Drive device, compressor drive system, and refrigeration cycle device

Abstract

A drive device (4) that drives an alternating current motor (1) by an inverter (11) includes a current detection unit (12) that detects a current flowing through the alternating current motor, a rotor position calculation unit (14) that calculates rotor position information which is information on the position of a rotor of the alternating current motor, a voltage command determination unit (15) that determines a voltage command based on the current vector of the current and the rotor position information, a beatless control unit (18) that suppresses pulsation of a current caused by periodic pulsation of a DC bus voltage by operating the voltage phase of the voltage command, and a resonance detection unit (20) that detects an inductor-capacitor resonance between an alternating current power supply (5) that supplies power to the inverter and the inverter. The beatless control unit reduces an operation amount of the voltage phase when the inductor-capacitor resonance is excited.

Description

Technical Field

[0001] The present disclosure relates to a drive device that drives a drive target using alternating current power, a compressor drive system, and a refrigeration cycle device.

Background Art

[0002] A drive device that combines an AC (Alternating Current) - DC (Direct Current) converter and a DC - AC converter (inverter) is frequently used to drive an alternating current motor. A capacitor is used in the DC bus section of this drive device. In addition, a reactor (inductance) is often inserted on the AC - DC converter side for the purpose of power factor improvement and boosting. For this drive device, miniaturization and downsizing of the capacitor and the reactor have been studied for the purpose of cost reduction.

[0003] However, when a capacitor or reactor with a small capacitance is used, the periodic pulsation of the DC bus voltage increases, which has an adverse effect on the current control of the alternating current motor. In particular, when the pulsation frequency (disturbance frequency) of the DC bus voltage and the frequency of the phase current of the alternating current motor are close to each other, a low - frequency current pulsation called beat vibration occurs. When the phase current pulsates due to beat vibration, demerits such as deterioration of motor efficiency due to an increase in the current peak value, a decrease in the maximum output of the motor due to the constraints of overcurrent protection, and an increase in the vibration or noise of the alternating current motor occur.

[0004] Various methods for beatless control, which is control for suppressing beat vibration, have been proposed so far. For example, a drive device that performs beatless control detects at least one of the DC bus voltage and the motor current, extracts the pulsation component included in the detected DC bus voltage or motor current by a band - pass filter or the like, and performs feedback control so that the extracted pulsation component is reduced, thereby suppressing beat vibration.

[0005] The beatless control of Patent Document 1 extracts the cosine component and sine component included in the pulsation of the q-axis current based on the principle of Fourier series, proportionally controls or integrally controls these components respectively, and then restores the control result to an AC signal to correct the dq-axis voltage output to the inverter.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0007] However, in the technology of the above Patent Document 1, in the voltage saturation region of the inverter voltage, a desired voltage operation cannot be performed, so there is a problem that the pulsation cannot be reduced as intended.

[0008] The present disclosure has been made in view of the above, and an object thereof is to obtain a drive device capable of reducing pulsation as intended even in the voltage saturation region of the inverter voltage.

Means for Solving the Problems

[0009] In order to solve the above-described problems and achieve the object, a drive device according to the present disclosure is a drive device that drives an alternating current motor by an inverter, and includes a current detection unit that detects a current flowing through the alternating current motor, and a rotor position calculation unit that calculates rotor position information that is information on the position of a rotor included in the alternating current motor. Further, the drive device according to the present disclosure includes a voltage command determination unit that determines a voltage command based on a current vector of a current and the rotor position information, and a beatless control unit that suppresses pulsation of a current caused by a periodic pulsation of a DC bus voltage by operating a voltage phase of the voltage command. Further, the drive device according to the present disclosure includes a resonance detection unit that detects an inductor-capacitor resonance between an alternating current power supply that supplies power to the inverter and the inverter. The beatless control unit reduces the operation amount of the voltage phase when the inductor-capacitor resonance is excited It includes a norm calculation unit that calculates the norm of the current vector, a pulsation extraction unit that extracts the pulsation component of the norm, a pulsation suppression unit that determines an operation amount by which the pulsation component is reduced, and a limit value determination unit that calculates a limit value for restricting the operation amount based on the inductor-capacitor resonance detected by the resonance detection unit. The pulsation suppression unit determines an operation amount by which the pulsation component is reduced within the range of the limit value.

Advantages of the Invention

[0010] The drive device according to the present disclosure has an effect that pulsation can be reduced as intended even in a voltage saturation region of an inverter voltage.

Brief Description of the Drawings

[0011]

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

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26

Figure 27

Figure 28

Figure 29

Figure 30

Figure 31

Figure 32

Figure 33

Figure 34

Embodiment for Carrying Out the Invention

[0012] Hereinafter, a driving device, a compressor drive system, and a refrigeration cycle device according to an embodiment of the present disclosure will be described in detail with reference to the drawings.

[0013] Embodiment 1. FIG. 1 is a diagram showing the configuration of a driving device according to Embodiment 1. The driving device 4 is a device that converts AC power into desired power to drive an AC motor 1. The rotating shaft of the AC motor 1 is mechanically connected to a mechanical device 2. The mechanical device 2 is, for example, a refrigerant compression mechanism. The compressor 3 is constituted by the AC motor 1 and the mechanical device 2. A system including the driving device 4 and the compressor 3 is a compressor drive system 800 described later.

[0014] Note that the compression mechanism is merely an example of the mechanical device 2 driven by the AC motor 1, and the driving device 4 can also be applied to other types of mechanical devices. The driving device 4 converts the AC power input from the AC power supply 5 and drives the AC motor 1. There is a power supply impedance (parasitic impedance) between the driving device 4 and the AC power supply 5. The power supply inductor 6 is the inductance component of the power supply impedance.

[0015] For convenience of explanation, here, the case where the AC power supply 5 as the input power supply is a three-phase AC will be described, but the driving device 4 of Embodiment 1 is also applicable when the AC power supply 5 is a single-phase AC.

[0016] The driving device 4 includes a diode rectifier 7, a DC reactor 8, a capacitor 9, a DC bus voltage detection unit 10, an inverter 11, a current detection unit 12, and a control unit 400.

[0017] In the drive device 4, an AC-DC converter is constituted by a diode rectifier 7, a DC reactor 8, a capacitor 9, etc. The drive device 4 converts the input AC voltage into a DC voltage by this AC-DC converter. Although the AC-DC converter shown in FIG. 1 is a very simple AC-DC converter, when power factor improvement or boosting is required, another type of AC-DC converter may be used.

[0018] One end of the DC reactor 8 is connected to the positive output point of the diode rectifier 7, and the other end of the DC reactor 8 is connected to the positive input point of the inverter 11. Also, one end of the capacitor 9 is connected to the other end of the DC reactor 8. The other end of the capacitor 9 is connected to the negative output point of the diode rectifier 7 and the negative input point of the inverter 11. The DC reactor 8 and the capacitor 9 are provided to smooth the DC power output from the diode rectifier 7.

[0019] The DC bus voltage detection unit 10 detects the voltage across both ends of the capacitor 9 as the DC bus voltage V DC and outputs it to the control unit 400. The inverter 11 converts the DC voltage into an AC voltage and drives the AC motor 1 with the AC voltage.

[0020] The current detection unit 12 detects the phase current (phase current vector I uvw ) flowing from the inverter 11 to the AC motor 1 and outputs it to the control unit 400. The control unit 400 performs a series of control operations for driving the AC motor 1. The detailed hardware configuration of the control unit 400 will be described later.

[0021] The control unit 400 includes a modulation unit 13, a rotor position calculation unit 14, a voltage command determination unit 15, a coordinate conversion unit 16, a coordinate conversion unit 17, a beatless control unit 18, an adder 19, and a resonance detection unit 20. The coordinate conversion unit 17 is the first coordinate conversion unit, and the coordinate conversion unit 16 is the second coordinate conversion unit.

[0022] The control unit 400 controls the AC motor 1 in the rotating two-phase coordinate system. Here, the case where the control unit 400 performs control in the dq rotating coordinate system based on the direction of the rotor magnet will be described, but the control unit 400 may perform control in a coordinate system other than the dq rotating coordinate system.

[0023] In order to perform control in the rotating two-phase coordinate system, the control unit 400 obtains the angular difference between the fixed two-phase coordinate and the rotating two-phase coordinate. The drive device 4 of Embodiment 1 calculates the dq-axis voltage command vector V * dq and the dq-axis current vector I dq to estimate the estimated magnetic pole position θ^ e and the estimated angular velocity ω^ e . Hereinafter, the information indicated by " * " is command information, and the information indicated by "^" is estimated information.

[0024] The dq-axis voltage command vector V * dq corresponds to the voltage applied by the inverter 11 to the AC motor 1. The dq-axis voltage command vector V * dq is the voltage command vector in the dq rotating coordinate system.

[0025] The dq-axis current vector I dq is the current vector actually detected by the current detection unit 12 and coordinate-converted into the dq rotating coordinate system. The estimated magnetic pole position θ^ e is the information obtained by estimating the magnetic pole position of the rotor included in the AC motor 1. The estimated angular velocity ω^ e is the information obtained by estimating the angular velocity of the rotor.

[0026] As a method for estimating the magnetic pole position of the rotor from the speed electromotive force generated while the AC motor 1 is rotating, there are various methods such as an adaptive flux observer and an extended induced voltage observer. Further, the control unit 400 may directly observe the magnetic pole position of the rotor using a position sensor such as an encoder or a resolver. The control unit 400 calculates the angular velocity of the rotor based on the magnetic pole position of the rotor.

[0027] The coordinate conversion unit 17 receives the phase current vector (three-phase current vector) I of the AC motor 1 from the current detection unit 12. uvw The coordinate conversion unit 17 receives the phase current vector I. uvw and converts it into the dq-axis current vector I. dq That is, the coordinate conversion unit 17 converts the three-phase current vector into two vectors (d-axis current vector and q-axis current vector) that are current vectors in the rotating two-phase coordinate system. The coordinate conversion unit 17 uses the estimated magnetic pole position θ^ e for this coordinate conversion. That is, the coordinate conversion unit 17 performs a rotating two-phase conversion based on the estimated magnetic pole position θ^ e which is the rotor position information. The coordinate conversion unit 17 outputs the dq-axis current vector I dq to the rotor position calculation unit 14, the voltage command determination unit 15, and the vectorless control unit 18.

