 18 is on and the lower arm is off. The second transition state is a state in which the upper arm of the second inverter 19 is off and the lower arm is on while the upper arm of the first inverter 18 is on and the lower arm is off. The third transition state is a state in which all arms of the first inverter 18 are off while the upper arm of the second inverter 19 is off and the lower arm is on.

[0111] FIG. 8 is a time chart showing an example of the PWM signals D uu1 * ~D wl1 * , D uu2 * ~D wl2 * in the second ASC mode (ASC2). In FIG. 8, the PWM signals D uu1 * ~D wl1 * , D uu2 * ~D wl2 * are shown in a scene similar to that of FIG. 7 except that the second ASC mode is selected.

[0112] As shown in FIG. 8, when the second ASC mode is selected, the duty command values Duty_UVW1 and Duty_UVW2 for the first inverter 18 and the second inverter 19 in the control period P3 are both 50 [%]. Also, the delay time T delay related to the switching of the first inverter 18 is zero. Therefore, the PWM signals D 11 ) for the upper arms (Q7, Q9, Q uu2 * , D vu2 * , D wu2 * of the second inverter 19 turn from on to off at time t a , and the PWM signals D uu1 * , D vu1 * , D wu1 * for the upper arms (Q1, Q3, Q5) of the first inverter 18 turn from on to off at time t c , and they are substantially at the same time (t a = t c ). As a result, with a dead time DT, the PWM signals D 11 ) for the upper arms (Q7, Q9, Q ul2 * , D vl2 * , D wl2 * of the second inverter 19 turn from on to off at time t b , and the PWM signals D ul1 * , D vl1 * , D wl1 * for the lower arms (Q2, Q4, Q6) of the first inverter 18 turn from on to off at time t d , and they are also substantially at the same time (t a = t c ).

[0113] Therefore, in the second ASC mode, when switching between the upper short - circuit control and the lower short - circuit control, the start state (time t aFrom the previous state, the switching of the second inverter 19 and the first inverter 18 starts simultaneously, and after the elapse of the dead time DT, the final state (time t d The subsequent state) is formed.

[0114] That is, in the first ASC mode, a transition state is formed by delaying the switching of the first inverter 18, whereas in the second ASC mode, a transition state is not formed, which is different.

[0115] FIG. 9 is an explanatory diagram showing the transition of the phase currents i u , i v , i w in the first ASC mode (ASC1). In FIG. 9, the U-phase current i u flows from the first inverter 18 through the U-phase stator coil (U) to the second inverter 19, and the V-phase current i v and the W-phase current i w flow from the second inverter 19 through the stator coils (V, W) of each phase to the first inverter 18, showing a scene of switching from the upper short-circuit control to the lower short-circuit control.

[0116] Specifically, FIG. 9(A) shows the initial state in which the upper short-circuit control is being performed. That is, in FIG. 9(A), all the upper arms (Q1, Q3, Q5) of the first inverter 18 are on, all the lower arms (Q2, Q4, Q6) of the first inverter 18 are off, all the upper arms (Q7, Q9, Q 11 ) of the second inverter 19 are on, and all the lower arms (Q8, Q 10 , Q 12 ) of the second inverter 19 are off, showing the flow of the phase currents i u , i v , i w .

[0117] FIG. 9(B) shows the first transition state. That is, in FIG. 9(B), while maintaining the switching state of all the arms (Q1~Q6) of the first inverter 18 in the switching state in the initial state, all the arms (Q7~Q 12The phase current i when u , i v , i w flows is shown.

[0118] Fig. 9(C) shows the second transition state. That is, in Fig. 9(C), with the switching states of all the arms (Q1 to Q6) of the first inverter 18 maintained in the switching states in the initial state, all the upper arms (Q7, Q9, Q 11 ) turn off, and all the lower arms (Q8, Q 10 , Q 12 ) of the second inverter 19 turn on, the phase current i u , i v , i w flows is shown.

[0119] Fig. 9(D) shows the third transition state. That is, in Fig. 9(D), with all the upper arms (Q7, Q9, Q 11 ) off and all the lower arms (Q8, Q 10 , Q 12 ) of the second inverter 19 on, when all the arms (Q1 to Q6) of the first inverter 18 turn off, the phase current i u , i v , i w flows is shown.

[0120] And Fig. 9(E) shows the final state in which the lower short - circuit control is being performed. Therefore, in Fig. 9(E), all the upper arms (Q1, Q3, Q5) of the first inverter 18 are off, all the lower arms (Q2, Q4, Q6) of the first inverter 18 are on, all the upper arms (Q7, Q9, Q 11 ) of the second inverter 19 are off, and all the lower arms (Q8, Q 10 , Q 12 ) of the second inverter 19 are on. That is, in Fig. 9(E), the switching states of all the arms (Q1 to Q6) of the first inverter 18 are reversed with respect to the initial state, and the switching states of all the arms (Q7 to Q 12 ) of the second inverter 19 are reversed with respect to the initial state, the phase current i u, i v , i w shows the flow of

[0121] As shown in Fig. 9(A), in the initial state where upper short - circuit control is being performed, the U - phase current i u flows from the upper arm (Q1) of the first inverter 18 into the U - phase stator coil (U) and circulates through the free - wheeling diode of the upper arm (Q7) of the second inverter 19. On the other hand, the V - phase current i v flows from the upper arm (Q7) of the second inverter 19 into the V - phase stator coil (V) and circulates through the free - wheeling diode of the upper arm (Q3) of the first inverter 18. Similarly, the W - phase current i w flows from the upper arm (Q 11 ) of the second inverter 19 into the W - phase stator coil (W) and circulates through the free - wheeling diode of the upper arm (Q5) of the first inverter 18. Also, at this time, the DC current component flowing through the smoothing capacitor 39 (hereinafter simply referred to as DC current I dc ) is zero (I dc = 0).

[0122] Therefore, in the initial state, the phase currents i u , i v , i w generated by the induced voltage of the rotating electrical machine 10 are consumed by circulating through the inverter 12 and the rotating electrical machine 10, and the smoothing capacitor 39 is neither discharged nor charged.

[0123] As shown in Fig. 9(B), in the first transition state, the U - phase current i u flows from the upper arm (Q1) of the first inverter 18 into the U - phase stator coil (U) and circulates through the free - wheeling diode of the upper arm (Q7) of the second inverter 19. On the other hand, the V - phase current i v flows from the free - wheeling diode of the lower arm (Q 10 ) of the second inverter 19 into the V - phase stator coil (V) and flows through the upper arm (Q3) of the first inverter 18. Similarly, the W - phase current i w flows from the lower arm (Q 12) flows from the freewheeling diode of v to the W-phase stator coil (W) and flows through the upper arm (Q5) of the first inverter 18. Therefore, the V-phase current i w and the W-phase current i

[0124] flow into the smoothing capacitor 39. u , i v , i w are consumed by circulating through the inverter 12 and the rotating electrical machine 10, while charging the smoothing capacitor 39.

