JP6563135B2 - Motor control device - Google Patents

Description

  The present invention relates to a motor control device, and more particularly to sensorless control for driving a motor without using a position sensor for detecting a rotation angle of a rotor.

  Conventionally, in motor control using a rotor rotation position (rotation angle) typified by vector control, sensorless control using a rotation angle obtained by estimation calculation without arranging a position sensor is known. In sensorless control, since the rotation angle is estimated by magnetic flux calculation based on the motor current, the control accuracy tends to decrease due to a decrease in estimation accuracy in a low speed region where the motor current is small.

  In order to cope with this point, Japanese Patent No. 4370754 (Patent Document 1) discloses a sensorless system configured to improve estimation accuracy in a low speed region by superimposing a high frequency signal on an inverter output voltage. A control device is described.

Japanese Patent No. 4370754

  However, the control device described in Patent Document 1 requires a configuration for generating a harmonic signal and further extracting the same frequency component as the harmonic signal. For this reason, there is a risk of cost increase due to the need for a processor with a high processing speed due to the complexity of the control configuration. Furthermore, there is a concern that harmonic components are superimposed on the motor current, which may cause vibration due to torque pulsation and increase in noise. Therefore, it is desired to improve the accuracy of sensorless control in the low speed region without superimposing such harmonic signals.

  The present invention has been made to solve such problems, and an object of the present invention is to improve control accuracy in a low speed region of sensorless control of a motor with a simple control configuration.

  A motor control device according to the present invention is a motor control device in which a position sensor is not disposed, and includes a DC power supply, an inverter, a current detection unit, and a control unit. The inverter is configured to convert a DC voltage from a DC power source into an AC voltage applied to the motor. The current detection unit is configured to detect a motor current flowing through the motor. The control unit is configured to generate a control signal for controlling on / off of the plurality of semiconductor switching elements constituting the inverter. The control unit generates a control signal so as to have a timing at which the motor current decreases during the operation of increasing the rotation speed of the motor during the feedback control of the motor current based on the detection value by the current detection unit.

  According to the present invention, the control accuracy in the low speed region of the sensorless control of the motor can be enhanced with a simple control configuration.

It is a block diagram which shows the structural example of the motor system to which the control apparatus of the motor according to this Embodiment was applied. It is a functional block diagram for demonstrating the control structure by the motor control apparatus according to this Embodiment. It is a functional block diagram for demonstrating the structure of the motor control calculating part shown by FIG. It is a conceptual graph for demonstrating an example of switching of starting control and steady state control. It is a functional block diagram for demonstrating in detail the control action by the steady control calculating part shown by FIG. It is a conceptual diagram for demonstrating the three-phase / two-phase conversion in the coordinate transformation part shown by FIG. 6 is a conceptual graph for explaining processing for setting a current command value for the γ-axis by the γ-axis control switching unit shown in FIG. 5. It is a wave form diagram which shows the simulation result of the motor current waveform in the low speed area | region by the motor current control according to a comparative example. It is a wave form diagram which shows an example of the control signal of the inverter provided with dead time. It is a wave form diagram which shows the simulation result of the motor current waveform by the motor control apparatus according to this Embodiment.

  Embodiments of the present invention will be described below in detail with reference to the drawings. In the following, the same parts in the drawings are denoted by the same reference numerals, and the description thereof will not be repeated in principle.

  FIG. 1 is a block diagram showing a schematic configuration of a system to which a motor control device according to the present embodiment is applied.

  Referring to FIG. 1, motor system 10 drives and controls motor 15 with electric power from AC power supply 12. The motor 15 is provided as a component of various devices. For example, the motor 15 can be provided as an outdoor fan motor of an air conditioner.

  The AC power supply 12 can typically be constituted by a commercial power supply. The motor 15 is constituted by, for example, a permanent magnet synchronous motor having a permanent magnet on a rotor (rotor) of the motor. In the example of FIG. 1, the motor 15 is configured by a three-phase permanent magnet synchronous motor.

  Rectifier 20 rectifies the AC voltage from AC power supply 12 and outputs the rectified voltage between power lines PL and NL. The rectifying unit 20 can be typically constituted by a diode bridge. Further, smoothing capacitor 30 is connected between power lines PL and NL. Thereby, a DC voltage is generated between power lines PL and NL.

  Voltage detector 35 detects a voltage between power lines PL and NL. A detection signal from the voltage detection unit 35 is input to the control unit 100. Hereinafter, the DC voltage detected by the voltage detection unit 35 is denoted as Vdc.

  As described above, the AC power supply 12, the rectifying unit 20, and the smoothing capacitor 30 constitute an example of a “DC power supply” that outputs a DC voltage. Note that in this embodiment, the configuration of the DC power supply is not particularly limited, and the DC power supply can be configured using a power storage device such as a secondary battery, a generator, or the like.