[0028] The voltage command determination unit 15 determines the voltage command to be applied to the AC motor 1. That is, the voltage command determination unit 15 executes speed control calculation and current control calculation to determine the dq-axis voltage command vector V dq from the dq-axis current vector I. * dq The voltage command determination unit 15 determines, for example, by vector control, the d-axis current vector and the q-axis current vector by decomposing the dq-axis current vector I dq and determines the d-axis voltage command vector and the q-axis voltage command vector. Then, the voltage command determination unit 15 determines the dq-axis voltage command vector V * dq from the d-axis voltage command vector and the q-axis voltage command vector. The voltage command determination unit 15 determines the dq-axis voltage command vector V * e so that the speed command ω e matches the estimated angular velocity ω^ * dq sent from the rotor position calculation unit 14. The speed command ω * e is a command for the angular velocity of the rotor. The voltage command determination unit 15 obtains the speed command ω * e from the upper-level program used by the upper-level device of the drive device 4.to obtain. The voltage command determination unit 15 outputs the dq-axis voltage command vector V * dq to the rotor position calculation unit 14 and the coordinate conversion unit 16.

[0029] The rotor position calculation unit 14 calculates the position information of the rotor of the AC motor 1 (rotor position information). Specifically, the rotor position calculation unit 14 calculates the estimated magnetic pole position θ^ * dq and the estimated angular velocity ω^ dq from the dq-axis voltage command vector V e and the dq-axis current vector I e and estimates them.

[0030] In addition, when a position sensor for detecting the position of the rotor is arranged in the compressor 3, the rotor position calculation unit 14 estimates the estimated magnetic pole position θ^ e and the estimated angular velocity ω^ e based on the position of the rotor detected by the position sensor. The rotor position calculation unit 14 outputs the estimated magnetic pole position θ^ e to the coordinate conversion unit 17 and the adder 19. Further, the rotor position calculation unit 14 outputs the estimated angular velocity ω^ e to the voltage command determination unit 15.

[0031] The beatless control unit 18 suppresses the pulsation of the motor current (the current of the AC motor 1) caused by the periodic pulsation of the DC bus voltage V DC by operating the voltage phase of the voltage command to the inverter 11. The beatless control unit 18 extracts the component of the disturbance frequency f dis included in the dq-axis current vector I dq based on the disturbance frequency f dis input by the user, and determines the operation amount (phase change amount) θ b of the voltage phase to reduce the extracted component. The disturbance frequency f dis is the pulsation frequency of the DC bus voltage V DC . The beatless control unit 18 outputs the operation amount θ b of the voltage phase to the adder 19. In this way, the beatless control unit 18 outputs the operation amount θ bBy outputting this, the voltage phase of the voltage command to the inverter 11 is manipulated, thereby suppressing the pulsation of the motor current.

[0032] At this time, the resonance detection unit 20 detects the occurrence of LC (inductor-capacitor) resonance based on the pulsation of the DC bus voltage V DC and, when detecting the occurrence of resonance, transmits a resonance detection signal indicating that resonance has occurred to the beatless control unit 18. LC resonance is the inductor-capacitor resonance between the AC power supply 5 that supplies power to the inverter 11 and the inverter 11. The resonance detection unit 20 detects the occurrence of resonance based on the DC bus voltage V DC and a preset disturbance frequency f dis . Details of the internal configuration and effects of the beatless control unit 18 will be described later.

[0033] The adder 19 determines the phase angle θ^ e by adding the operation amount θ b of the voltage phase sent from the beatless control unit 18 to the estimated magnetic pole position θ^ eb sent from the rotor position calculation unit 14. That is, the adder 19 determines the sum of the operation amount θ b of the voltage phase and the estimated magnetic pole position θ^ e as the phase angle θ^ eb . The phase angle θ^ eb is the magnetic pole position of the rotor adjusted to suppress the pulsation corresponding to the disturbance frequency f dis . The adder 19 sends the phase angle θ^ eb to the coordinate conversion unit 16.

[0034] The coordinate conversion unit 16 converts the dq-axis voltage command vector V eb using the phase angle θ^ * dq into a three-phase voltage command vector V * uvw . That is, the coordinate conversion unit 16 generates the three-phase voltage command vector V eb based on the magnetic pole position (phase angle θ^ * uvw ) of the rotor adjusted to suppress the pulsation. Specifically, the coordinate conversion unit 16 uses the phase angle θ^eb Using this, the voltage phase of the dq-axis voltage command vector V * dq is manipulated, and the manipulated dq-axis voltage command vector V * dq is converted into a three-phase voltage command vector V * uvw . The coordinate conversion unit 16 outputs the three-phase voltage command vector V * uvw to the modulation unit 13.

[0035] The modulation unit 13 determines a PWM (Pulse Width Modulation) signal for operating the inverter 11. Specifically, the modulation unit 13 determines the PWM signal based on the DC bus voltage V DC sent from the DC bus voltage detection unit 10 and the three-phase voltage command vector V * uvw sent from the coordinate conversion unit 16, and outputs it to the inverter 11.

[0036] The inverter 11 outputs a voltage corresponding to the PWM signal to the AC motor 1. As a result, the AC motor 1 is driven by the drive device 4. Generally, when the DC reactor 8 or the capacitor 9 is reduced in capacity, the DC bus voltage V DC tends to pulsate greatly. When the input is a three-phase AC power supply, it is known that the disturbance frequency, which is the pulsation frequency of the DC bus voltage V DC , is six times the power supply frequency. Also, in the case of a single-phase AC power supply, it is known that the disturbance frequency, which is the pulsation frequency of the DC bus voltage V DC , is twice the power supply frequency. Furthermore, harmonic pulsations of these integral multiple frequencies occur. Due to such pulsations of the DC bus voltage V DC , pulsations occur in the motor current.

[0037] When viewed in terms of the dq-axis current, the frequency of this current pulsation coincides with the disturbance frequency and the frequencies that are integral multiples of the disturbance frequency. When this current pulsation is observed on the three-phase coordinates, the sum and difference frequencies |f dis of the disturbance frequency f e and the frequency f dis of the phase current of the AC motor 1e |becomes the current pulsation of the. This current pulsation has a frequency of the difference |f dis -f e |is likely to become apparent when is small.

[0038] DC bus voltage V DC The resulting phase current pulsation of this low frequency (disturbance frequency f dis ) is called beat vibration, and various countermeasures have been studied for a long time. When the phase current pulsates due to beat vibration, demerits such as deterioration of motor efficiency due to an increase in the current peak value, a decrease in the maximum output of the AC motor 1 due to constraints on overcurrent protection, and an increase in vibration or noise of the AC motor 1 occur.

[0039] The drive device 4 of Embodiment 1 automatically minimizes beat vibration in order to suppress the pulsation of the motor current. The drive device 4 minimizes beat vibration even in the voltage saturation region of the inverter voltage where the amplitude of the voltage command cannot be operated by operating the voltage phase.

[0040] FIG. 2 is a diagram for explaining beat vibration when the drive device according to Embodiment 1 does not perform voltage phase operation. FIG. 3 is a diagram for explaining the principle of suppressing beat vibration by voltage phase operation of the drive device according to Embodiment 1.

[0041] The horizontal axis of the graphs shown in FIGS. 2 and 3 is the d-axis, and the vertical axis is the q-axis. In FIGS. 2 and 3, the voltage that the drive device 4 outputs on average is represented by (v d0 , v q0 ), and the current that the drive device 4 outputs on average is represented by (i d0 , i q0 ). That is, on average, the drive device 4 outputs voltages of v d0 and v q0 , and it is assumed that currents of i d0 and i q0 are flowing through the AC motor 1.

[0042] When the inverter voltage, which is the voltage of the inverter 11, is saturated, the inverter voltage is the DC bus voltage VDC It expands and contracts as shown in Fig. 2 due to the pulsation (disturbance voltage VD). As a result, the dq-axis current draws an elliptical locus centered at the point (i d0 , i q0 ). That is, the current locus Ita, which is the locus of the dq-axis current, becomes an elliptical locus. The larger this elliptical current locus Ita of the dq-axis current becomes, the larger the beat vibration appearing in the phase current becomes.

[0043] In Embodiment 1, since the drive device 4 cannot operate the amplitude of the voltage command in the voltage saturation region of the inverter voltage, it operates the voltage phase to reduce the beat vibration.

[0044] Generally, it is known that if an appropriate change is given to the voltage phase with respect to the dq-axis voltage command, the elliptical locus of the dq-axis current becomes smaller. Although various voltage phase operation type beatless controls have been studied, there has been no study on what voltage phase change (the operation amount θ b ) should be given in order to efficiently reduce the locus of the elliptical current locus Ita of the dq-axis current.

[0045] The drive device 4 efficiently reduces the elliptical current locus Itb of the dq-axis current as shown in Fig. 3 by appropriately changing the voltage phase. In Fig. 3, the case is shown where the drive device 4 changes the voltage phase by the phase change amount Pc, the locus of the voltage of the drive device 4 becomes the voltage locus Vt, and the elliptical current locus of the dq-axis current becomes the current locus Itb.

[0046] Further, the drive device 4 drives the AC motor 1 so that the major and minor axis directions of the elliptical current locus Itb of the dq-axis current become the angles desired by the user in order to enhance the effect of the beatless control.

[0047] In Embodiment 1, in order for the drive device 4 to reduce the elliptical current locus Itb of the dq-axis current and make the major and minor axis directions the desired angles set by the user, the configuration of the beatless control unit 18 is set to the configuration shown in Fig. 5 described later.

[0048] In a conventional drive device, since the power supply inductance 6 is sufficiently smaller than the DC reactor 8, the size of the power supply inductance 6 has rarely been a problem. However, as the DC reactor 8 is miniaturized, the influence of the power supply inductance 6 cannot be ignored. As described above, although various studies have been conducted on the voltage phase operation type beatless control, in recent years, the downsizing of the DC reactor 8 and the capacitor 9 has further advanced, and there have been an increasing number of cases where a sufficient pulsation reduction effect cannot be obtained by the conventional method.