[0125] The direct current I dc in the first transition state is generally expressed by the following equation (9).

[0126]

Equation

[0127] And the direction in which current flows out of the smoothing capacitor 39 is the positive direction of the direct current I dc , and the direction in which current flows into the smoothing capacitor 39 is the negative direction of the direct current I dc . That is, when the direct current I dc is positive, the smoothing capacitor 39 discharges, and when the direct current I dc is negative, the smoothing capacitor 39 is charged. And when the direct current I dc is zero, the smoothing capacitor 39 is neither discharged nor charged. In the scene shown in Fig. 9(B), specifically, I dc = -i u = -i v - i w . Therefore, as described above, in the first transition state of Fig. 9(B), the smoothing capacitor 39 is charged.

[0128] As shown in Fig. 9(C), in the second transition state, the U-phase current i uflows from the upper arm (Q1) of the first inverter 18 into the U-phase stator coil (U), passes through the lower arm (Q8) of the second inverter 19, and becomes the V-phase current i v and the W-phase current i w . The V-phase current i v flows from the freewheeling diode of the lower arm (Q 10 ) of the second inverter 19 into the V-phase stator coil (V), passes through the freewheeling diode of the upper arm (Q3) of the first inverter 18, and becomes the U-phase current i u . Similarly, the W-phase current i w flows through the freewheeling diode of the lower arm (Q 12 ) of the second inverter 19 into the W-phase stator coil (W), passes through the upper arm (Q5) of the first inverter 18, and becomes the U-phase current i u . That is, in the second transition state, the phase currents i u , i v , i w circulate through the upper arms (Q1, Q3, Q5) of the first inverter 18, the stator coils (U, V, W) of each phase, and the lower arms (Q8, Q 10 , Q 12 ) of the second inverter 19. Therefore, a part of the phase currents i u , i v , i w generated by the induced voltage of the rotating electrical machine 10 is consumed by circulating through the inverter 12 and the rotating electrical machine 10.

[0129] Also, when the phase currents i u , i v , i w circulate as described above, the DC current I dc in the smoothing capacitor 39 is equal to the zero-phase current i0 (I dc = i0). And in this embodiment, the ASC switching signal S UL2 is adjusted so that the switching between the upper short-circuit control and the lower short-circuit control is performed when the zero-phase current i0 is greater than zero (i0 > 0). Therefore, in the second transition state, since the DC current I dc becomes positive, the smoothing capacitor 39 is discharged.

[0130] As shown in FIG. 9(D), in the third transition state, the U-phase current i u flows from the freewheeling diode of the lower arm (Q2) of the first inverter 18 into the U-phase stator coil (U), passes through the lower arm (Q8) of the second inverter 19, and becomes the V-phase current i v and the W-phase current i w . The V-phase current i v flows from the freewheeling diode of the lower arm (Q 10 ) of the second inverter 19 into the V-phase stator coil (V), and then flows into the smoothing capacitor 39 through the freewheeling diode of the upper arm (Q3) of the first inverter 18. Similarly, the W-phase current i w flows from the freewheeling diode of the lower arm (Q 12 ) of the second inverter 19 into the W-phase stator coil (W), and then flows into the smoothing capacitor 39 through the freewheeling diode of the upper arm (Q5) of the first inverter 18.

[0131] Also, the DC current I dc in the third transition state is expressed by the aforementioned equation (9) as in the case of the first transition state. In the scene shown in FIG. 9(D), specifically, I dc =-i u =-i v -i w . Therefore, in the third transition state of FIG. 9(D), the smoothing capacitor 39 is charged.

[0132] Therefore, in the third transition state, the phase currents i u , i v , i w generated by the induced voltage of the rotating electrical machine 10 are consumed while circulating through the inverter 12 and the rotating electrical machine 10, and at the same time, the smoothing capacitor 39 is charged.

[0133] As shown in FIG. 9(E), in the final state switched to the lower short-circuit control, the U-phase current i u flows from the freewheeling diode of the lower arm (Q2) of the first inverter 18 into the U-phase stator coil (U), passes through the lower arm (Q8) of the second inverter 19, and becomes the V-phase current i vand the W-phase current i w is obtained. And the V-phase current i v flows from the freewheeling diode of the lower arm (Q 10 ) of the second inverter 19 into the V-phase stator coil (V), passes through the lower arm (Q4) of the first inverter 18, and becomes the U-phase current i u . Similarly, the W-phase current i w flows from the freewheeling diode of the lower arm (Q 12 ) of the second inverter 19 into the W-phase stator coil (W), passes through the lower arm (Q6) of the first inverter 18, and becomes the U-phase current i u . Also, at this time, the direct current I dc flowing through the smoothing capacitor 39 is zero (I dc = 0).

[0134] Therefore, in the final state, the phase currents i u , i v , i w generated by the induced voltage of the rotating electrical machine 10 are consumed by circulating through the inverter 12 and the rotating electrical machine 10, and the smoothing capacitor 39 is neither discharged nor charged.

[0135] Here, when the U-phase current i u flows from the first inverter 18 through the U-phase stator coil (U) to the second inverter 19, and the V-phase current i v and the W-phase current i w flow from the second inverter 19 through the stator coils (V, W) of each phase to the first inverter 18, a scene of switching from the upper short-circuit control to the lower short-circuit control is shown, but the same applies to other scenes. Therefore, in the switching between the upper short-circuit control and the lower short-circuit control, the state of the smoothing capacitor 39 changes as follows.

[0136] FIG. 10 is a graph illustrating the transition of the direct current I dc in the first ASC mode. As shown in FIG. 10, when switching between the upper short-circuit control and the lower short-circuit control, at time t aIn the initial state where upper short - circuit control (or lower short - circuit control) has been performed previously, the direct - current I flowing through the smoothing capacitor 39 dc is zero. Therefore, the smoothing capacitor 39 is neither discharged nor charged.

[0137] And at time t a when all the arms (Q7 - Q 12 ) of the second inverter 19 turn off and transition to the first transition state, the direct - current I dc becomes a negative value. Therefore, during the first transition state, the smoothing capacitor 39 is charged.

[0138] At time t a when the dead - time DT has elapsed from time t b to time t 10 and the lower arms (Q8, Q 12 ) of the second inverter 19 turn on and transition to the second transition state, the direct - current I dc becomes a positive value. Therefore, during the second transition state, the smoothing capacitor 39 is discharged.