  Inverter 40 has a circuit configuration of a general three-phase inverter, converts a DC voltage on power lines PL and NL into a three-phase AC voltage, and applies it to each phase of motor 15. Inverter 40 has a general three-phase inverter circuit configuration and includes power semiconductor switching elements (hereinafter also simply referred to as “switching elements”) Q1 to Q6. Switching elements Q1 to Q6 constitute an upper arm and a lower arm of the U phase, the V phase, and the W phase, respectively.

  An intermediate node Nu of the U-phase arm, that is, a connection node of switching elements Q1 and Q2 connected in series between power lines PL and NL is connected to a U-phase coil winding (not shown) of motor 15 by power line 41U. The Similarly, intermediate node Nv (connection node of switching elements Q3 and Q4) of the V-phase arm is connected to a V-phase coil winding (not shown) of motor 15 by power line 41V. Intermediate node Nw (connection node of switching elements Q5 and Q6) of the W-phase arm is connected to a W-phase coil winding (not shown) of motor 15 by power line 41W.

  The configuration of the inverter 40 shown in FIG. 1 is an example, and the inverter 40 has any function as long as it has a function of converting a DC voltage from a DC power source into an AC voltage applied to the motor 15. A circuit configuration can be applied.

  A current detection unit 50 for detecting a motor current (phase current) flowing through the motor 15 is disposed on the power lines 41U, 41V, 41W. A detection signal from the current detection unit 50 is input to the control unit 100. Hereinafter, the motor current of each phase detected by the current detection unit 50 is also referred to as a U-phase current Iu, a V-phase current Iv, and a W current Iw. In addition, when including three phases, it is also simply referred to as motor currents Iu, Iv, and Iw.

  The current detection unit 50 includes current detectors provided in at least two phases of the power lines 41U, 41V, and 41W. Each current detector can be constituted by a current transformer or a shunt resistor. As in the example of FIG. 1, current detectors may be arranged in all three phases, but only two phases are utilized by utilizing that the sum of the three-phase instantaneous currents is 0 (Iu + Iw + Iv = 0). It is also possible to provide a current detector in the control unit 100 and calculate the remaining one phase by calculation.

  The control unit 100 can typically be configured by a microcomputer. The control unit 100 controls the output of the motor 15 by controlling the power conversion in the inverter 40 by the control signals S1 to S6. As can be understood from FIG. 1, the motor 15 is not provided with a position sensor for detecting the rotation angle of the rotor, and the output of the motor 15 is controlled by sensorless control. Control unit 100 uses DC voltage Vdc from voltage detection unit 35 and motor currents Iu, Iv, Iw from current detection unit 50 to control signals S1 to S6 for controlling on / off of switching elements Q1 to Q6 of inverter 40. Is generated.

  FIG. 2 is a functional block diagram for illustrating a control configuration by the motor control device according to the present embodiment. 2 and each of a plurality of functional block diagrams described later, the function of each block is to execute a predetermined program (software processing) by the control unit 100 and / or an electronic circuit (hardware) built in the control unit 100. Hardware processing).

  Referring to FIG. 2, motor control unit 110 includes a speed command value generation unit 120, a motor control calculation unit 130, and a control signal generation unit 140.

  The speed command value generation unit 120 generates a speed command value ω * for the motor 15. For example, when the motor 15 is an outdoor fan motor of an air conditioner, the speed command value ω * is generated based on the operating state of the air conditioner.

  The motor control calculation unit 130 generates a U phase, a V phase, and a W phase of the inverter 40 based on the DC voltage Vdc detected by the voltage detection unit 35 and the motor currents Iu, Iv, Iw detected by the current detection unit 50. Voltage command values Vu *, Vv *, and Vw * are generated. For example, the voltage command values Vu *, Vv *, and Vw * for each phase have the same amplitude and have a sinusoidal AC voltage waveform whose phases are shifted from each other by 120 ° (electrical angle). The motor control calculation unit 130 can control the phase and the phase of the AC voltage so that the rotation speed of the motor 15 matches the speed command value ω *.

  Control signal generation unit 140 generates control signals S1 to S6 for switching elements Q1 to Q6 in accordance with voltage command values Vu *, Vv *, and Vw * for each phase of inverter 40. For example, control signals S1 to S6 are pulse signals for controlling on / off of switching elements Q1 to Q6.

  Typically, control signal generation section 140 performs pulse width modulation (PWM) control based on voltage comparison between voltage command values Vu *, Vv *, Vw * of each phase and a carrier wave (for example, a triangular wave) having a predetermined frequency. Thus, the control signals S1 to S6 can be generated. Specifically, in the U phase, in a period in which the voltage command value Vu * having an AC voltage waveform is higher than the voltage of the carrier wave (triangular wave), the upper arm switching element Q1 is turned on while Vu * is the carrier. In a period lower than the wave voltage, the control signals S1 and S2 are generated so that the lower arm switching element Q2 is turned on.