[0049] Also, when a plurality of devices with a low input impedance, such as the drive device 4 of Embodiment 1, are connected in parallel to the AC power supply 5, the apparent power supply impedance increases, so the influence of the power supply inductance 6 becomes even greater. This is because, when viewed from the drive device 4, the other devices (such as other drive devices) connected in parallel appear to perform the same function as the power supply impedance. The more the number of such devices connected in parallel increases, the larger the power supply inductance 6 becomes. And when the power supply inductance 6 becomes large, the LC resonance frequency decreases, which has an adverse effect on the beatless control.

[0050] In the beatless control described with reference to FIG. 3, the drive device 4 changes the voltage phase in synchronization with the disturbance angular frequency ω dis =2πf dis to change the dq-axis voltage and suppress the beat vibration. In this case, when the LC resonance angular frequency ω LC of the system including the drive device 4 and the AC power supply 5 is close to the disturbance angular frequency ω dis , the LC resonance is excited by performing the beatless control. When the LC resonance is excited, the pulsation of the DC bus voltage V DC increases. As a result, the current pulsation increases, so the effect of the beatless control decreases.

[0051] Even in such a case, if it is desired to make the current ripple zero, it is conceivable to use integral control. However, when using integral control, in order to cancel the current ripple, the operation amount of the beatless control is further increased, resulting in an even worse cycle of increasing the ripple of the DC bus voltage V DC and falling into a vicious cycle. Due to this vicious cycle, the beatless control may become unstable.

[0052] Generally, the power supply impedance changes over time and is an unknown value, and it is difficult to know its exact value in real time. In order to pursue further reduction in the capacitance of the DC reactor 8 or the capacitor 9, a beatless control that operates reliably regardless of the magnitude of the power supply impedance is required.

[0053] The drive device 4 of Embodiment 1 changes the behavior of the beatless control when it detects that the LC resonance angular frequency ω LC is approaching the disturbance angular frequency ω dis and the LC resonance is excited by the voltage operation by the beatless control. Thereby, the drive device 4 prevents an excessive increase in the ripple of the DC bus voltage V DC and reduces the ripple components included in the power supply current and the motor current. In order to realize such reduction of the ripple components, in Embodiment 1, the drive device 4 is provided with a resonance detection unit 20.

[0054] FIG. 4 is a diagram showing a configuration example of the resonance detection unit included in the drive device according to Embodiment 1. The resonance detection unit 20 has a pulsation amplitude calculation unit 401 and a comparator 402. Here, the case where the resonance detection unit 20 determines the presence or absence of LC resonance based on the pulsation amplitude of the DC bus voltage V DC will be described.

[0055] In addition, when a current sensor is arranged in the immediate vicinity of the AC-DC converter, the DC bus voltage V DCAlternatively, the pulsating amplitude of the current flowing through the AC-DC converter may be detected. In this case, the resonance detection unit 20 determines the presence or absence of LC resonance based on the pulsating amplitude of the current flowing through the AC-DC converter. For example, a current sensor is attached to any of the power lines connected to the diode rectifier 7, and the presence or absence of LC resonance is determined from the current detected by this current sensor.

[0056] The pulsating amplitude calculation unit 401 receives the DC bus voltage V DC from the DC bus voltage detection unit 10 and also receives the disturbance frequency f dis input by the user. The pulsating amplitude calculation unit 401 extracts the pulsating component included in the DC bus voltage V dis based on the disturbance frequency f DC . At this time, the frequency components extracted by the pulsating amplitude calculation unit 401 are the component of the disturbance frequency (fundamental frequency) f DC of the DC bus voltage V dis and the harmonic components that are integer multiples of the disturbance frequency f dis .

[0057] When the beatless control excites the LC resonance, the pulsations of these frequency components increase. The pulsating amplitude calculation unit 401 may extract the pulsating component included in the DC bus voltage V DC by any extraction method.

[0058] The pulsating amplitude calculation unit 401 extracts the pulsating component included in the DC bus voltage V DC , for example, using the principle of Fourier series. Also, the pulsating amplitude calculation unit 401 may extract the pulsating component using a band-pass filter. Here, the pulsating amplitude of the DC bus voltage V DC extracted by the pulsating amplitude calculation unit 401 is represented by the symbol |V dcDis |. The pulsating amplitude calculation unit 401 outputs the extracted |V dcDis | to the beatless control unit 18 and the comparator 402.

[0059] The comparator 402 compares |V dcDisCompare it with a preset threshold value and output a detection flag, which is the comparison result, to the beatless control unit 18. The detection flag is a binary signal of on or off, and dcDis when |V dcDis is greater than the threshold value, it indicates on, and

[0060] in Embodiment 1, the pulsation amplitude of the DC bus voltage V DC is denoted as |V dcDis and a signal including at least one of |V dcDis and the detection flag is called a resonance detection signal. That is, it is sufficient that at least one of |V

[0061] and the detection flag is included in the resonance detection signal.

[0062] Based on the resonance detection signal, the beatless control unit 18 can diagnose the occurrence situation of LC resonance. When a current sensor (not shown in FIG. 1) is used, the pulsation amplitude of the current will be detected. In that case, the pulsation amplitude of the current may be included in the resonance detection signal. Further, the resonance detection unit 20 may compare the pulsation amplitude of the current with a preset threshold value and output a detection flag, which is the comparison result, to the beatless control unit 18.

[0062] FIG. 5 is a diagram showing the configuration of the beatless control unit included in the drive device according to Embodiment 1. The beatless control unit 18 includes a norm calculation unit 101, a weight coefficient setting unit 102, a pulsation extraction unit 103, a pulsation suppression unit 104, and a limit value determination unit 105. The dq-axis current vector I dq is input to the norm calculation unit 101, and the disturbance frequency f dis stored in advance is input to the pulsation extraction unit 103 and the pulsation suppression unit 104.

[0063] The norm calculation unit 101 calculates the norm (absolute value) or the weighted norm of the dq-axis current vector I dq . The beatless control unit 18 is the dq-axis current vector I dqIt is only necessary to have a weight coefficient setting unit 102 when calculating the weighted norm, and when calculating the norm of the dq-axis current vector I dq (when not calculating the weighted norm), it is not necessary to have the weight coefficient setting unit 102. The weight coefficient setting unit 102 stores the weight coefficients w1 and w2 to be set by the user, and sets the weight coefficients w1 and w2 for the norm calculation unit 101.

[0064] The weight coefficient w1 is the weight for the d-axis current, and the weight coefficient w2 is the weight for the q-axis current. The weights of the d-axis current and the q-axis current are adjusted according to the ratio of the weight coefficients w1 and w2. There are various types of norms, and the most well-known norm is the L2 norm calculated using the following formula (1).

[0065]

Equation

[0066] However, i d and i q in Equation (1) are the dq-axis currents (d-axis current and q-axis current). The norm calculation unit 101 calculates the norm |I dq |2 using Equation (1). |I dq |2 corresponds to the absolute value of the dq-axis current vector I dq . Note that the norm calculation unit 101 may use the L1 norm, L∞ norm, etc. instead of the L2 norm.

[0067] Also, the norm calculation unit 101 may calculate the weighted norm |I dqw | using the following formula (2) to which the weight coefficients w1 and w2 are applied. In this case, the norm calculation unit 101 calculates |I dqw | using the weight coefficients w1 and w2 stored in the weight coefficient setting unit 102.

[0068]

Equation

[0069] The weighting coefficients w1 and w2 in Equation (2) shall be any values equal to or greater than zero. In Equation (2), i d 2 and i q 2 may be replaced by i d 3 and i q 3 may be applied, or i d 4 and i q 4 etc. may be applied. Also, |I dq |2 is equal to |I dqw | when w1 = w2 = 1. FIG. 5 shows the configuration of the beatless control unit 18 when the norm calculation unit 101 calculates |I dqw | and outputs it to the pulsation extraction unit 103.

[0070] The beatless control unit 18 of Embodiment 1 reduces the pulsation component of the norm (|I dq |2 or |I dqw |) calculated by the norm calculation unit 101. The effect obtained by the beatless control changes depending on which norm pulsation component the beatless control unit 18 suppresses.

[0071] For example, when it is desired to suppress the peak value of the phase current, the beatless control unit 18 greatly suppresses the pulsation of |I dq |2. Also, when it is desired to suppress the vibration and noise of the AC motor 1, the beatless control unit 18 sets w1 = 0 and w2 = 1 to suppress the pulsation of the q-axis current. Further, when the beatless control unit 18 aims for an intermediate state between these, the weighting coefficients w1 and w2 may be changed such as w1 = 0.5 and w2 = 0.5. Also, the beatless control unit 18 may change the weighting coefficients w1 and w2 such as w1 = 0.25 and w2 = 0.75, or w1 = 0.75 and w2 = 0.25. Further, in the case where the d-axis current is larger than the q-axis current, the beatless control unit 18 may set w1 = 1 and w2 = 0 to suppress the pulsation of the d-axis current.

[0072] The operation in the norm operation unit 101 is an operation for setting the major and minor axes of the elliptical current locus Itb of the dq-axis current to the angles desired by the user, and is an operation for effectively applying the beatless control.

[0073] The weight coefficient setting unit 102 sets weight coefficients w1 and w2 according to the purpose of the beatless control (what control effect is desired by the beatless control). The weight coefficients w1 and w2 may be arbitrarily set by the user of the drive device 4. In the following, there may be cases where an unweighted norm is described, but the norm may be a weighted norm. The weight coefficient setting unit 102 may adjust the weight coefficients w1 and w2 based on the dq-axis current vector I dq and the rotational speed of the AC motor 1, etc.

[0074] The pulsation extraction unit 103 extracts the pulsation component (periodic pulsation) of the norm (the unweighted norm |I dqw | or the weighted norm |I dqw |) calculated by the norm operation unit 101. The frequency components extracted by the pulsation extraction unit 103 at this time are the components of the disturbance frequency f dis and the harmonic components that are integer multiples of the disturbance frequency f dis (fundamental frequency). The pulsation extraction unit 103 may extract the pulsation component of the norm by any extraction method.