[0139] Next, at time t b when the additional time T p has elapsed from time t c to time t dc and all the arms (Q1 - Q6) of the first inverter 18 turn off and transition to the third transition state, the direct - current I

[0140] becomes a negative value. Therefore, during the third transition state, the smoothing capacitor 39 is charged again. c After that, at time t d when the dead - time DT has elapsed from time t dc and the lower arms (Q2, Q4, Q6) of the first inverter 18 turn on and transition to the final state, the direct - current I

[0141] becomes zero. Therefore, in the final state, similar to the initial state, the smoothing capacitor 39 is neither discharged nor charged.Therefore, if the charge amount of the smoothing capacitor 39 in the first transition state is -P1, the discharge amount of the smoothing capacitor 39 in the second transition state is +P2, and the charge amount of the smoothing capacitor 39 in the third transition state is -P3, the balance of the discharge and charge of the smoothing capacitor 39 in the switching between the upper short-circuit control and the lower short-circuit control is -P1 + P2 - P3. Therefore, if -P1 + P2 - P3 > 0, the smoothing capacitor 39 is net-discharged in the switching between the upper short-circuit control and the lower short-circuit control. Therefore, the additional time T is determined so that the charge amounts (-P1 - P3) in the first transition state and the third transition state are offset by the discharge amount (+P2) in the second transition state and -P1 + P2 - P3 > 0. p is determined.

[0142] As described above, the direct current I in the first transition state and the third transition state dc is represented by Equation (9), and the direct current I in the second transition state dc becomes equal to the zero-phase current i0. The zero-phase current i0 is represented by using the direct current voltage V dc and the resistance R and inductance L of the stator coils (U, V, W). Therefore, the condition for the smoothing capacitor 39 to be net-discharged in the switching between the upper short-circuit control and the lower short-circuit control is expressed by the following Equation (11) using the dead time DT and the additional time T p . Here, it is assumed that the changes in the phase currents i u , i v , i w in the first transition state and the third transition state are small and substantially constant. Also, i0(0) represents the initial value of the zero-phase current i0 in the second transition state, that is, the zero-phase current i0 at the time t b .

[0143]

Equation

[0144] Therefore, the ASC switching processing unit 44 determines the additional time T according to the zero-phase current i0 so as to satisfy Equation (11). pis determined. In particular, in the present embodiment, more simply, assuming that the change in the zero-phase current i0 in the second transition state is also small, the ASC switching processing unit 44 determines the additional time T by the following equation (12). p is determined. The predetermined coefficient k and the offset current i in equation (12). ofs are determined in advance by adaptation so that the additional time T p satisfies equation (11), based on experiments, simulations, or the like.

[0145]

Equation

[0146] As is clear from equation (12), the additional time T p is equal to or greater than zero, and in principle, the additional time T p is greater than zero as long as no abnormalities occur. Therefore, the delay time T delay = DT + T p relating to the switching of the first inverter 18 in the first ASC mode is longer than the dead time DT.

[0147] Also, in principle, the additional time T p is set to be longer as the magnitude |i0| of the zero-phase current i0 is smaller. This is to cancel out the charging of the smoothing capacitor 39 in the first and third transition states and the discharging of the smoothing capacitor 39 in the second transition state. Therefore, the delay time T delay = DT + T p relating to the switching of the first inverter 18 in the first ASC mode is set to be longer as the magnitude |i0| of the zero-phase current i0 is smaller.

[0148] As described above, in the first ASC mode, when switching between the upper short-circuit control and the lower short-circuit control, the charging of the smoothing capacitor 39 in the first transition state and the third transition state and the discharging of the smoothing capacitor 39 in the second transition state cancel each other out, and as a result, the smoothing capacitor 39 is discharged. That is, the first ASC mode is an ASC mode that discharges the smoothing capacitor 39 by switching between the upper short-circuit control and the lower short-circuit control. Therefore, in the present embodiment, the second ASC mode is selected in a scene where the DC voltage V dc is high (V dc > V th ) and it is preferable to discharge the smoothing capacitor 39.

[0149] Here, in the transition state of the first ASC mode, the additional time T p is determined so that the smoothing capacitor 39 is net-discharged. However, in the first ASC mode, the additional time T p may be set so that the smoothing capacitor 39 is neither substantially discharged nor charged. In this case, the first ASC mode is an ASC mode that maintains the power stored in the smoothing capacitor 39. Therefore, more precisely, the first ASC mode is an ASC mode that can discharge or maintain the power stored in the smoothing capacitor 39 when the DC voltage V dc is high (V dc > V th ).

[0150] FIG. 11 is an explanatory diagram showing the transition of the phase currents i u , i v , i w in the second ASC mode (ASC2). In FIG. 11, the U-phase current i u flows from the first inverter 18 through the U-phase stator coil (U) to the second inverter 19, and the V-phase current i v and the W-phase current i w flow from the second inverter 19 through the stator coils (V, W) of each phase to the first inverter 18, showing a scene of switching from the upper short-circuit control to the lower short-circuit control.

[0151] Specifically, FIG. 11(A) shows the initial state in which upper-side short-circuit control is being performed. That is, in FIG. 11(A), all the upper arms (Q1, Q3, Q5) of the first inverter 18 are on, all the lower arms (Q2, Q4, Q6) of the first inverter 18 are off, all the upper arms (Q7, Q9, Q 11 ) of the second inverter 19 are on, and when all the lower arms (Q8, Q 10 , Q 12 ) of the second inverter 19 are off, the phase currents i u , i v , i w are shown flowing.

[0152] FIG. 11(B) shows the phase currents i u , i v , i w when all the arms (Q1 to Q12) of the first inverter 18 and the second inverter 19 are turned off due to the dead time DT.

[0153] And FIG. 11(C) shows the final state in which lower-side short-circuit control is being performed. Therefore, in FIG. 11(C), all the arms (Q1, Q3, Q5) of the first inverter 18 are off, all the lower arms (Q2, Q4, Q6) of the first inverter 18 are on, all the upper arms (Q7, Q9, Q 11 ) of the second inverter 19 are off, and all the lower arms (Q8, Q 10 , Q 12 ) of the second inverter 19 are on. That is, in FIG. 11(C), the switching states of all the arms (Q1 to Q6) of the first inverter 18 are reversed with respect to the initial state, and when the switching states of all the arms (Q7 to Q 12 ) of the second inverter 19 are reversed with respect to the initial state, the phase currents i u , i v , i w are shown flowing.

[0154] As shown in FIG. 11(A), in the initial state in which upper-side short-circuit control is being performed, even in the second ASC mode, the U-phase current i uflows from the upper arm (Q1) of the first inverter 18 into the U-phase stator coil (U) and circulates through the freewheeling diode of the upper arm (Q7) of the second inverter 19. On the other hand, the V-phase current i v flows from the upper arm (Q7) of the second inverter 19 into the V-phase stator coil (V) and circulates through the freewheeling diode of the upper arm (Q3) of the first inverter 18. Similarly, the W-phase current i w flows from the upper arm (Q 11 ) of the second inverter 19 into the W-phase stator coil (W) and circulates through the freewheeling diode of the upper arm (Q5) of the first inverter 18. At this time, the direct current I dc flowing through the smoothing capacitor 39 is zero (I dc = 0).

[0155] Therefore, in the initial state, the phase currents i u , i v , i w generated by the induced voltage of the rotating electrical machine 10 are consumed by circulating through the inverter 12 and the rotating electrical machine 10, and the smoothing capacitor 39 is neither discharged nor charged.