  Thus, in each phase, the upper arm elements (Q1, Q3, Q5) and the lower arm elements (Q2, Q4, Q6) are crossed at the intersections of the voltage command values Vu *, Vv *, Vw * and the carrier wave. ON / OFF is switched. Thereby, the three-phase pseudo sine wave voltage by PWM can be applied to the three-phase coil winding (not shown) of the motor 15 from the inverter 40.

  In each phase arm of inverter 40, when both upper and lower arm elements (for example, U-phase switching elements Q1, Q2) of the same arm are turned on, a short-circuit path is formed between power lines PL and NL. Therefore, when switching on / off between the upper arm elements (Q1, Q3, Q5) and the lower arm elements (Q2, Q4, Q6), it is common to provide a dead time for both the upper and lower arm elements. It is.

  FIG. 3 is a functional block diagram for explaining the configuration of the motor control calculation unit 130 shown in FIG.

  Referring to FIG. 3, motor control calculation unit 130 includes a startup control calculation unit 150, a steady control calculation unit 200, and a control switching unit 160.

  Each of start control calculation unit 150 and steady control calculation unit 200 is a voltage command value of inverter 40 for operating motor 15 according to speed command value ω * based on DC voltage Vdc and motor currents Iu, Iv, Iw. It has a function of generating Vu *, Vv *, and Vw *.

  The control switching unit 160 outputs the voltage command values Vu *, Vv *, Vw * from one of the activation control calculation unit 150 and the steady control calculation unit 200 to the control signal generation unit 140 (FIG. 2). .

  FIG. 4 shows a conceptual graph for explaining an example of switching between start control and steady control by the motor control device according to the present embodiment.

  Referring to FIG. 4, when driving of motor 15 is started from time t0, start control is first applied. The speed command value ω * shown on the vertical axis is gradually increased from 0 to a predetermined starting rotational speed ω1, and then maintained at ω * = ω1 for a certain period.

  Thereby, the motor 15 is controlled by the start control so as to rotate at a constant speed after the rotation speed is increased to the start rotation speed ω1. For example, in the start control, the voltage command values Vu *, Vv *, and Vw * can be set so that a large amount of reactive current flows in order to start the motor 15 stably. At time tx during the period maintained at ω * = ω1, switching from the start control to the steady control is executed. Thereby, the steady control can be started with respect to the motor 15 from a stable state at a constant rotational speed at the starting rotational speed ω1.

  According to the example of FIG. 4, the control switching unit 160 of FIG. 3 uses the voltage command values Vu *, Vv *, and Vw * calculated by the activation control calculation unit 150 (FIG. 3) between times t0 and tx. And output to the control signal generator 140 (FIG. 2). On the other hand, after time tx, the control switching unit 160 uses the voltage command values Vu *, Vv *, and Vw * calculated by the steady control calculation unit 200 (FIG. 3) to the control signal generation unit 140 (FIG. 2). Output. The steady control calculation unit 200 calculates voltage command values Vu *, Vv *, Vw * by feedback control of the motor currents Iu, Iv, Iw according to the current command values.

  As described above, the control switching unit 160 may switch between the start control and the steady control depending on whether the voltage command values Vu *, Vv *, and Vw * from the start control calculation unit 150 or the steady control calculation unit 200 are output. it can. In the example of FIG. 4, the switching from the start control to the steady control is controlled based on the elapsed time after ω * = ω1 according to the speed command value ω *.

  As will be apparent from the following description, in the present embodiment, since the motor control accuracy in the low speed region in the steady control is increased, the control content in the start control and the change from the start control to the steady control are described. There is no particular limitation on the switching method, and the point that any method can be applied will be described in a confirming manner.

  FIG. 5 is a functional block diagram for explaining in detail the control operation by the steady control calculation unit 200 of FIG.

  Referring to FIG. 5, steady control calculation unit 200 includes coordinate conversion unit 210, state estimation unit 220 for estimating the magnetic flux, speed, and position of motor 15, magnetic flux command value generation unit 230, and magnetic flux control unit. 240, a current command value fixing unit 250, a γ-axis control switching unit 260, a speed control unit 270, a current control unit 280, and a voltage command calculation unit 290.

  The coordinate conversion unit 210 converts the three-phase (U, V, W) motor currents Iu, Iv, Iw into the two-phase currents Iγ, Iδ on the γ-δ axis defined in FIG.

FIG. 6 is a conceptual diagram for explaining the three-phase / two-phase conversion in the coordinate conversion unit 210.
Referring to FIG. 6, the U axis, the V axis, and the W axis are three-phase stationary coordinate systems that correspond to the U phase, the V phase, and the W phase of the motor 15 and that have different phases by 120 ° (electrical angle). . Motor currents Iu, Iv, Iw and voltage command values Vu *, Vv *, Vw * are current values or voltage values on the U axis, V axis, and W axis.