[0075] The pulsation extraction unit 103 separates, for example, the sin component and the cos component of the pulsation of the norm using the principle of Fourier series to extract the pulsation component of the norm. Also, the pulsation extraction unit 103 may extract the pulsation component using a band-pass filter. In the following, the case where the pulsation extraction unit 103 extracts the pulsation component of the norm using the principle of Fourier series will be described.

[0076] Here, the sin component and the cos component of the pulsation of the norm extracted by the pulsation extraction unit 103 are represented by the symbols y sin and y cos respectively. The pulsation extraction unit 103 outputs y sin and y cos to the pulsation suppression unit 104.

[0077] The limit value determination unit 105 receives a resonance detection signal from the resonance detection unit 20. The limit value determination unit 105 determines a limit value for beatless control based on the resonance detection signal. The limit value is determined based on a manipulated variable θ b The limit value determining unit 105 outputs the limit value to the pulsation suppressing unit 104.

[0078] When beatless control excites LC resonance, even if the current pulsation can be reduced to zero, a good current waveform cannot be obtained. For this reason, the driving device 4 of the first embodiment uses the manipulated variable θ b The limit value for reducing the LC resonance is determined, and beatless control is executed using the limit value. As a result, the drive unit 4 can obtain a good current waveform even when the beatless control excites LC resonance, thereby improving motor efficiency and reducing noise.

[0079] The pulsation suppression unit 104 executes beatless control based on the limit value. That is, the pulsation suppression unit 104 performs beatless control based on the limit value. sin , y cos The voltage phase control amount θ that reduces b Calculate.

[0080] Fig. 6 is a diagram for explaining a first method for determining a limit value determined by the drive device according to the first embodiment. Fig. 7 is a diagram for explaining a second method for determining a limit value determined by the drive device according to the first embodiment. The horizontal axis of the graphs shown in Fig. 6 and Fig. 7 is time, and the vertical axis is limit value.

[0081] 6 and 7, a case will be described where the resonance detection signal is detected at time t1. When the limit value determination unit 105 detects the resonance detection signal, it sets the limit value Θ lim In Figures 6 and 7, the initial limit value Θ lim Θ LB and the reduced limit value Θlim is represented by Θ LS as shown.

[0082] In FIG. 6, when the resonance detection signal is detected at the time t1 when the LC resonance occurs, the limit value determination unit 105 steps the limit value Θ lim is represented by Θ LB from Θ LS to Θ

[0083] In FIG. 7, from the time t1 when the resonance detection signal is detected when the LC resonance occurs to the time t2 after a certain time has elapsed, the limit value determination unit 105 decreases the limit value Θ in a ramp shape lim is represented by Θ LB from Θ LS to Θ

[0084] In this way, the limit value determination unit 105 may decrease the limit value Θ stepwise as shown in FIG. 6 lim or may decrease the limit value Θ in a ramp shape as shown in FIG. 7. Further, the limit value determination unit 105 may smoothly decrease the limit value Θ using a low-pass filter (LPF: Low Pass Filter) or the like. Note that the time t2 and the limit value Θ lim after the decrease may be freely set by the user of the drive device 4. lim after the decrease may be freely set by the user of the drive device 4. lim which is Θ LS after the decrease may be freely set by the user of the drive device 4.

[0085] The limit value determination unit 105 may automatically adjust the limit value Θ by feedback control. FIG. 8 is a diagram showing a configuration example of the limit value determination unit included in the beatless control unit according to Embodiment 1. lim The limit value determination unit 105 may automatically adjust the limit value Θ by feedback control. FIG. 8 is a diagram showing a configuration example of the limit value determination unit included in the beatless control unit according to Embodiment 1.

[0086] The limit value determination unit 105 includes a subtractor 701 and a PID (Proportional-Integral-Differential) controller 702 with a limiter. The limit value determination unit 105 receives the pulsating amplitude |V of the DC bus voltage V DC from the resonance detection unit 20dcDis Obtain a resonance detection signal including |. Further, the limit value determination unit 105 obtains the command value |V DC of the DC bus voltage V * dcDis from the upper program used by the upper device of the drive device 4. In FIGS. 1 and 5, the illustration of the command value |V DC of the DC bus voltage V * dcDis is omitted.

[0087] The limit value determination unit 105 determines the limit value Θ DC of the pulsation amplitude |V dcDis of the DC bus voltage V * dcDis so that it matches the command value |V lim for beatless control. Specifically, the subtractor 701 calculates the deviation between the command value |V * dcDis and the pulsation amplitude |V dcDis and outputs the deviation to the PID controller 702. The PID controller 702 determines the limit value Θ * dcDis so that the deviation between the command value |V dcDis and the pulsation amplitude |V lim becomes zero. That is, the PID controller 702 performs PID control using the deviation from the subtractor 701 to determine the limit value Θ lim and outputs the limit value Θ lim to the pulsation suppression unit 104. Here, a PID controller 702 with a limiter is used for feedback control, but as long as the limit processing can be appropriately performed, another type of controller that performs I control, PI control, etc. may be used.

[0088] FIG. 9 is a diagram for explaining the limit processing executed by the drive device according to Embodiment 1. Here, the method for determining the limit value Θ lim when the PID controller 702 is a PID controller with a limiter will be described. The horizontal axis of the graph shown in FIG. 9 is the input value, and the vertical axis is the output value. The input value here is the limit value Θ limand the output value is the limit value Θ after the PID controller 702 performs the limit process lim That is. The PID controller 702 outputs a limit value Θ that does not exceed the threshold value lim .

[0089] When the PID controller 702 is a PID controller with a limiter, since the PID controller 702 has an integrator inside, if a limiter is not inserted in the final output stage part or the integration process part, the output may diverge. The limit value Θ in the beatless control lim must be zero or more. Therefore, when the output of the PID controller 702 or the output of the integrator becomes a value of zero or less, the PID controller 702 clamps those outputs to zero

[0090] Also, since the PID controller 702 must also avoid the output of the PID controller 702 or the output of the integrator becoming larger than a specific value, when the output value becomes larger than the specific value, the output is clamped with a value smaller than the specific value. In FIG. 9, when the output from the PID controller 702 or the output of the integrator of the PID controller 702 is the limit value Θ lim which is Θ LB in the initial state, when these outputs exceed Θ LB they are clamped at Θ LB . In this way, the PID controller 702 limits the limit value Θ lim so that the PID controller 702 does not become less than zero and does not exceed Θ

[0091] When the PID controller 702 automatically adjusts the limit value Θ as described in FIGS. 6 and 7, when the pulsation amplitude |V lim | is smaller than the command value |V dcDis | (normal time), due to the action of the integrator arranged inside the PID controller 702 with a limiter, the limit value Θ of the beatless control * dcDis becomes equal to Θ lim in the initial state. On the other hand, when the pulsation amplitude |V LB | is the command value |V dcDis |* dcDis When it is larger (during LC resonance detection), the PID controller 702 sets |V dcDis | = |V * dcDis | until the limit value Θ of the beatless control reaches lim and decreases it. The pulsation amplitude |V dcDis | and the command value |V * dcDis | are the same size, the PID controller 702 sets |V dcDis | = |V * dcDis | and does not change the limit value Θ lim to maintain it.

[0092] In this way, the drive device 4 of the first embodiment detects that LC resonance is excited by voltage operation with beatless control, and reduces the limit value Θ lim of the beatless control. The pulsation suppression unit 104 calculates the operation amount θ lim of the voltage phase such that y sin , y cos is reduced within the range of the limit value Θ b .

[0093] Here, the calculation performed by the pulsation suppression unit 104 will be described in detail. FIG. 10 is a diagram showing the configuration of the pulsation suppression unit included in the beatless control unit according to the first embodiment. The pulsation suppression unit 104 includes subtractors 201A and 201B, a rotation calculation unit 202, integral control units (integral controllers) 203A and 203B, an AC restoration unit 204, and a rotation amount adjustment unit 205. The integral control unit 203A is the first integral control unit, and the integral control unit 203B is the second integral control unit. The rotation calculation unit 202 may be arranged inside the pulsation extraction unit 103. * between the target value r * of the pulsation component and the cos component of the norm, and rotates the difference between the target value r b of the pulsation component and the sin component of the norm to calculate the operation amount θ

[0094] of the voltage phase.

[0095] To the pulsation suppression unit 104, the target value r of the pulsation component stored in advance * and y which is the cos component of the pulsation of the norm or the weighted norm cos , and y which is the sin component sin and the disturbance frequency f dis are input. Usually, the target value r of the pulsation component * is set to zero, but the target value r of the pulsation component * may be set to a non-zero value (a value other than zero).

[0096] The subtractor 201A calculates the difference between the target value r of the pulsation component * and y which is the cos component of the pulsation of the norm or the weighted norm cos , and outputs the calculation result as the deviation e of the cos component cos to the rotation calculation unit 202 and the rotation amount adjustment unit 205.

[0097] The subtractor 201B calculates the difference between the target value r of the pulsation component * and y which is the sin component of the pulsation of the norm or the weighted norm sin , and outputs the calculation result as the deviation e of the sin component sin to the rotation calculation unit 202 and the rotation amount adjustment unit 205.

[0098] The rotation calculation unit 202 performs a rotation calculation on e cos and e sin using the following formula (3). Here, the cos component after the rotation calculation is denoted as e Rcos , and the sin component after the rotation calculation is denoted as e Rsin .

[0099]

Equation

[0100] θ in formula (3) R is the rotation amount of the rotation calculation. The rotation amount adjustment unit 205 sets the rotation amount θ R set by the user in advance to the rotation calculation unit 202. Also, the rotation amount adjustment unit 205 uses the deviation e of the cos component cos, and the deviation e of the sin component sin is used to correct the rotation amount θ R The rotation amount adjustment unit 205 outputs the rotation amount θ R to the rotation calculation unit 202.

[0101] The rotation calculation unit 202 receives the rotation amount θ R from the rotation amount adjustment unit 205. The rotation calculation unit 202 uses the rotation amount θ R received from the rotation amount adjustment unit 205 to perform a rotation calculation on e cos and e sin .