[0156] Then, in the second ASC mode, in order to switch between the upper short-circuit control and the lower short-circuit control, the first inverter 18 and the second inverter 19 are switched simultaneously. That is, the delay time T delay related to the switching of the first inverter 18 in the second ASC mode is zero (T delay = 0). However, in the relationship between the upper arm and the lower arm of the same phase, a dead time DT is provided. For this reason, as shown in FIG. 11(B), a state occurs in which all the arms (Q1 to Q 12 ) of the first inverter 18 and the second inverter 19 are turned off.

[0157] At this time, the U-phase current i u flows from the freewheeling diode of the lower arm (Q2) of the first inverter 18 into the U-phase stator coil (U), passes through the freewheeling diode of the upper arm (Q7) of the second inverter 19, and flows into the smoothing capacitor 39. The V-phase current iv flows from the freewheeling diode of the lower arm (Q 10 ) of the second inverter 19 into the V-phase stator coil (W), and then flows into the smoothing capacitor 39 through the freewheeling diode of the upper arm (Q3) of the first inverter 18. Similarly, the W-phase current i w flows from the freewheeling diode of the lower arm (Q 12 ) of the second inverter 19 into the W-phase stator coil (W), and then flows into the smoothing capacitor 39 through the freewheeling diode of the upper arm (Q5) of the first inverter 18.

[0158] At this time, the DC current I dc is expressed by the following formula (13). That is, in the second ASC mode, when all the arms (Q1 to Q 12 ) of the first inverter 18 and the second inverter 19 are turned off, the DC current I dc is negative. Therefore, the phase currents i u , i v , i w (zero-phase current i0) charge the smoothing capacitor 39.

[0159]

Equation

[0160] As shown in Fig. 11(C), also in the second ASC mode, in the final state switched to the lower short-circuit control, the U-phase current i u flows from the freewheeling diode of the lower arm (Q2) of the first inverter 18 into the U-phase stator coil (U), and then becomes the V-phase current i v and the W-phase current i w through the lower arm (Q8) of the second inverter 19. And the V-phase current i v flows from the freewheeling diode of the lower arm (Q 10 ) of the second inverter 19 into the V-phase stator coil (V), and then becomes the U-phase current i u through the lower arm (Q4) of the first inverter 18. Similarly, the W-phase current i w flows from the freewheeling diode of the lower arm (Q12 ) flows from the freewheeling diode of to the W-phase stator coil (W), passes through the lower arm (Q6) of the first inverter 18, and becomes the U-phase current i u . At this time, the direct current I flowing through the smoothing capacitor 39 dc is zero (I dc = 0).

[0161] Therefore, in the final state, the phase currents i u , i v , i w generated by the induced voltage of the rotating electrical machine 10 are consumed by circulating through the inverter 12 and the rotating electrical machine 10, and the smoothing capacitor 39 is neither discharged nor charged.

[0162] As described above, the second ASC mode is an ASC mode in which the smoothing capacitor 39 is charged when switching between the upper short-circuit control and the lower short-circuit control. Therefore, in the present embodiment, the second ASC mode is selected in a scene where it is desirable to charge the smoothing capacitor 39 in view of a low DC voltage V dc (V dc ≤ V th ) and for restarting the system (recharging the smoothing capacitor 39) and the like.

[0163] If the conventional ASC control is used, the upper short-circuit control and the lower short-circuit control are always switched by the above-described second ASC mode. For this reason, even in a scene where it is desirable to discharge the smoothing capacitor 39 when the DC voltage V dc is high (V dc > V th ) or in a scene where it is desirable neither to charge nor discharge the smoothing capacitor 39, the smoothing capacitor 39 is charged. Therefore, in these scenes, the smoothing capacitor 39 may be overcharged and damaged.

[0164] On the other hand, the ASC control of the present embodiment provides a transition state (first to third transition states) when switching between the upper short-circuit control and the lower short-circuit control in the first ASC mode, so that the phase current i generated by the induced voltage of the rotating electrical machine 10 u, i v , i w while consuming it in a short - circuit formed by the inverter 12 and the rotating electrical machine 10, and discharging (or maintaining) the smoothing capacitor 39. And the second ASC mode is the phase current i u , i v , i w while consuming it in a short - circuit formed by the inverter 12 and the rotating electrical machine 10, and charging the smoothing capacitor 39 is selected only in a desirable scene. Therefore, in the ASC control of this embodiment, the smoothing capacitor 39 is protected when switching between the upper - side short - circuit control and the lower - side short - circuit control.

[0165] [Second Embodiment] As described in the above - mentioned first embodiment, the direct - current current I dc in the second transition state of the first ASC mode is equal to the zero - phase current i0 (I dc = i0). And in the first embodiment, the switching timing is adjusted so that the switching between the upper - side short - circuit control and the lower - side short - circuit control is performed when the zero - phase current i0 is greater than zero. Thereby, in the first embodiment, if the second inverter 19 is switched first (the first inverter 18 is switched later), it is ensured that the smoothing capacitor 39 is discharged (or maintained) in the second transition state.

[0166] However, in the second transition state, the method of making the direct - current current I dc (= zero - phase current i0) positive and discharging the smoothing capacitor 39 is not limited to this. Specifically, since the first inverter 18 and the second inverter 19 have a symmetrical configuration with respect to the rotating electrical machine 10, by changing which of the first inverter 18 and the second inverter 19 is switched first (later) according to the zero - phase current i0, in the second transition state, the direct - current current I dc can be made positive and the smoothing capacitor 39 can be discharged.

[0167] Hereinafter, in the second embodiment, an example of changing which of the first inverter 18 and the second inverter 19 is switched first (delayed) according to the zero-phase current i0 in the first ASC mode will be described.

[0168] FIG. 12 is a block diagram of the first PWM control unit 33 and the second PWM control unit 34 according to the second embodiment. FIG. 12 shows the configuration of the part related to the ASC control in particular.

[0169] As shown in FIG. 12, in the second embodiment, the first PWM control unit 33 and the second PWM control unit 34 include an upper and lower switching signal generation unit 41, a three-phase sum processing unit 42, and an ASC switching processing unit 44, and the timing adjustment processing unit 43 is omitted.

[0170] The upper and lower switching signal generation unit 41 and the three-phase sum processing unit 42 have the same configuration as in the first embodiment. That is, the upper and lower switching signal generation unit 41 generates an ASC switching signal S that switches the timing of the upper short-circuit control and the lower short-circuit control. UL1 Also, the three-phase sum processing unit 42 calculates the zero-phase current i0 that is the sum of the phase currents i u , i v , i w .

[0171] The ASC switching processing unit 44 generates duty command values (Duty_UVW1, Duty_UVW2) for the ASC mode based on the carrier signal CS and the duty command value generation mode S mode . Also, the ASC switching processing unit 44 determines an additional time T p according to the zero-phase current i0 so that the balance between the charge amount and the discharge amount of the smoothing capacitor 39 becomes "discharge" (charge amount < discharge amount) in the transition state. These are the same as in the first embodiment.