  On the other hand, of the d-axis and q-axis of the rotating coordinate system indicated by the dotted line, the d-axis indicates the actual magnetic pole axis of the motor 15, and the rotor of the motor 15 is between the U-axis of the stationary coordinate system. It has a phase difference corresponding to the rotational position (rotational angle θ). The q axis is defined as a coordinate axis orthogonal to the d axis.

  In sensorless control used in the motor control device according to the present embodiment, the γ-δ axes based on rotor rotation angle estimation value θ # are further defined. That is, the phase difference between the γ-axis and the U-axis corresponds to the rotor rotation angle estimated value θ #. The δ axis is defined as a coordinate axis orthogonal to the γ axis.

  The dq axis and the γ-δ axis are coordinate axes that rotate around the origin at the rotational speed ω of the motor 15. Thus, since the d-axis corresponds to the estimated d-axis and the δ-axis corresponds to the estimated q-axis, when the error of the rotor rotation angle estimated value θ # is 0 (that is, θ # = θ), The γ-axis matches, and the q-axis and δ-axis match.

  In the sensorless control, the d-axis and the q-axis cannot be directly estimated, and therefore the motor 15 is controlled using the rotation angle estimation value θ # having the angle estimation error Δθ with respect to the actual rotation angle θ that cannot be detected. . The rotation angle estimation value θ # is obtained by the state estimation unit 220 as described later.

  As described above, in this embodiment, the motor 15 is sensorlessly controlled by so-called vector control using the motor current converted into the rotating coordinate system. However, the definition of the coordinate system described with reference to FIG. It is possible to perform the motor control by applying the coordinate transformation.

  Referring to FIG. 5 again, state estimation unit 220 estimates magnetic flux using γ-axis current Iγ and δ-axis current Iδ obtained by coordinate conversion unit and voltage command values Vγ * and Vδ * of γ-axis and δ-axis. A value φγ #, a speed estimation value ω #, and a rotation angle estimation value θ # are output.

  For example, state estimating unit 220 can estimate magnetic flux estimated value φγ #, speed estimated value ω #, and rotational angle estimated value θ # according to the following equations (1) to (3).

  In equation (1), the total magnetic flux vector by the permanent magnet can be estimated by an operation of integrating a voltage obtained by subtracting the voltage drop due to the resistance R of the coil winding from the voltage command value of the γ-axis using a low-pass filter. That is, ωc in equation (1) can be a fixed value corresponding to the cutoff frequency of the low-pass filter. By blocking the frequency component lower than the cutoff frequency ωc, it is possible to prevent the noise error of the input signal from being amplified and increase the error of the estimated magnetic flux φγ #.

  In the equation (2), the velocity estimated value ω is obtained by dividing (Vδ−R · Iδ) by the magnetic flux estimated value φγ # obtained by the equation (1) from the relational expression of V = ω · φ with respect to the δ axis. # Can be estimated. Further, in the equation (3), the rotation angle estimated value θ # can be estimated by integrating the speed estimated value ω # obtained by the equation (2).

  Speed estimation value ω # and magnetic flux estimation value φγ # calculated by state estimation unit 220 are input to speed control unit 270 and magnetic flux control unit 240, respectively.

  The estimated rotation angle value θ # calculated by the state estimation unit 220 is input to the coordinate conversion unit 210 and used to calculate the γ-axis current Iγ and the δ-axis current Iδ. Further, the estimated rotation angle θ # is input to the voltage command calculation unit 290 and used for inverse conversion from the two-phase coordinates (γ-δ axis) to the three-phase coordinates (U axis, V axis, W axis). It is done. As a result, the error of the rotation angle estimation value θ # decreases the accuracy of the motor control by the steady control calculation unit 200 that performs the feedback control of the motor current through the error in the coordinate conversion of the motor current by the coordinate conversion unit 210. It is understood that

  Speed control unit 270 generates a current command value Iδ * for the δ axis based on speed command value ω * and speed estimated value ω # estimated by state estimating unit 220. For example, speed control unit 270 generates δ-axis current command value Iδ * in accordance with PID control of speed deviation Δω (Δω = ω * −ω #) or the like. By performing feedback control of the δ-axis current Iδ according to the current command value Iδ *, torque for converging Δω to 0 can be output from the motor 15.

  The γ-axis current command value Iγ * is set by the γ-axis control switching unit 260 by selecting one of the magnetic flux control unit 240 and the current command value fixing unit 250. The current command value fixing unit 250 outputs Iγ * const which is a constant value.

  On the other hand, the magnetic flux command value generation unit 230 generates a magnetic flux command value φγ * according to the speed command value ω *. Typically, the magnetic flux command value generation unit 230 variably sets the magnetic flux command value φγ * according to the speed command value ω * so that the magnetic flux vector is controlled to be constant.