[0102] The integral control unit 203A integrates the deviation e cos corresponding to y Rcos (the rotated cos component), and the integral control unit 203B integrates the deviation e sin corresponding to y Rsin (the rotated sin component). That is, the integral control unit 203A determines x Rcos by performing integral control on e cos , and the integral control unit 203B determines x Rsin by performing integral control on e sin . x cos , x sin are the cos component and the sin component of the output signal of the beatless control, respectively. The integral control units 203A and 203B accumulate the error between the target value r * of the pulsation component and the actual value, and perform integral control by adding an amount proportional to this accumulated value to the operation amount θ b of the voltage phase.

[0103] The integral control unit 203A performs integral control so that e Rcos , which is the rotated cos component, becomes zero. That is, the integral control unit 203A determines x * from e cos so that the deviation e cos , which is the difference between the target value r Rcos of the pulsation component and the cos component y cos of the norm or weighted norm of the pulsation, becomes zero. That is, the integral control unit 203A is ycos Integral control is performed so that it approaches the target value r of the pulsation component, and x * is determined. cos

[0104] Also, the integral control unit 203B performs integral control so that e, which is the rotated sin component, Rsin becomes zero. That is, the integral control unit 203B performs integral control so that the target value r of the pulsation component * and y, which is the sin component of the pulsation of the norm or weighted norm, sin the deviation e, which is the difference between them, sin becomes zero, and x is determined from e. That is, the integral control unit 203B performs integral control so that y Rsin approaches the target value r of the pulsation component, and x sin is determined. sin * sin

[0105] Here, the pulsation suppression unit 104 uses the integral control units 203A and 203B, but the pulsation suppression unit 104 may use other control units. For example, the pulsation suppression unit 104 may use a control unit that performs P (Proportional) control, a control unit that performs PI (Proportional-Integral) control, or a control unit that performs PID control.

[0106] The integral control units 203A and 203B operate within the range of the limit value Θ. lim In other words, the integral control units 203A and 203B determine x and x within the range of Θ lim ≧√(x cos 2 +x sin 2 ). When the limit value Θ cos is lowered, the integral control units 203A and 203B determine x and x so that Θ sin lim lim ≧√(x cos 2 +x sin 2 ). cos sin ​​​​​​​Lower it. In Embodiment 1, two controllers (integral control units 203A and 203B) for the cos component and the sin component are arranged in the pulsation suppression unit 104, and the limit value Θ lim How to distribute it between the integral control unit 203A, which is the cos-side controller, and the integral control unit 203B, which is the sin-side controller, may be arbitrarily determined by the designer or user of the drive device 4. Information on how to distribute it is set in advance by the designer or user in the integral control units 203A and 203B.

[0107] The AC restoration unit 204 converts the output of the integral control into AC and determines the operation amount θ of the voltage phase b That is, the AC restoration unit 204 determines the operation amount θ of the voltage phase based on x cos x sin and f dis Based on this, the operation amount θ of the voltage phase b is calculated.

[0108] Here, the calculation formula for the operation amount θ of the voltage phase to be optimized will be described. First, the conversion result of converting the disturbance frequency f b into the angular frequency is defined as the disturbance angular frequency ω dis Let it be ω dis The relational expression between ω dis and f dis is given by the following formula (4).

[0109]

Equation

[0110] Here, the integral result of integrating ω dis with respect to time t is represented by the symbol θ dis

[0111]

Equation

[0112] Here, since f dis can be regarded as a constant, θ dis ​can be expressed as a first-order function of time t. At this time, the operation amount θ of the voltage phase b can be expressed, for example, by the following formula (6).

[0113] [Equation]

[0114] In formula (6), x cos , x sin are the cosine component and sine component of the output signal of the beatless control, respectively. The AC restoration unit 204 applies x cos , x sin , and f dis to calculate the operation amount θ of the voltage phase b . By applying f dis to formulas (4) to (5), a sine wave and a cosine wave corresponding to f dis are derived, and the operation amount θ of the voltage phase corresponding to f dis is derived b .

[0115] FIG. 11 is a diagram showing the operation pattern of the deviation vector when the beatless control unit according to Embodiment 1 executes beatless control when the rotation amount is in an optimal state. FIG. 12 is a diagram showing the operation pattern of the deviation vector when the beatless control unit according to Embodiment 1 executes beatless control when the rotation amount is in a state within the range of less than ±90 degrees from the optimal value. FIG. 13 is a diagram showing the operation pattern of the deviation vector when the beatless control unit according to Embodiment 1 executes beatless control when the rotation amount is in a state away from the optimal value by more than ±90 degrees.

[0116] When the rotation amount θ R is in the range of less than ±90 degrees from the optimal value, it is the case where the absolute value of the difference between the rotation amount θ R and the optimal value is less than 90 degrees. When the rotation amount θ R is in a state away from the optimal value by more than ±90 degrees, it is the case where the absolute value of the difference between the rotation amount θ R and the optimal value is greater than 90 degrees.

[0117] In FIGS. 11 to 13, the horizontal axis is the cos component (e cos ), and the vertical axis is the sin component (e sin ). In FIGS. 11 to 13, when the beatless control unit 18 executes beatless control, the deviation vector trajectory Et of the deviation vector E (not shown) composed of e cos and e sin is shown. The deviation vector E is a vector directed from the origin to (e cos , e sin ). The beatless control unit 18 controls so that (e cos , e sin ) approaches the origin, that is, the absolute value of the deviation vector E becomes smaller. The deviation vector trajectory Et has "Start" as the starting point and "Goal" as the ending point. The operation pattern of the deviation vector E by beatless control (the deviation vector trajectory Et corresponding to the behavior of the control unit 400) can be roughly classified into four types according to the rotation amount θ R .

[0118] The first operation pattern shown in FIG. 11 is when the rotation amount θ R is in the optimal state. When the beatless control unit 18 starts beatless control in this state, the deviation vector E composed of e cos and e sin approaches the origin at the shortest distance by the action of integral control.

[0119] The second operation pattern shown in FIG. 12 is when the rotation amount θ R is within the range of less than ±90 degrees from the optimal value. When the beatless control unit 18 starts beatless control at this time, the deviation vector E approaches the origin while drawing a spiral.

[0120] The third operation pattern shown in FIG. 13 is the rotation amount θ Rwhen it is more than ±90 degrees away from the optimum value. At this time, when the beatless control unit 18 starts beatless control, the deviation vector E moves away from the origin while drawing a spiral.

[0121] A fourth operation pattern (not shown) is when the rotation amount θ R is in the worst state (180 degrees away from the optimum value). At this time, when the beatless control unit 18 starts beatless control, the deviation vector E moves away from the origin in a straight line.

[0122] As described above, if an appropriate rotation amount θ R is given, the pulsation suppression unit 104 can control the deviation e cos and the deviation e sin to zero. However, if the rotation amount θ R is inappropriate, the deviation e cos, e sin will diverge.

[0123] When an integral control unit is not arranged in the beatless control unit 18 (for example, when P control is used), although the deviation e cos , e sin will not diverge, it is the same as the case where an integral control unit is arranged in the beatless control unit 18 in that the control effect decreases when the rotation amount θ R is inappropriate.

[0124] The drive device 4 of the first embodiment calculates and adjusts the optimum value of the changed rotation amount θ R even when the optimum value of the rotation amount θ R changes due to a change in the power impedance. Whether the beatless control excites the LC resonance is determined by the power impedance. For this reason, when the beatless control excites the LC resonance, the pulsation amplitude of the DC bus voltage V DC increases. Along with this, the optimum value of the rotation amount θ R changes. Therefore, the rotation amount θ RWhen it is set as a fixed value, depending on the value of the power supply impedance, the beatless control may become unstable as shown in FIG. 13. As described above, the power supply impedance changes over time and has an unknown value, and it is difficult to know its exact value in real time. Therefore, in the conventional technology, the optimal value of the rotation amount θ R cannot be calculated.

[0125] In the beatless control unit 18 of Embodiment 1, when y cos , which is the cos component of the pulsation of the norm of the dq-axis current or the weighted norm, or y sin , which is the sin component, increases unintentionally, this increase is detected, and the rotation amount θ R is adjusted. Thereby, the beatless control unit 18 can always suppress the beat vibration regardless of the power supply impedance.

[0126] FIG. 14 is a diagram showing an operation pattern of a deviation vector when the beatless control unit according to Embodiment 1 executes beatless control. The horizontal axis in FIG. 14 is the cos component (e cos ), and the vertical axis is the sin component (e sin ). In FIG. 14, the deviation vector locus Et, which is the operation pattern of the deviation vector E (not shown) composed of e cos and e sin when the beatless control unit 18 executes beatless control, is shown. The deviation vector locus Et has "Start" as the starting point and "Goal" as the ending point.

[0127] In the drive device 4 of Embodiment 1, when y cos , which is the cos component of the pulsation of the norm of the dq-axis current or the weighted norm, or y sin , which is the sin component, increases unintentionally, this increase is detected. When the rotation amount adjustment unit 205 detects the unintentional increase in y cos and y sin , it determines that the beatless control is in an abnormal state (the rotation amount θ R is in an inappropriate state), and corrects the rotation amount θ R . The rotation amount adjustment unit 205 corrects ycos and y sin By finding a search direction that reduces

[0128] Here, the concept of the search direction modified by the beatless control unit 18 will be described. FIG. 15 is a first explanatory diagram for explaining the search direction modified by the beatless control unit according to the first embodiment. FIG. 16 is a second explanatory diagram for explaining the search direction modified by the beatless control unit according to the first embodiment. FIG. 17 is a third explanatory diagram for explaining the search direction modified by the beatless control unit according to the first embodiment. FIG. 18 is a fourth explanatory diagram for explaining the search direction modified by the beatless control unit according to the first embodiment.