[0172] However, in the second embodiment, instead of the final ASC switching signal S UL2 , the ASC switching processing unit 44 uses the basic ASC switching signal S generated by the upper and lower switching signal generation unit 41 UL1Refer to it. Also, in the second embodiment, the ASC switching processing unit 44 uses the carrier signal CS, the basic ASC switching signal S UL1 , and in addition to the duty command value generation mode S mode , according to the zero-phase current i0, it generates duty command values (Duty_UVW1, Duty_UVW2) for the ASC mode. This is the difference from the first embodiment.

[0173] Specifically, in the second embodiment, the ASC switching processing unit 44, according to Table 3 below, uses the carrier signal CS, the basic ASC switching signal S UL1 , the duty command value generation mode S mode , and generates duty command values (Duty_UVW1, Duty_UVW2) for the ASC mode according to the zero-phase current i0.

[0174]

Table 3

[0175] As shown in Table 3 above, when the duty command value generation mode S mode is the first ASC mode (ASC1), the ASC switching processing unit 44 changes the inverter that delays or advances the switching depending on whether the zero-phase current i0 is greater than zero (i0>0) or the zero-phase current i0 is less than or equal to zero (i0≦0).

[0176] Specifically, when the zero-phase current i0 is greater than zero (i0>0) in the first ASC mode, the ASC switching processing unit 44 sets the second duty command value Duty_UVW2 for the second inverter 19 to 50 [%], and the first duty command value Duty-UVW1 for the first inverter 18 is calculated by 50±100×(DT+T p ) / τ [%]. Thereby, the ASC switching processing unit 44 relatively advances the switching of the second inverter 19 and relatively delays the switching of the first inverter 18. At this time, the delay time T delay related to the switching of the first inverter 18 is T delay = DT+Tp This is the same as the first embodiment.

[0177] On the other hand, when the zero-phase current i0 is zero or less in the first ASC mode (i0 ≦ 0), the ASC switching processing unit 44 sets the second duty command value Duty_UVW2 for the second inverter 19 to 50 ± 100 × (DT + T p ) / τ [%] for calculation, and sets the first duty command value Duty_UVW1 for the first inverter 18 to 50 [%]. Thereby, the ASC switching processing unit 44 relatively delays the switching of the second inverter 19 and relatively advances the switching of the first inverter 18. At this time, the delay time T delay related to the switching of the second inverter 19 is T delay = DT + T p This is the same as the first embodiment.

[0178] For other scenes such as the second ASC mode, it is the same as the first embodiment.

[0179] Also, the flow related to the ASC control of the second embodiment in which the first PWM control unit 33 and the second PWM control unit 34 are configured as described above is the same as the first embodiment (see FIG. 4) except that step S16 is omitted. However, step S20 is subdivided as follows.

[0180] FIG. 13 is a flowchart related to the calculation of the ASC control duty command values (Duty_UVW1, Duty_UVW2) in the second embodiment.

[0181] As shown in FIG. 13, when calculating the ASC control duty command values (Duty_UVW1, Duty_UVW2), the ASC switching processing unit 44 checks in step S31 whether the duty command value generation mode S mode is the first ASC mode (ASC1).

[0182] In step S31, whether the duty command value generation mode S modeWhen it is not in the first ASC mode, it proceeds to step S32, and the ASC switching processing unit 44 generates duty command values Duty_UVW1 and Duty_UVW2 for simultaneously switching the first inverter 18 and the second inverter 19 for switching between the upper short-circuit control and the lower short-circuit control. That is, in step S32, duty command values Duty_UVW1 and Duty_UVW2 for realizing the second ASC mode (ASC2) are generated.

[0183] In step S31, when the duty command value generation mode S mode When it is not in the first ASC mode, it proceeds to step S33, and the ASC switching processing unit 44 further checks whether the zero-phase current i0 is greater than zero (that is, the direction of the zero-phase current i0).

[0184] In step S33, when the zero-phase current i0 is greater than zero (i0 > 0), it proceeds to step S34, and the ASC switching processing unit 44 generates duty command values Duty_UVW1 and Duty_UVW2 for switching the second inverter 19 first (delaying the first inverter 18) in the same manner as in the first embodiment.

[0185] On the other hand, in step S33, when the zero-phase current i0 is zero or less (i0 ≤ 0), it proceeds to step S35, and the ASC switching processing unit 44 generates duty command values Duty_UVW1 and Duty_UVW2 for switching the first inverter 18 first (delaying the second inverter 19), contrary to the first embodiment.

[0186] As described above, when performing ASC control in the first ASC mode (ASC1), by switching the inverter to be switched first (delayed) depending on the direction of the zero-phase current i0, that is, whether the zero-phase current i0 is greater than zero, in the second transition state, the direct current I dc becomes positive. Therefore, the additional time T pIf it is appropriately determined in the same manner as in the first embodiment, the smoothing capacitor 39 is discharged (or maintained). Therefore, even if the switching timings of the upper short-circuit control and the lower short-circuit control are not adjusted, by switching the inverter to be switched earlier (or later) according to the direction of the zero-phase current i0, the same operational effects as in the first embodiment can be obtained.

[0187] Here, in the second transition state, when the zero-phase current i0 flows from the first inverter 18 to the second inverter 19 via the rotating electrical machine 10, the smoothing capacitor 39 is discharged. For this reason, the first inverter 18 is switched later and the second inverter 19 is switched earlier. However, depending on the specific connection relationships of the battery 11, the smoothing capacitor 39, the first inverter 18 and the second inverter 19, and the rotating electrical machine 10, in the second transition state, when the zero-phase current i0 flows from the second inverter 19 to the first inverter 18 via the rotating electrical machine 10, there may be a system in which the smoothing capacitor 39 is discharged. In this case, the first inverter 18 may be switched earlier and the second inverter 19 may be switched later.

[0188] [Modification Example] In the above-described first and second embodiments, overspeed and overcurrent are mainly detected as system abnormalities, and ASC control is executed when these abnormalities occur. However, as described above, for safety reasons, the relay 16 may be turned off and ASC control may be performed even when sensors such as the current sensor 14 malfunction. And the ASC control of the first and second embodiments can also be used when the sensors malfunction. However, in the above-described first and second embodiments, when generating the duty command values Duty_UVW1 and Duty_UVW2 for performing ASC control, it is premised that the phase currents i u , i v , i w can be detected normally (see FIGS. 4 and 12). Therefore, hereinafter, particularly when the current sensor 14 malfunctions, the phase currents i u , i v , i wis estimated, and using the estimated phase currents (hereinafter referred to as the estimated phase currents i u ′, i v ′, i w ′), a modification for performing the ASC control of the first embodiment or the second embodiment will be described.

[0189] FIG. 14 is a block diagram showing the configuration of the first PWM control unit 33 and the second PWM control unit 34 according to the modification. As shown in FIG. 14, the first PWM control unit 33 and the second PWM control unit 34 of the modification include a current selection unit 301 in addition to the up / down switching signal generation unit 41, the three-phase sum processing unit 42, the timing adjustment processing unit 43, and the ASC switching processing unit 44. In this modification, the state signal S output by the abnormality detection unit 28 state represents the states of "normal" and "abnormal", and when its status is "abnormal", it includes information indicating the location where the abnormality occurred or the content of the abnormality.