  For example, the magnetic flux command value generation unit 230 can be configured using a lookup table that predefines the correspondence between the speed command value ω * and the magnetic flux command value φγ *. Alternatively, the magnetic flux command value generation unit 230 can be configured by preparing in advance an arithmetic expression for calculating the magnetic flux command value φγ * using the speed command value ω * as a variable.

  The magnetic flux control unit 240 generates the γ-axis current command value Iγ * based on the magnetic flux command value φγ * from the magnetic flux command value generation unit 230 and the estimated magnetic flux value φγ # calculated by the state estimation unit 220. .

  For example, the magnetic flux control unit 240 generates a current command value Iγ * for the γ axis in accordance with PID control or the like of the magnetic flux deviation Δφγ (Δφγ = φγ * −φγ #). By performing feedback control of the γ-axis current Iγ according to the current command value Iγ *, the estimated magnetic flux value φγ # can be matched with the magnetic flux command value φγ * (Δφγ = 0), that is, the magnetic flux vector can be controlled to be constant.

  The γ-axis control switching unit 260 selects one of the current command value Iγ * from the magnetic flux control unit 240 and Iγ * const from the current command value fixing unit 250, and sets the current command value Iγ * of the γ-axis as the current command value Iγ *. Output to the control unit 280.

  FIG. 7 is a conceptual graph for explaining the setting process of the γ-axis current command value Iγ * by the γ-axis control switching unit 260 shown in FIG.

  Referring to FIG. 7, speed command value ω * is set in the same pattern as in FIG. 4, and steady control according to FIG. 5 is executed from time tx. Further, at time ty, ω * reaches a predetermined threshold value ωt (ω * = ωt), and after time ty, ω * exceeds the threshold value ωt (ω *> ωt). The threshold value ωt is set higher than the starting rotational speed ω1.

  In the steady control, in the region of ω * ≦ ωt (time tx to ty), the γ-axis current command value Iγ * is set using Iγ * const (constant value) from the current command value fixing unit 250 ( Iγ * = Iγ * const). In contrast, in the region of ω *> ωt (after time ty), the current command value Iγ * from the magnetic flux control unit 240 is used to set the current command value Iγ * of the γ axis during the steady control. That is, the γ-axis control switching unit 260 can operate to select one of the magnetic flux control unit 240 and the current command value fixing unit 250 based on the comparison between the speed command value ω * and the threshold value ωt.

  Referring again to FIG. 5, current control unit 280 includes current command value Iγ * output from γ-axis control switching unit 260, current command value Iδ * output from speed control unit 270, and coordinate conversion unit 210. Voltage command values Vγ * and Vδ * are calculated based on the γ-axis current Iγ and the δ-axis current Iδ calculated by the above. The current command value Iγ * corresponds to the “excitation current command value”, and the current command value Iδ * corresponds to the “torque current command value”.

  The current control unit 280 calculates the voltage command value Vγ * by γ-axis current feedback control. For example, the γ-axis voltage command value Vγ * for converging the current deviation ΔIγ to 0 can be generated according to PID control of the current deviation ΔIγ (ΔIγ = Iγ * −Iγ) of the γ-axis. Similarly, the current control unit 280 calculates the voltage command value Vδ * by δ-axis current feedback control. For example, the voltage command value Vδ * for converging the current deviation ΔIδ to 0 can be generated according to the PID control of the current deviation ΔIδ (ΔIδ = Iδ * −Iδ) on the δ axis.

  The voltage command calculation unit 290 converts the voltage command values Vγ * and Vδ * calculated by the current control unit 280 from two-phase coordinates (γ-δ axis) to three-phase coordinates (U axis, V axis, W axis). By doing so, voltage command values Vu *, Vv *, and Vw * for U phase, V phase, and W phase are calculated. The rotation angle estimation value θ # calculated by the state estimation unit 220 is used for the two-phase / three-phase conversion.

  At this time, the output voltage phase θv of the inverter 40 is expressed by the following equation (4) using the voltage command values Vγ * and Vδ * by the current control unit 280.

  Regarding the three-phase voltage command values Vu *, Vv *, and Vw * obtained by the two-phase / three-phase conversion by the voltage command calculation unit 290, the U-phase voltage command value Vu * is expressed by the following equation (5): A sinusoidal voltage with controlled amplitude and phase.

  The voltage command value Vv * is a sine wave voltage whose phase is delayed by 120 ° (electrical angle) from Vu * in Equation (5), and the voltage command value Vv * is further 120 ° (electrical) from Vv *. Square) Delayed sine wave voltage.

  As described above, the steady control calculation unit 200 (FIGS. 3 and 5) controls the output (rotation speed) of the motor 15 by generating the control signals S1 to S6 of the inverter 40 by feedback control of the motor current. Can do.