[0129] In FIGS. 15 to 18, an image of the search direction modified by the beatless control unit 18 is shown. The horizontal axis in FIGS. 15 to 18 is the cos component (e cos , e Rcos , x cos ), and the vertical axis is the sin component (e sin , e Rsin , x sin ). In FIGS. 15 to 17, compared with FIGS. 11 to 13, a deviation vector locus ERt which is an operation pattern (locus) of the deviation vector ER (not shown) after the rotation operation, an output signal locus xt which is an operation pattern of the output signal vector x (not shown), and a sector region representing an image of the search direction are added. The deviation vector ER after the rotation operation is a vector of the rotation result obtained by rotating the deviation vector E before the rotation operation by the rotation operation unit 202. The output signal locus xt is the locus of the output signal vector x of the beatless control.

[0130] The deviation vector ER is a vector directed from the origin to (e Rcos , e Rsin ), and the deviation vector E is, as described above, a vector directed from the origin to (e cos , e sin ). Here, the case where the ideal values of the deviation vector E before the rotation operation and the deviation vector ER after the rotation operation are zero (the origin) will be described.

[0131] In FIGS. 15 to 18, the image of the search direction is shown by a fan-shaped figure (sector area) imitating the human visual field. When the center of the sector area is the position of the output signal vector x at a certain moment, the arc portion of the sector area is the image of the forward visual field at that moment.

[0132] Specifically, in FIG. 15, the image of the search direction is shown by the search direction image SD1, and in FIG. 16, the images of the search direction are shown by the search direction images SD2 and SD3. Further, in FIG. 17, the images of the search direction are shown by the search direction images SD4, SD5, and SD6, and in FIG. 18, the images of the search direction are shown by the search direction images SD7 and SD8.

[0133] The search direction image SD1 shown in FIG. 15 is the search direction image when the rotation amount θ R is the optimal value. When the rotation amount θ R is optimal, the search direction image SD1 does not need to be changed from the start to the completion of the search.

[0134] The search direction images SD2 and SD3 shown in FIG. 16 are the search direction images when the rotation amount θ R is an appropriate value. The search direction image SD2 is the image of the search direction at the start of the search, and the search direction image SD3 is the image of the search direction when the search can be completed. That is, the search direction image SD3 is the search direction image after a specific time has elapsed since the start of the search using the search direction image SD2.

[0135] The search direction images SD4, SD5, and SD6 shown in FIG. 17 are the search direction images when the rotation amount θ RIt is an image of the search direction when the value is inappropriate. The search direction image SD4 is the image of the search direction at the start of the search, the search direction image SD5 is the image of the search direction during the search, and the search direction image SD6 is the image of the search direction when the search fails. That is, the search direction image SD5 is the search direction image after a specific time has elapsed since the start of the search using the search direction image SD4. The search direction image SD6 is the search direction image after a specific time has elapsed since the search using the search direction image SD5.

[0136] In FIGS. 15 to 17, the trajectory of the deviation vector E before the rotation operation as it attempts to move from the point of the initial value Es of the deviation vector E to the point of the ideal value Ei of the deviation vector E (the origin) is illustrated as the deviation vector trajectory Et.

[0137] Also, in FIGS. 15 to 17, the trajectory of the deviation vector ER after the rotation operation as it attempts to move from the point of the initial value ERs of the deviation vector ER to the point of the ideal value Ei of the deviation vector ER (the origin) is illustrated as the deviation vector trajectory ERt.

[0138] Also, in FIG. 18, as shown in FIG. 17, the rotation amount θ R is inappropriate and the search fails. After the rotation amount θ R is adjusted, the trajectory of the deviation vector ER as it attempts to move from the point of the initial value ERs of the deviation vector ER to the point of the ideal value Ei of the deviation vector ER (the origin) is illustrated as the deviation vector trajectory ERt2.

[0139] Here, it is assumed that if the ideal value of the output signal vector x can be searched, the deviation vector E before the rotation operation and the deviation vector ER after the rotation operation can be set to zero. However, the beatless control unit 18 cannot know in advance where the ideal value of the output signal vector x is in the figure.

[0140] The beatless control unit 18 can determine whether or not it has succeeded in searching for the ideal value of the output signal vector x only by observing the deviation vectors E and ER. However, the search range of the output signal vector x is limited. That is, the output signal vector x is provided with a searchable range SR which is the range where searching is possible. When the ideal value of the output signal vector x is not within the searchable range SR, the deviation vectors E and ER cannot be set to zero. Therefore, the beatless control unit 18 searches for the output signal vector x that minimizes the deviation vectors E and ER within the searchable range SR.

[0141] The initial value xs of the output signal vector x is an arbitrary value and is often set to zero, but here it is set to a non-zero value for convenience of drawing. Also, the initial values of the deviation vectors E and ER are set to arbitrary non-zero values. In FIGS. 15 to 17, the initial value of the deviation vector E is indicated by the initial value Es, and the initial value of the deviation vector ER is indicated by the initial value ERs. Note that the initial value ERs of the deviation vector ER may be the same as the initial value Es of the deviation vector E.

[0142] In FIG. 15, the beatless control unit 18 according to Embodiment 1 R represents the operation patterns of the deviation vectors E and ER and the output signal vector x when beatless control is executed when the rotation amount θ

[0143] The deviation vector locus Et which is the locus of the deviation vector E and the deviation vector locus ERt which is the locus of the deviation vector ER after rotation operation are R different in phase by the amount of rotation θ R When the rotation amount θ is in an ideal state, the deviation vectors E and ER move from the initial values Es and ERs toward the ideal value Ei (the origin), which is the ideal point of the deviation vectors E and ER, at the shortest distance. For this purpose, the output signal vector x also needs to move from the initial value xs toward the ideal value xi, which is the ideal point of the output signal vector x, at the shortest distance.

[0144] Since the output signal vector x is a vector obtained by integrating the deviation vector ER, if the direction of the deviation vector ER does not match the moving direction of the ideal output signal vector x, it is impossible for the output signal vector x to move at the shortest distance. In this case, since the parameter for adjusting the direction of the deviation vector ER is the rotation amount θ R if the beatless control unit 18 makes the rotation amount θ R optimal, it can move the deviation vectors E and ER to the ideal value Ei at the shortest distance.

[0145] Since the output signal vector x moves along the direction of the deviation vector ER seen from the origin, the direction of the deviation vector ER seen from the origin can be considered as the search direction of the output signal vector x. Alternatively, it can be said that the search direction of the beatless control is approximately the traveling direction of the output signal vector x.

[0146] The sector region (search direction image SD1) shown in FIG. 15 corresponds to the visual field when the rotation amount θ R is the optimal value. In this case, since the ideal value xi of the output signal vector x is on the extension line of the search direction, the search for the ideal value xi of the output signal vector x succeeds smoothly.

[0147] The sector regions (search direction images SD2 and SD3) shown in FIG. 16 correspond to the visual field when the rotation amount θ R is within the range of ±90 degrees from the optimal value. In this case, there is a slight deviation between the direction of the ideal value xi of the output signal vector x and the search direction. Since the internal integral control (integral control by the integral control units 203A and 203B) in the beatless control unit 18 allows a slight deviation in the search direction, the beatless control unit 18 can finally reach the ideal value Ei for the deviation vectors E and ER. However, the deviation vectors E and ER will decrease in a spiral shape because the output signal vector x does not head towards the ideal value xi of the output signal vector x at the shortest distance.

[0148] The sector regions (search direction images SD4, SD5, and SD6) shown in FIG. 17 are for the rotation amount θR This corresponds to the field of view when it is more than ±90 degrees away from the optimum value. In this case, there is a large deviation between the direction of the ideal value xi of the output signal vector x and the search direction. In this case, since the deviation vectors E and ER do not decrease as intended even when the output signal vector x is changed, the search direction is not uniquely determined, and the output signal vector x is manipulated so that the deviation vectors E and ER point to a point different from the ideal value Ei. For this reason, the beatless control unit 18 increases the deviation vectors E and ER in a trajectory such that they draw a spiral, and the beatless control fails.

[0149] In FIG. 17, the points of the deviation vectors E and ER when it is detected that the search has failed are indicated by the deviation vectors Ef and ERf, respectively. Also, in FIG. 17, the point of the output signal vector x when it is detected that the search has failed is indicated by the arrival value xf.

[0150] In the first embodiment, in preparation for the case where the deviation vectors E and ER perform the operation as shown in FIG. 17, the beatless control unit 18 is provided with a search failure detection unit that detects that the search for the optimum value of the rotation amount θ R has failed. The reason for the failure of the search in the operation described with reference to FIG. 17 is that the search direction was inappropriate (in other words, the rotation amount θ R was inappropriate). When the search fails, the beatless control unit 18 detects the failure of the search and corrects the search direction, so that eventually the search for the optimum value of the rotation amount θ R can succeed.

[0151] In FIG. 18, an operation image is shown when, after the search for the optimum value of the rotation amount θ R has failed, the search direction is corrected and the search is executed again. That is, the behavior shown in FIG. 18 is that when the beatless control unit 18 detects the failure of the search, it adjusts the rotation amount θ R and adjusts the rotation amount θ RThis is the behavior when reset to an appropriate value. Note that, as in Fig. 17, in Fig. 18, the points of the deviation vectors E and ER when it is detected that the search has failed are indicated by the deviation vectors Ef and ERf, respectively.

[0152] In Fig. 18, the rotation amount before correction is indicated by the rotation amount Bθ R and the rotation amount after correction is indicated by the rotation amount Aθ R Also, in Fig. 18, for the search direction image SD7 when it is detected that the search has failed, the beatless control unit 18 indicates the search direction image after adjusting the rotation amount Bθ R by the search direction image SD8. When the beatless control unit 18 adjusts the rotation amount Bθ R , the rotation amount becomes the rotation amount Aθ R and the deviation vector ER and the search direction after the rotation operation change.

[0153] In Fig. 18, the locus of the output signal vector x after the rotation amount Bθ R is adjusted is indicated by the locus xt2. Also, in Fig. 18, the locus of the deviation vector ER rotationally calculated by the adjusted rotation amount Aθ R is indicated by the deviation vector locus ERt2. Also, in Fig. 18, the locus of the deviation vector E rotationally calculated by the adjusted rotation amount Aθ R is indicated by the deviation vector locus Et2.