[0190] The current selection unit 301 checks whether the abnormality is due to a failure of the current sensor 14 when the state signal S obtained from the abnormality detection unit 28 state is "abnormal". When the state signal S state indicates "abnormal" due to a cause other than the failure of the current sensor 14, the current selection unit 301 acquires the phase currents i from the current sensor 14 u , i v , i w and inputs the detected values of the acquired phase currents i u , i v , i w to the three-phase sum processing unit 42.

[0191] On the other hand, when the cause for the state signal S state indicating "abnormal" includes the failure of the current sensor 14, the current selection unit 301 estimates the phase currents i based on the electrical angle θ u , i v , i w . Specifically, the current selection unit 301 estimates the phase currents i according to the following formula (14), based on the electrical angle θ, the magnet flux Φ a , and the d-axis inductance L dBased on this, the estimated phase currents i u ′, i v ′, i w ′ are calculated. Then, the current selection unit 301 inputs the estimated phase currents i u , i v , i w ′, i u ′, i v ′, i w ′ to the three-phase sum processing unit 42 instead of the detected values of the phase currents i

[0192]

Number

[0193] Note that since the magnet flux Φ a and the d-axis inductance L d are the inherent parameters of the rotating electrical machine 10, they are known parameters. Also, [-Φ a / L d , 0] in Equation (14) represents the dq-axis currents [i d , i q when the rotating electrical machine 10 rotates without power supply.

[0194] Also, in this modified example, a configuration is shown such that the first PWM control unit 33 and the second PWM control unit 34 of the first embodiment can execute the ASC control even when a fault occurs in the current sensor 14, but it is not limited to this. If the above current selection unit 301 is added to the first PWM control unit 33 and the second PWM control unit 34 of the second embodiment, the ASC control can also be executed in the second embodiment even when a fault occurs in the current sensor 14.

[0195] As described above, the inverter control method according to the first embodiment, the second embodiment, and the modification example is an inverter control method for controlling a first inverter 18 that connects one end of a stator coil (U, V, W) of an open-winding type rotating electrical machine (10) to a DC power supply (11) via a smoothing capacitor 39, and a second inverter 19 that connects the other end of the stator coil (U, V, W) to the DC power supply (11) via the smoothing capacitor 39. In this control method, the upper arms (Q1, Q3, Q5, Q7, Q9, Q 11 ) of the first inverter 18 and the second inverter 19 are turned on, and the lower arms (Q2, Q4, Q6, Q8, Q 10 , Q 12 ) of the first inverter 18 and the second inverter 19 are turned off for upper short-circuit control, and the upper arms (Q1, Q3, Q5, Q7, Q9, Q 11 ) of the first inverter 18 and the second inverter 19 are turned off, and the lower arms (Q2, Q4, Q6, Q8, Q 10 , Q 12 ) of the first inverter 18 and the second inverter 19 are turned off for lower short-circuit control, and these are alternately executed. Then, when switching between the upper short-circuit control and the lower short-circuit control, while maintaining the state of one of the first inverter 18 or the second inverter 19 in the same state as the start state with respect to the start state in which the upper short-circuit control or the lower short-circuit control is being performed, a transition state is formed in which the switching of the other one of the first inverter 18 or the second inverter 19 is started in advance, and after a predetermined time (T delay ) has elapsed since the switching of the other inverter was started in advance, by starting the switching of one inverter, a final state is formed in which the states of the upper arm and the lower arm of one inverter are reversed with respect to the start state, and the states of the upper arm and the lower arm of the other inverter are reversed with respect to the start state.

[0196] Thus, when switching between the upper short-circuit control and the lower short-circuit control in the open-winding type rotating electrical machine 10, by forming a transition state in which the switching of one of the first inverter 18 or the second inverter 19 is advanced (delayed), it is possible to switch between the upper short-circuit control and the lower short-circuit control while protecting the smoothing capacitor 39. Specifically, when switching between the upper short-circuit control and the lower short-circuit control without going through the above-described transition state, during the dead time DT, all arms are turned off, and the phase currents i u , i v , i w caused by the induced voltage of the rotating electrical machine 10 may charge the smoothing capacitor 39, resulting in overcharging and failure. On the other hand, when providing the above-described transition state when switching between the upper short-circuit control and the lower short-circuit control, it is possible to switch between the upper short-circuit control and the lower short-circuit control without substantially charging the smoothing capacitor 39. Therefore, according to the inverter control method according to the above-described embodiment and the like, the smoothing capacitor 39 can be protected when switching between the upper short-circuit control and the lower short-circuit control.

[0197] In the inverter control methods according to the first embodiment, the second embodiment, and the modification, specifically, as the transition state, while maintaining the states of the upper arm and the lower arm of either one of the first inverter 18 or the second inverter 19 in the same state as the start state, the upper arm and the lower arm of the other one of the first inverter 18 or the second inverter 19 are both turned off to form a first transition state, and while maintaining the states of the upper arm and the lower arm of one inverter in the same state as the start state, a second transition state in which the states of the upper arm and the lower arm of the other inverter are reversed with respect to the start state, and while maintaining a state in which the states of the upper arm and the lower arm of the other inverter are not reversed with respect to the start state, a third transition state in which the upper arm and the lower arm of one inverter are both turned off are formed.

[0198] Thus, when forming the first transition state, the second transition state, and the third transition state as transition states, although the smoothing capacitor 39 is charged in the first transition state and the third transition state, the smoothing capacitor 39 is discharged in the second transition state. Therefore, the charging of the smoothing capacitor 39 in the first transition state and the third transition state is offset by the discharging in the second transition state. Thus, as described above, when forming the first transition state, the second transition state, and the third transition state as transition states, it is easy to substantially discharge the smoothing capacitor 39. As a result, when switching between the upper short-circuit control and the lower short-circuit control, it is particularly easy to protect the smoothing capacitor 39.

[0199] In the inverter control methods according to the first embodiment, the second embodiment, and the modification, a predetermined time (T delay ) is longer than the dead time DT required when switching on and off the paired upper arm and lower arm.

[0200] Thus, when making the delay time T delay related to the switching of the first inverter 18 or the second inverter 19 longer than the dead time DT, a period (the second transition state) during which the smoothing capacitor 39 is discharged in the switching between the upper short-circuit control and the lower short-circuit control is ensured. Therefore, by making the delay time T delay longer than the dead time DT, it is possible to switch between the upper short-circuit control and the lower short-circuit control without substantially charging the smoothing capacitor 39. That is, in the switching between the upper short-circuit control and the lower short-circuit control, the smoothing capacitor 39 is easily protected.

[0201] In the inverter control methods according to the first embodiment, the second embodiment, and the modification, the length of the predetermined time (T delay ) is determined based on the zero-phase current i0 that is the sum of the currents (i u , i v , i w ) flowing through the respective phases of the open-winding type rotating electrical machine (10).