  Referring to FIG. 7 again, in a low speed region where the rotational speed command value ω * is lower than the threshold value ωt, by setting Iγ * = Iγ * const, it is compared with the case where the magnetic flux vector constant control is performed in the speed region. Thus, the γ-axis current Iγ is increased. That is, in the region sandwiching ω * = ωt, the current command value fixing unit is set so that Iγ * const> Iγ * (ωt) with respect to Iγ * (ωt) in the constant magnetic flux vector control at ω * = ωt. Iγ * const (constant value) by 250 is set.

As a result, when the γ-axis current command value Iγ * is set according to FIG. 7, when steadily increasing the rotational speed of the motor 15 by the steady control calculation unit 200, the timing at which ω * = ωt is passed. The current command value Iγ * decreases. Thereby, during the above operation, the steady control calculation unit 200 has a timing at which the amplitude of the motor current that is proportional to the square of the square sum of the γ-axis current Iγ and the δ-axis current Iδ (√ (Iγ ² + Iδ ² )) decreases. It will be appreciated that the inverter 40 is controlled to occur.

  Here, as a comparative example, a problem when the motor current is controlled so that the magnetic flux vector by the magnetic flux controller 240 is constant even in a low speed region where ω <ωt will be described.

  FIG. 8 is a waveform diagram showing a simulation result of the motor current waveform in the low speed region by the motor current control according to the comparative example.

  Referring to FIG. 8, U-phase, V-phase and W-phase motor currents Iu, Iv and Iw should ideally be controlled to sine wave currents that are out of phase by 120 ° (electrical angle). However, according to the motor current control according to the comparative example, pulsation and distortion are generated in the motor currents Iu, Iv, and Iw in the low speed region. The inventors have found that this phenomenon is greatly affected by the distortion of the m-current due to the dead time in the inverter 40 and the occurrence of an output voltage error of the inverter 40. .

  FIG. 9 is a waveform diagram showing an example of a control signal of the inverter 40 provided with a dead time. FIG. 9 shows waveforms of the control signal S1 of the switching element Q1 (upper arm) and the control signal S2 of the switching element Q2 (lower arm) that constitute the U phase as an example.

  Referring to FIG. 9, it is assumed that the voltages of U-phase voltage command value Vu * and the carrier wave become equal at times t1 and t2 by PWM control. As a result, at time t1, it is necessary to change the control signals S1 and S2 so that the switching element Q2 is switched from on to off and the switching element Q1 is switched from off to on. On the other hand, at time t2, it is necessary to change the control signals S1 and S2 so that the switching element Q1 is switched from on to off and the switching element Q2 is switched from off to on.

  When switching on / off of the upper arm element and the lower arm element in the same phase as described above, the ideal switching element can simultaneously change the levels of the control signals S1 and S2 at times t1 and t2. It is.

  However, in an actual switching element, a delay time occurs in the on / off operation, so that a period in which both the upper arm element and the lower arm element are in an on state occurs and a short circuit path is formed between the power lines PL and NL. In order to avoid this, it is necessary to provide a dead time (Td).

  As a result, the level of the control signal S1 is changed to turn on the switching element Q1 after the dead time Td has elapsed after the level of the control signal S2 has changed to turn off the switching element Q2 at time t1. . Similarly, at time t2, a dead time Td is ensured between the time when the level of the control signal S1 changes and the time when the level of the control signal S2 changes.

  In the dead time period indicated by hatching in FIG. 9, both the upper arm element and the lower arm element are turned off. In the output voltage in the phase of the inverter 40 in the period, an offset error voltage is generated depending on the flow direction of the current flowing in the motor 15 as a load. The absolute value of the error voltage ΔVof is given by the following equation (6). Note that fc in equation (6) is the switching frequency of the inverter 40, and in PWM control, the frequency of the carrier wave corresponds to fc.

| ΔVof | = Td · fc · Vdc (6)
As described above, since the motor current is an alternating current (ideally, a sine wave current), the positive / negative of the error voltage ΔVof changes depending on the direction (positive / negative) of the motor current in the phase during the dead time period. As a result, the inverter output voltage has a waveform in which the pulsed error voltage ΔVof in the dead time period is superimposed on the pseudo sine wave voltage according to the voltage command values Vu *, Vv *, and Vw * generated by the PWM control. To have.

  As a result, due to the influence of the error voltage ΔVof, harmonic distortion is generated in the motor current according to the dead time generation frequency. Further, the magnitude (effective value) of the inverter output voltage has an error with respect to the original voltage command values Vu *, Vv *, Vw *.

  In the low speed region of the motor 15, the motor current and output voltage of the inverter 40 are small. For this reason, since the influence of the error voltage ΔVof due to the dead time becomes large, the error of the inverter output voltage (effective value) and the harmonic distortion of the motor current become large. As a result, as shown in the current waveform of FIG. 9, the motor current may be significantly disturbed.

  Referring to FIG. 5 again, the influence of dead time on steady control (sensorless control) by the motor control device according to the present embodiment will be further described.