[0154] If the rotation amount Aθ R becomes the optimum value by adjusting the rotation amount Bθ R , the ideal value xi of the output signal vector x comes on the extension line of the search direction, so the search for the optimum value is completed smoothly.

[0155] Note that, in Fig. 18, for convenience of explanation, the case where the rotation amount θ R is adjusted once and the rotation amount θ R is adjusted to the optimum value has been described. However, in reality, since the optimum value of the rotation amount θ R is unknown, the adjustment of the rotation amount θ R is performed little by little in multiple times by the integral control units 203A and 203B.

[0156] Also, the beatless control unit 18 always corrects the rotation amount θ during the operation of the beatless control by a method combining the outer product operation and PID control described later. R It may continue to be corrected. Due to such a correction process of the rotation amount θ R the deviation vector E may approach the origin through a complex deviation vector locus Et as shown in FIG. 14. Also, as shown in FIG. 18, after the correction of the rotation amount θ R the deviation vector E may head toward the origin along a linear deviation vector locus Et2.

[0157] In beatless control, it is unknown where the ideal value xi of the output signal vector x is located. If the output signal vector x is operated in an inappropriate direction, the beat vibration will increase. Therefore, the search for the ideal value xi of the output signal vector x needs to be carried out carefully. For this reason, in Embodiment 1, the beatless control unit 18 appropriately corrects the search direction of the beatless control so that the output signal vector x can surely reach the ideal value xi.

[0158] The rotation amount adjustment unit 205 may adjust the rotation amount θ R by any method. The rotation amount adjustment unit 205 corrects (adjusts) the rotation amount θ R by, for example, a method combining the outer product operation and PID control described later. Also, the rotation amount adjustment unit 205 may automatically search for the rotation amount θ R using AI (Artificial Intelligence) or machine learning.

[0159] Here, the rotation amount θ RAn example of the adjustment method will be described. FIG. 19 is a diagram showing a first behavior of a deviation vector when the beatless control unit according to Embodiment 1 executes beatless control. FIG. 20 is a diagram showing a second behavior of a deviation vector when the beatless control unit according to Embodiment 1 executes beatless control. FIG. 21 is a diagram showing a third behavior of a deviation vector when the beatless control unit according to Embodiment 1 executes beatless control.

[0160] The horizontal axis of FIGS. 19 to 21 is the cos component (e cos ), and the vertical axis is the sin component (e sin ). In FIGS. 19 to 21, the behavior of the deviation vector E composed of e cos and e sin when the pulsation suppression unit 104 executes beatless control is schematically shown.

[0161] The pulsation suppression unit 104 can determine whether the beatless control is properly performed by examining the deviation vector E and the time differential vector (d / dt)E of the deviation vector E. Hereinafter, the time differential vector (d / dt)E of the deviation vector E may be referred to as the time differential vector (d / dt)E.

[0162] When the deviation vector E and the time differential vector (d / dt)E are out of phase as in the first behavior of the deviation vector E shown in FIG. 19, the pulsation suppression unit 104 determines that the beatless control is being performed well.

[0163] Also, when the direction of the time differential vector (d / dt)E is inside (the origin side) with respect to the perpendicular line of the deviation vector E as in the second behavior of the deviation vector E shown in FIG. 20, the pulsation suppression unit 104 determines that the beatless control is being performed appropriately to some extent.

[0164] When the direction of the time differential vector (d / dt)E is outside with respect to the perpendicular line of the deviation vector E as in the third behavior of the deviation vector E shown in FIG. 21, the pulsation suppression unit 104 determines that the beatless control is not being performed appropriately.

[0165] If the state of the deviation vector E shown in FIG. 21 is left unchanged, the beatless control becomes unstable and diverges. Therefore, the pulsation suppression unit 104 adjusts the rotation amount θ so that the deviation vector E assumes the state shown in FIG. 19 or FIG. 20. R to correct it.

[0166] In addition, as shown in FIG. 20, when the direction of the time differential vector (d / dt)E is more inward than the perpendicular line of the deviation vector E, the pulsation suppression unit 104 adjusts the rotation amount θ so that the direction of the time differential vector (d / dt)E becomes even more inward. R may be corrected.

[0167] The pulsation suppression unit 104 R evaluates whether the beatless control is being properly performed using a quantitative numerical value (evaluation value) in order to correct the rotation amount θ. The pulsation suppression unit 104 corrects the rotation amount θ with a correction amount corresponding to the evaluation value. R to correct it.

[0168] FIG. 22 is a diagram for explaining an evaluation value for evaluating whether the pulsation suppression unit according to Embodiment 1 is properly performing beatless control. The horizontal axis in FIG. 22 is the cos component, and the vertical axis is the sin component.

[0169] FIG. 22 shows an example of the definition of the evaluation value. The pulsation suppression unit 104 evaluates the validity of the operation of the beatless control by using, for example, the cross product of the deviation vector E and the time differential vector (d / dt)E of the deviation vector E (the area of the parallelogram formed by the two vectors). Here, the evaluation value for evaluating whether the beatless control is being properly performed is denoted by C d as shown. The cross product of the deviation vector E and the time differential vector (d / dt)E of the deviation vector E is C d as described above.

[0170] The smaller the area of the parallelogram formed by the deviation vector E and the time differential vector (d / dt)E of the deviation vector E, the smaller the rotation amount θ RReduce the amount of correction. As a result, the closer the angle formed by the deviation vector E and the time derivative vector (d / dt)E is to 180 degrees or 0 degrees, the smaller the rotation amount θ R of the correction amount becomes. That is, the pulsation suppression unit 104 increases the rotation amount θ as the angle formed by the deviation vector E and the time derivative vector (d / dt)E approaches 90 degrees R of the correction amount.

[0171] FIG. 23 is a diagram showing the configuration of the rotation amount adjustment unit included in the pulsation suppression unit according to Embodiment 1. The rotation amount adjustment unit 205 of Embodiment 1 includes an outer product calculation unit 300, a dead zone 305, a PID control unit 306, and an adder 309.

[0172] The outer product calculation unit 300 includes pseudo differentiators 302A and 302B, multipliers 303A and 303B, and a subtractor 304.

[0173] Note that the symbol s shown in FIG. 23 is a Laplace operator. For the rotation amount adjustment unit 205, a differentiator without a low-pass filter (LPF) may be used instead of the pseudo differentiators 302A and 302B. However, in FIG. 23, the case where the pseudo differentiators 302A and 302B with a low-pass filter added to remove differential noise are used will be described.

[0174] To the outer product calculation unit 300, the deviation e cos is input from the subtractor 201A, and the deviation e sin is input from the subtractor 201B.

[0175] The outer product calculation unit 300 calculates the evaluation value C d by the above-described outer product calculation. In the outer product calculation unit 300, the deviation e cos is input to the pseudo differentiator 302A and the multiplier 303B, and the deviation e sin is input to the pseudo differentiator 302B and the multiplier 303A.

[0176] The pseudo differentiator 302A differentiates the deviation e cos with respect to time t and passes it through a low-pass filter to obtain (d / dt)ecos is calculated and output to the multiplier 303A. The multiplier 303A multiplies e sin by (d / dt)e cos and outputs the multiplication result to the subtractor 304.

[0177] The pseudo differentiator 302B differentiates the deviation e sin with respect to time t and passes it through a low-pass filter to calculate (d / dt)e sin and outputs it to the multiplier 303B. The multiplier 303B multiplies e cos by (d / dt)e sin and outputs the multiplication result to the subtractor 304.

[0178] The subtractor 304 calculates the evaluation value C by subtracting the multiplication result output from the multiplier 303B from the multiplication result output from the multiplier 303A. The subtractor 304 outputs C d to the dead zone 305. d

[0179] The dead zone 305 stops adjusting the rotation amount θ R after the beatless control converges to the final value. Adjusting the rotation amount θ R more than necessary is not preferable in terms of the stability of the beatless control. Therefore, after the beatless control converges to the final value, the rotation amount adjustment unit 205 in Embodiment 1 stops adjusting the rotation amount θ R by the dead zone 305. Note that the rotation amount adjustment unit 205 using the dead zone 305 is an example, and other circuits or the like other than the dead zone 305 may be used as a mechanism for stopping the adjustment of the rotation amount θ R

[0180] The PID control unit 306 performs PID control on the signal output from the dead zone 305 and outputs the rotation amount θ R to the adder 309. Note that the adjustment of the rotation amount θ R may be performed using any circuit. The adjustment of the rotation amount θ RThe adjustment may be performed by, for example, a PI control unit that executes PI control. If better control results are expected, the rotation amount θ may be adjusted using a different type of control unit or AI. R It may be adjusted.

[0181] The adder 309 adds the default value (for example, 90 degrees) stored in advance to the rotation amount θ sent from the PID control unit 306 to correct the rotation amount θ. R The adder 309 outputs the rotation amount θ, which is the addition result, to the rotation calculation unit 202. R The adder 309 outputs the rotation amount θ, which is the addition result, to the rotation calculation unit 202. R The adder 309 outputs the rotation amount θ, which is the addition result, to the rotation calculation unit 202.

[0182] Thus, the rotation amount adjustment unit 205 of Embodiment 1 determines the rotation amount θ based on the sum of the default value and the output from the PID control unit 306. R Thus, the rotation amount adjustment unit 205 of Embodiment 1 determines the rotation amount θ based on the sum of the default value and the output from the PID control unit 306.

[0183] When the rotation amount adjustment unit 205 changes the rotation amount θ, the operation of the beatless control is optimized and the evaluation value C decreases. R When the rotation amount adjustment unit 205 changes the rotation amount θ, the operation of the beatless control is optimized and the evaluation value C decreases. d When the evaluation value C decreases to a specific value, the change in the rotation amount θ stops. d At this time, since the rotation amount θ is an appropriate value, the deviation vector E converges to zero. R At this time, since the rotation amount θ is an appropriate value, the deviation vector E converges to zero. R At this time, since the rotation amount θ is an appropriate value, the deviation vector E converges to zero.