[0202] Thus, the delay time T related to the switching of the first inverter 18 or the second inverter 19 delay (especially the additional time T p part) is determined based on the zero-phase current i0, and in the transition state, the smoothing capacitor 39 can surely be discharged net. Therefore, the smoothing capacitor 39 is easily protected in the switching between the upper short-circuit control and the lower short-circuit control.

[0203] In the inverter control methods according to the first embodiment, the second embodiment, and the modification, the smaller the magnitude |i0| of the zero-phase current i0 is, the longer the predetermined time (T delay ).

[0204] Thus, the smaller the magnitude |i0| of the zero-phase current i0 is, the longer the delay time T related to the switching of the first inverter 18 or the second inverter 19 delay (especially the additional time T p part) is, and in the transition state, the smoothing capacitor 39 can surely be discharged net. Therefore, the smoothing capacitor 39 is easily protected in the switching between the upper short-circuit control and the lower short-circuit control.

[0205] In the inverter control methods according to the first embodiment and the modification, when the zero-phase current i0, which is the sum of the currents (i u , i v , i w ) flowing through each phase of the open-winding type rotating electrical machine (10), is negative, the timing of switching between the upper short-circuit control and the lower short-circuit control is delayed until the zero-phase current i0 becomes larger than zero.

[0206] Thus, when the timing adjustment process of delaying the timing of switching between the upper short-circuit control and the lower short-circuit control is performed until the zero-phase current i0 becomes larger than zero, in the transition state, the smoothing capacitor 39 can be surely discharged net particularly. More specifically, in the second transition state, the direct current I flowing through the smoothing capacitor 39 dcSince it is equal to the zero-phase current i0, if the upper short-circuit control and the lower short-circuit control are switched at a timing when the zero-phase current i0 is greater than zero as described above, the smoothing capacitor 39 can be surely discharged in the second transition state. As a result, the charging in the first transition state and the third transition state can be surely offset by the discharging in the second transition state.

[0207] In the control method of the inverter according to the second embodiment and the modification, based on the direction of the zero-phase current i0 which is the sum of the currents (i u , i v , i w ) flowing through each phase of the open-winding type rotating electrical machine (10), which of the first inverter 18 and the second inverter 19 is to be the other inverter that switches relatively earlier and which is to be the one inverter that switches relatively later is switched.

[0208] In this way, by switching the inverter that switches earlier / delays according to the direction of the zero-phase current i0 (whether the zero-phase current i0 is greater than zero), even if the timing adjustment process as described above is not performed, the smoothing capacitor 39 can be surely discharged net in the transition state. Also, there is an advantage that it is easy to suppress the time required for switching between the upper short-circuit control and the lower short-circuit control as compared with performing the timing adjustment process.

[0209] In the control method of the inverter according to the first embodiment, the second embodiment, and the modification, when the smoothing capacitor 39 discharges when the zero-phase current i0 flows from the first inverter 18 to the second inverter 19 through the open-winding type rotating electrical machine (10) in the transition state, the first inverter 18 is switched later and the second inverter 19 is switched earlier.

[0210] Thus, when the zero-phase current i0 flows from the first inverter 18 to the second inverter 19 via the rotating electrical machine 10 and the smoothing capacitor 39 is discharged, if the switching of the first inverter 18 is delayed and the switching of the second inverter 19 is advanced, the smoothing capacitor 39 can be discharged more reliably, particularly in the transition state (especially the second transition state). That is, in the transition state, the smoothing capacitor 39 can be discharged net more reliably.

[0211] In the inverter control methods according to the first embodiment, the second embodiment, and the modification, the predetermined time (T delay ) is variable. When the DC voltage V dc , which is the voltage across the smoothing capacitor 39, is higher than a predetermined threshold value (V th ), the transition state is formed by setting the predetermined time (T delay ) to be longer than the dead time DT required for switching the on and off states of the paired upper arm and lower arm. When the DC voltage V dc is equal to or lower than the threshold value (V th ), the transition state is omitted by setting the predetermined time (T delay ) to zero.

[0212] In the ASC control, the preferable scene for charging the smoothing capacitor 39 can be determined based on whether the DC voltage V dc is lower than the voltage threshold value V th . Therefore, when V dc > V th and it is preferable to discharge the smoothing capacitor 39, the ASC control is executed in the first ASC mode where the predetermined time (delay time T delay ) is set longer than the dead time DT. When V dc ≤ V th , the ASC control is executed in the second ASC mode where the predetermined time (delay time T delay ) is set to zero as described above. Thereby, the smoothing capacitor 39 can be charged to a certain extent for the restart of the system (recharging of the smoothing capacitor 39).

[0213] In the control methods of the inverters according to the above-described first embodiment, second embodiment, and modification, when a current sensor 14 for detecting phase currents i u , i v , i w flowing through each phase of the open-winding type rotating electrical machine (10) fails, based on the electrical angle θ of the open-winding type rotating electrical machine (10), the phase currents i u , i v , i w are estimated, and duty command values Duty_UVW1 and Duty_UVW2 for controlling the first inverter 18 and the second inverter 19 are calculated based on the estimated phase currents (i u ′, i v ′, i w ′).

[0214] Thus, by using the estimated phase currents i u ′, i v ′, i w ′, it is possible to perform ASC control in an appropriate mode even when the current sensor 14 fails. Therefore, when switching between the upper short-circuit control and the lower short-circuit control, it is possible to reliably protect the smoothing capacitor 39 especially surely.

[0215] The control device of the inverter according to the above-described first embodiment, second embodiment, and modification is a control device (controller 13) of an inverter that controls a first inverter 18 that connects one end of a stator coil (U, V, W) of an open-winding type rotating electrical machine (10) to a DC power supply (11) via a smoothing capacitor 39 and a second inverter 19 that connects the other end of the stator coil (U, V, W) to the DC power supply (11) via the smoothing capacitor 39. This control device (controller 13) turns on the upper arms (Q1, Q3, Q5, Q7, Q9, Q 11 ) of the first inverter 18 and the second inverter 19, and turns off the lower arms (Q2, Q4, Q6, Q8, Q 10 , Q 12 ) of the first inverter 18 and the second inverter 19, and upper short-circuit control for turning off the upper arms (Q1, Q3, Q5, Q7, Q9, Q11 ) is turned off, and the lower arms (Q2, Q4, Q6, Q8, Q 10 , Q 12 ) of the first inverter 18 and the second inverter 19 are turned off, and a PWM control unit 24 that alternately executes lower short-circuit control is provided. When switching between upper short-circuit control and lower short-circuit control, the PWM control unit 24, with respect to the initial state in which upper short-circuit control or lower short-circuit control is being performed, while maintaining the state of one of the first inverter 18 or the second inverter 19 in the same state as the initial state, forms a transition state in which the switching of the other inverter among the first inverter 18 or the second inverter 19 is started first, and after a predetermined time (T delay ) has elapsed since the switching of the other inverter was started first, by starting the switching of one inverter, the states of the upper arm and the lower arm of one inverter are reversed with respect to the initial state, and a final state in which the states of the upper arm and the lower arm of the other inverter are reversed with respect to the initial state is formed.