  First, when the motor currents Iu, Iv, and Iw are distorted due to the dead time, the distortion is propagated to the γ-axis current Iγ and the δ-axis current Iδ calculated by the coordinate conversion unit 210. In state estimation unit 220, estimated magnetic flux value φγ #, estimated speed value ω #, and estimated rotational angle value θ # are estimated using voltage command values Vγ * and Vδ *, and γ-axis current Iγ and δ-axis current Iδ. The Therefore, the distortion generated in motor currents Iu, Iv, and Iw also propagates to estimated magnetic flux value φγ #, estimated speed value ω #, and estimated rotational angle value θ #.

  Further, rotation angle estimated value θ # is used for calculation of γ-axis current Iγ and δ-axis current Iδ in coordinate conversion section 210. For this reason, the influence of the distortion propagated from the motor current to the γ-axis current Iγ and the δ-axis current Iδ returns to the γ-axis current Iγ and the δ-axis current Iδ, and an internal circulation that amplifies the error occurs. Resulting in. As a result, the angle estimation error Δθ (Δθ = θ−θ #) with respect to the actual rotation angle θ described with reference to FIG. 6 becomes unstable. As a result, the motor current is pulsated and the control accuracy of the motor 15 is reduced. There is a concern that it may increase noise.

  Second, if an error occurs in the magnitude (absolute value) of the actual output voltage of the inverter 40 due to dead time with respect to Vu *, Vv *, Vw * according to the voltage command values Vδ *, Vγ *, the voltage command value Vγ. An estimation error also occurs in the estimated magnetic flux value φγ #, estimated speed value ω #, and estimated rotational angle value θ # estimated using * and Vδ *. As a result, similarly to the distortion of the motor current, there is a concern that the controllability of the motor 15 by the steady control calculation unit 200 is deteriorated.

  As described above, the dead time Td provided in the inverter 40 has a great influence on the motor control by the feedback control of the motor current by the steady control calculation unit 200, and causes the disturbance of the motor current waveform as shown in FIG. There is concern.

  On the other hand, in the motor control device according to the present embodiment, as described with reference to FIG. 7, the motor current and the output voltage of inverter 40 become small, so that the influence of the dead time error becomes large. In ω * <ωt), the motor current is controlled without using the magnetic flux controller 240. Specifically, the motor current is controlled by setting Iγ * = Iγ * const so that the current command value Iγ * becomes a larger value than when the magnetic flux vector constant control is applied.

  As a result, since the motor currents Iu, Iv, and Iw also increase in accordance with the current command value Iγ *, distortion of the motor current due to dead time can be reduced. Further, regarding the output voltage error of the inverter 40, the voltage command value Vγ * also increases according to the current command value Iγ *, so that the influence of the error voltage ΔVof due to the dead time can be mitigated.

  FIG. 10 is a waveform diagram showing a simulation result of the motor current waveform by the motor control device according to the present embodiment. FIG. 10 shows a simulation result of the U-phase current Iu among the motor currents Iu, Iv, and Iw. Since the cycle of the motor current Iu is short with respect to the time axis scale in FIG. 10, the envelope of the Iu waveform shows the transition of the amplitude of Iu.

  Referring to FIG. 10, speed command value ω * is set in a pattern as in FIGS. 4 and 7. For this reason, activation control is applied until time tx. In the example of FIG. 10, the current command value is not set in the startup control. For example, the output of the inverter 40 is maintained at a constant startup stability so as to output a predetermined torque. Further, at the time of start-up, a large amount of exciting current (reactive current component) is flowed to ensure starting torque, so that the motor current becomes relatively large.

  When the steady control is started from the time tx, the motor current feedback control is started, so that the amplitude of the motor current decreases. This indicates that the efficiency of the motor 15 is increased in the steady control as compared with the startup control due to the reduction of the reactive current component.

  In steady-state control, Iγ * = Iγ * const is set so that the current command value Iγ * becomes larger than the time of application of constant magnetic flux vector control at times tx to ty (low speed region) where ω * ≦ ωt. . As a result, when the operation of continuously increasing the rotation speed of the motor 15 is performed, the current command value Iγ * decreases at the timing when ω * = ωt is passed. In the region of ω *> ωt (after time ty), the magnetic flux control unit 240 sets the current command value Iγ *.

  By setting the current command value Iγ * in this way, the motor current Iu is also controlled so as to decrease when the motor 15 transitions from the region of ω * ≦ ωt to the region of ω *> ωt. Is done. As a result, the motor current Iu in the low speed region can be secured, so that the influence of the dead time in the low speed region can be alleviated and the feedback control of the motor current can be stabilized. As a result, the controllability of the motor 15 can be improved as compared with the case where the magnetic flux control by the magnetic flux controller 240 is performed through steady control. That is, the control accuracy of sensorless control in the low speed region of the motor 15 can be increased with a simple configuration without complicating the control configuration by using harmonic signals as in Patent Document 1.

  Further, in the current feedback control by the steady control calculation unit 200 shown in FIG. 5, by fixing Iγ * = Iγ * const, the error is amplified by the internal circulation through the magnetic flux estimation by the state estimation unit 220 described above. The phenomenon can be suppressed. Also from this point, the controllability of the motor 15 can be improved by suppressing the influence of the dead time in the low speed region.

  In this embodiment, in order to suppress the influence of the dead time in the low speed region, the control example in which the motor current in the low speed region is increased by increasing the γ axis current Iγ has been described. However, the δ axis currents Iδ and γ It is also possible to increase the motor current in the low speed region by increasing both of the shaft currents Iγ. For example, when the operation of continuously increasing the rotational speed of the motor 15 is performed by limiting the output torque of the motor 15 so as not to adversely affect, the current command value Iδ is passed at the timing of passing ω * = ωt. The current command value Iδ * can be adjusted so that * decreases.

  In the configuration in which the motor current is feedback-controlled by coordinate transformation different from the coordinate system shown in the present embodiment (FIG. 6), the motor current of the motor 15 is relatively increased by relatively increasing the motor current at the low speed current. As long as the control that reduces the motor current passing through ω * = ωt can be realized when the operation for continuously increasing the rotation speed is performed, the sensorless control using the arbitrary motor current control, The present invention can be applied.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

DESCRIPTION OF SYMBOLS 10 Motor system, 12 AC power supply, 15 Motor, 20 Rectification part, 30 Smoothing capacitor, 35 Voltage detection part, 40 Inverter, 41U, 41V, 41W, NL, PL Power line, 50 Current detection part, 100 Control part, 110 Motor control Unit, 120 speed command value generation unit, 130 motor control calculation unit, 140 control signal generation unit, 150 start control calculation unit, 160 control switching unit, 200 steady state control calculation unit, 210 coordinate conversion unit, 220 state estimation unit, 230 magnetic flux Command value generation unit, 240 Magnetic flux control unit, 250 Current command value fixing unit, 260 Axis control switching unit, 270 Speed control unit, 280 Current control unit, 290 Voltage command calculation unit, Iγ *, Iδ * Current command value, Iu, Iv, Iw Motor current, Iγ γ-axis current, Iδ δ-axis current, Nu, Nv, Nw Intermediate node, Q1 Q6 switching element, S1 to S6 control signal, Td dead time, Vγ *, Vδ * voltage command value (gamma-[delta] axis), Vu *, Vv *, Vw * voltage command value (three-phase), Vdc DC voltage.

Claims (8)

A motor control device in which the position sensor is not arranged,
DC power supply,
An inverter that converts a DC voltage from the DC power source into an AC voltage applied to the motor;
A current detector for detecting a motor current flowing through the motor;
A control unit that generates a control signal for controlling on / off of a plurality of semiconductor switching elements constituting the inverter;
The control unit has a timing at which the motor current decreases during the operation of increasing the rotation speed of the motor during feedback control of the motor current based on a detection value by the current detection unit, and after the timing, A motor control device that generates the control signal such that the motor current increases as the rotational speed increases . The control unit reduces the motor current when the rotational speed transitions from a region lower than a predetermined threshold to a region higher than the threshold, and increases the rotational speed in a region higher than the threshold. The motor control device according to claim 1, wherein the control signal is generated so that the motor current increases in accordance with the control signal.   The motor control device according to claim 2, wherein the control unit feedback-controls the motor current so that a magnetic flux vector in the motor is controlled to be constant in a region where the rotation speed is higher than the threshold value. The control unit generates the control signal so as to make the excitation component and the torque component of the motor current coincide with the excitation current command value and the torque current command value during the feedback control of the motor current,
The motor control device according to claim 3, wherein the controller sets the excitation current command value to a constant value in a region where the rotation speed is lower than the threshold value. The control unit generates the control signal so as to make the excitation component and the torque component of the motor current coincide with the excitation current command value and the torque current command value during the feedback control of the motor current,
The motor control device according to claim 1, wherein the control unit reduces the excitation current command value at a timing when the motor current decreases.   During the feedback control of the motor current, the control unit displays the excitation current command value in a region where the rotation speed is lower than a predetermined threshold, and the magnetic flux in the motor when the rotation speed is the threshold. The motor control device according to claim 5, wherein the motor control device is set to a value higher than the excitation current command value set by feedback control of the motor current for controlling the vector to be constant.   The motor control device according to claim 6, wherein the control unit sets the excitation current command value in a region where the rotation speed is lower than a predetermined threshold value during feedback control of the motor current to a constant value. . After the start control for increasing the rotational speed to a predetermined speed, the motor current is decreased during the operation for increasing the rotational speed of the motor from the predetermined speed by feedback control of the motor current. The motor control device according to claim 1, which generates a control signal.