[0216] In this way, when switching between upper short-circuit control and lower short-circuit control in the open-winding type rotating electrical machine 10, by forming a transition state in which the switching of one of the first inverter 18 or the second inverter 19 is advanced (delayed), it is possible to switch between upper short-circuit control and lower short-circuit control while protecting the smoothing capacitor 39.

[0217] As described above, the embodiments and modified examples of the present invention have been described. However, the configurations described in the above embodiments and modified examples merely show a part of the application examples of the present invention and are not intended to limit the technical scope of the present invention.

Explanation of Reference Numerals

[0218] 10: Rotating electrical machine, 11: Battery, 12: Inverter, 13: Controller, 14: Current sensor, 15: Rotation sensor, 16: Relay, 17: Voltage sensor, 18: First inverter, 19: Second inverter, 21: Torque control unit, 22: Voltage distribution unit, 23: Coordinate conversion unit, 24: PWM control unit, 25: PWM signal generator, 26: Coordinate conversion unit, 27: Rotational speed calculation unit, 28: Abnormality detection unit, 31: First conversion unit, 32: Second conversion unit, 33: First PWM control unit, 34: Second PWM control unit, 35: Output determination unit, 36: First PWM signal generator, 37: Second PWM signal generator, 39: Smoothing capacitor, 41: Up / down switching signal generation unit, 42: Three-phase sum processing unit, 43: Timing adjustment processing unit, 44: ASC switching processing unit, 100: Electric vehicle, 301: Current selection unit

Claims

1. An inverter control method for controlling a first inverter connected to one end of a stator coil of an open-wound rotating electrical machine via a smoothing capacitor and a DC power supply, and a second inverter connected to the other end of the stator coil and the DC power supply via the smoothing capacitor, comprising: alternately executing an upper short-circuit control for turning on the upper arms of the first inverter and the second inverter and turning off the lower arms of the first inverter and the second inverter, and a lower short-circuit control for turning off the upper arms of the first inverter and the second inverter and turning off the lower arms of the first inverter and the second inverter; when switching between the upper short-circuit control and the lower short-circuit control, while maintaining the state of one of the first inverter or the second inverter in the same state as the starting state with respect to the starting state in which the upper short-circuit control or the lower short-circuit control is being performed, a transition state is formed by starting the switching of the other one of the first inverter or the second inverter in advance; after a predetermined time has elapsed since the switching of the other inverter was started in advance, by starting the switching of the one inverter, a final state is formed in which the states of the upper arm and the lower arm of the one inverter are reversed with respect to the starting state, and the states of the upper arm and the lower arm of the other inverter are reversed with respect to the starting state. An inverter control method.

2. The inverter control method according to claim 1, wherein as the transition state, a first transition state in which the upper arm and the lower arm of one of the first inverter or the second inverter are both turned off while maintaining the state of the upper arm and the lower arm of the one inverter in the same state as the starting state; a second transition state in which the upper arm and the lower arm of the other inverter are reversed with respect to the starting state while maintaining the state of the upper arm and the lower arm of the one inverter in the same state as the starting state; a third transition state in which the upper arm and the lower arm of the one inverter are both turned off while maintaining the state of the upper arm and the lower arm of the other inverter in a state that is not reversed with respect to the starting state. Forming A method for controlling an inverter. **Claim 3** A method for controlling an inverter according to claim 1 or 2, wherein the predetermined time is longer than the dead time required when switching on and off the paired upper arm and lower arm. A method for controlling an inverter. **Claim 4** A method for controlling an inverter according to claim 3, wherein the length of the predetermined time is determined based on the zero-phase current which is the sum of the currents flowing through each phase of the open-winding type rotating electric machine. A method for controlling an inverter. **Claim 5** A method for controlling an inverter according to claim 4, wherein the smaller the magnitude of the zero-phase current, the longer the predetermined time. A method for controlling an inverter. **Claim 6** A method for controlling an inverter according to claim 3, when the zero-phase current which is the sum of the currents flowing through each phase of the open-winding type rotating electric machine is negative, delaying the timing of switching between the upper short-circuit control and the lower short-circuit control until the zero-phase current becomes larger than zero. A method for controlling an inverter. **Claim 7** A method for controlling an inverter according to claim 3, based on the direction of the zero-phase current which is the sum of the currents flowing through each phase of the open-winding type rotating electric machine, switching which of the first inverter and the second inverter is the other inverter that switches relatively earlier and which is the one inverter that switches relatively later. A method for controlling an inverter. **Claim 8** A method for controlling an inverter according to claim 7, in the transition state, when the smoothing capacitor discharges when the zero-phase current flows from the first inverter through the open-winding type rotating electric machine to the second inverter, delaying the switching of the first inverter and advancing the switching of the second inverter. A method for controlling an inverter. **Claim 9** A method for controlling an inverter according to claim 1 or 2, wherein the predetermined time is variable, when the DC voltage which is the voltage across the smoothing capacitor is higher than a predetermined threshold value, forming the transition state by setting the predetermined time longer than the dead time required when switching on and off the paired upper arm and lower arm, when the DC voltage is below the threshold value, omitting the transition state by setting the predetermined time to zero. A method for controlling an inverter. **Claim 10** A method for controlling an inverter according to claim 1 or 2, wherein when a current sensor for detecting a phase current, which is a current flowing through each phase of the open-winding rotating electrical machine, fails, the phase current is estimated based on the electrical angle of the open-winding rotating electrical machine, a duty command value for controlling the first inverter and the second inverter is calculated based on the estimated phase current, A method for controlling an inverter.

11. An inverter control device for controlling a first inverter connected to one end of a stator coil of an open-winding rotating electrical machine via a smoothing capacitor and a DC power supply, and a second inverter connected to the other end of the stator coil and the DC power supply via the smoothing capacitor, comprising a PWM control unit that alternately executes an upper short-circuit control for turning on the upper arms of the first inverter and the second inverter and turning off the lower arms of the first inverter and the second inverter, and a lower short-circuit control for turning off the upper arms of the first inverter and the second inverter and turning off the lower arms of the first inverter and the second inverter, when the PWM control unit switches between the upper short-circuit control and the lower short-circuit control, while maintaining the state of one of the first inverter or the second inverter in the same state as the initial state with respect to the initial state in which the upper short-circuit control or the lower short-circuit control is being performed, a transition state is formed in which the switching of the other inverter among the first inverter or the second inverter is started in advance, after a predetermined time has elapsed since the switching of the other inverter was started in advance, by starting the switching of the one inverter, a final state is formed in which the states of the upper arm and the lower arm of the one inverter are reversed with respect to the initial state, and the states of the upper arm and the lower arm of the other inverter are reversed with respect to the initial state, An inverter control device.

Images