[0184] The reason for aiming to minimize the pulsation of the norm or weighted norm in Embodiment 1 is that the minimization is the best state that can be achieved in the voltage saturation region (inverter overmodulation region) of the inverter voltage. When the DC bus voltage V pulsates, pulsations occur in both the d-axis and q-axis currents. However, to suppress both pulsations of the d-axis and q-axis currents simultaneously, both the amplitude and phase of the voltage must be manipulated. This is obvious from the perspective of control freedom. In a state where only the phase of the voltage can be controlled, such as during inverter overmodulation, there is only one parameter that the control unit 400 can control. Therefore, in such a situation, the control unit 400 aims to minimize the pulsation of the norm or weighted norm by controlling the phase of the voltage. DC When the DC bus voltage V pulsates, pulsations occur in both the d-axis and q-axis currents. However, to suppress both pulsations of the d-axis and q-axis currents simultaneously, both the amplitude and phase of the voltage must be manipulated. This is obvious from the perspective of control freedom. In a state where only the phase of the voltage can be controlled, such as during inverter overmodulation, there is only one parameter that the control unit 400 can control. Therefore, in such a situation, the control unit 400 aims to minimize the pulsation of the norm or weighted norm by controlling the phase of the voltage.

[0185] In this way, the control unit 400 minimizes the pulsation of an arbitrary norm or weighted norm by appropriately operating the phase of the voltage. However, when attempting to determine (search for) the operation amount θ of the optimal voltage phase using integral control, control divergence as described with reference to FIG. 13 may occur. As the DC reactor 8 or the capacitor 9 is reduced in capacitance, it becomes more susceptible to the influence of the power supply impedance, and the beatless control becomes increasingly unstable. To prevent such destabilization of the control, the control unit 400 of Embodiment 1 automatically adjusts the rotation amount θ. b When attempting to determine (search for) the operation amount θ of the optimal voltage phase using integral control, control divergence as described with reference to FIG. 13 may occur. As the DC reactor 8 or the capacitor 9 is reduced in capacitance, it becomes more susceptible to the influence of the power supply impedance, and the beatless control becomes increasingly unstable. To prevent such destabilization of the control, the control unit 400 of Embodiment 1 automatically adjusts the rotation amount θ. R to automatically adjust.

[0186] Since the pulsation component of the norm or weighted norm of the current can always be made zero, LC resonance may be excited by beatless control. Therefore, the control unit 400 of Embodiment 1 detects LC resonance and reduces the limit value of beatless control. As a result, even in a case where the LC resonance angular frequency ω LC is close to the disturbance angular frequency ω dis , the control unit 400 can reduce the influence of LC resonance and obtain a good current waveform.

[0187] By configuring the drive device 4 as described above, the drive device 4 can cope with a wide range of power supply environments, and thus can accurately suppress the pulsation of the phase current due to beat vibration. The effect of beatless control varies depending on which norm pulsation the drive device 4 reduces. By detecting LC resonance and reducing the limit value of beatless control, the control unit 400 can prevent, for example, deterioration of motor efficiency due to an increase in the current peak value, a decrease in the maximum output of the AC motor 1 due to constraints on overcurrent protection, and an increase in vibration and noise of the AC motor 1.

[0188] FIG. 24 is a flowchart showing the processing procedure of the control process executed by the control unit of the drive device according to Embodiment 1. The control unit 400 is the phase current (phase current vector I flowing through the AC motor 1 detected by the current detection unit 12 uvwis acquired from the current detection unit 12 (step S10). Subsequently, the control unit 400 obtains the DC bus voltage V, which is the voltage across the capacitor 9 detected by the DC bus voltage detection unit 10 DC from the DC bus voltage detection unit 10 (step S20).

[0189] Thereafter, the coordinate conversion unit 17 performs a coordinate conversion operation on the current (step S30). That is, the coordinate conversion unit 17 converts the phase current vector I received from the current detection unit 12 uvw using the estimated magnetic pole position θ^ e into the dq-axis current vector I dq by coordinate conversion.

[0190] The rotor position calculation unit 14 calculates the rotor position (step S40). As a result, the rotor position calculation unit 14 obtains the position information and speed information of the rotor of the AC motor 1. Specifically, the rotor position calculation unit 14 estimates the estimated magnetic pole position θ^ * dq which is the position information of the rotor, and the estimated angular velocity ω^ dq which is the speed information of the rotor, from the dq-axis voltage command vector V e and the dq-axis current vector I e

[0191] The voltage command determination unit 15 calculates a voltage command for rotating the AC motor 1 at a desired speed and torque (step S50). Specifically, the voltage command determination unit 15 calculates the dq-axis voltage command vector V * e from the dq-axis current vector I e so that the speed command ω dq and the estimated angular velocity ω^ * dq match.

[0192] The vectorless control unit 18 executes a vectorless control operation (step S60). As a result, the vectorless control unit 18 determines the operation amount θ b of the voltage phase. Specifically, the vectorless control unit 18 determines the disturbance frequency f dq contained in the dq-axis current vector I​dis Extract the components of dis and determine the operation amount θ of the voltage phase for reducing the extracted components b Determine b .

[0193] The coordinate conversion unit 16 performs coordinate conversion on the voltage command to a value on the three-phase coordinates (step S70). Specifically, the coordinate conversion unit 16 uses the phase angle θ^ which is the sum of the operation amount θ of the voltage phase and the estimated magnetic pole position θ^ b and the estimated magnetic pole position θ^ e to convert the dq-axis voltage command vector V eb into the three-phase voltage command vector V * dq * uvw DC Convert.

[0194] The modulation unit 13 executes a modulation operation (step S80). Specifically, the modulation unit 13 determines a PWM signal based on the DC bus voltage V DC and the three-phase voltage command vector V * uvw and gives the PWM signal to the inverter 11.

[0195] Next, the operation of the beatless control unit 18 will be described. FIG. 25 is a flowchart showing the processing procedure of the beatless control process executed by the beatless control unit of the drive device according to Embodiment 1.

[0196] In the beatless control unit 18, the weight coefficient setting unit 102 sets the weight coefficients w1 and w2 stored in advance to the norm calculation unit 101 (step S110). The norm calculation unit 101 executes a norm calculation (step S120). That is, the norm calculation unit 101 calculates the norm or weighted norm of the dq-axis current vector I dq

[0197] The pulsation extraction unit 103 executes a pulsation extraction operation of the norm or weighted norm calculated by the norm calculation unit 101 (step S130). That is, the pulsation extraction unit 103 extracts the components of the disturbance frequency f dis and the disturbance frequency f dis ​Based on the harmonic components that are integer multiples of, the pulsation components included in the norm or weighted norm are extracted. The pulsation extraction unit 103 extracts the sin component and cos component of the pulsation of the norm, and outputs them to the pulsation suppression unit 104 as y sin , y cos respectively.

[0198] The limit value determination unit 105 executes a limit value calculation based on the resonance detection signal (step S140). The pulsation suppression unit 104 executes a pulsation suppression calculation (step S150). That is, the pulsation suppression unit 104 calculates the operation amount θ sin , y cos of the voltage phase such that y b is reduced within the range of the limit value.

[0199] Next, the operation of the pulsation suppression unit 104 will be described. FIG. 26 is a flowchart showing the processing procedure of the pulsation suppression process executed by the pulsation suppression unit of the driving device according to Embodiment 1. The pulsation suppression unit 104 determines the point where the pulsation of the norm or weighted norm is minimized according to the following procedure.

[0200] The subtractor 201A calculates the deviation e cos of the cos component, and the subtractor 201B calculates the deviation e sin of the sin component (step S210). The rotation amount adjustment unit 205 sets the rotation amount θ R to the rotation calculation unit 202, and corrects the rotation amount θ cos using the deviation e sin of the cos component and the deviation e R of the sin component (step S220). The rotation amount adjustment unit 205 sets the adjusted rotation amount θ R to the rotation calculation unit 202.

[0201] The rotation calculation unit 202 performs a rotation calculation of the vector using the rotation amount θ R (step S230). Specifically, the rotation calculation unit 202 applies e cos , e sin , and the rotation amount θ R to Equation (3) to obtain the cos component e Rcosand e, which is the sin component after the rotation operation Rsin are calculated.

[0202] The integral control units 203A and 203B execute integral control calculations (step S240). Specifically, the integral control unit 203A integrates e Rcos after the rotation operation to determine x cos which is the cos component of the output signal of the beatless control, and the integral control unit 203B integrates e Rsin after the rotation operation to determine x sin which is the sin component of the output signal of the beatless control.

[0203] The AC restoration unit 204 executes an AC restoration calculation (step S250). Specifically, the AC restoration unit 204 applies x cos , x sin , and f dis to equations (4) to (6) to calculate the operation amount θ b of the voltage phase. The drive device 4 drives the AC motor 1 using the operation amount θ b of the voltage phase, making it possible to effectively suppress the beat vibration of the current even when the operating conditions change significantly.

[0204] Next, the operation of the rotation amount adjustment unit 205 will be described. FIG. 27 is a flowchart showing the processing procedure of the rotation amount adjustment process executed by the rotation amount adjustment unit according to Embodiment 1.

[0205] The outer product calculation unit 300 calculates C d , which is an evaluation value for determining whether the beatless control is being properly performed, by outer product calculation (step S320). That is, the outer product calculation unit 300 calculates the evaluation value C cos of the cos component deviation e calculated by the subtractor 201A sin and the deviation e calculated by the subtractor 201B d .

[0206] The dead zone 305 of the rotation amount adjustment unit 205 executes dead zone processing (step S330). The PID control unit 306 of the rotation amount adjustment unit 205 performs PID control calculation (step S340), and the evaluation value C d is made zero for the rotation amount θ R to be corrected.

[0207] The adder 309 executes addition processing (step S370). Specifically, the adder 309 adds a default value (for example, 90 degrees) to the rotation amount θ R sent from the PID control unit 306 to adjust the rotation amount θ R thereby.

[0208] Next, the hardware configuration of the control unit 400 provided in the drive device 4 will be described. FIG. 28 is a diagram showing an example of the hardware configuration for realizing the control unit provided in the drive device according to the first embodiment.

[0209] The control unit 400 is realized by a processor 91, a memory 92, and peripheral devices 93. Note that, not only in the first embodiment but also in other embodiments, the control unit 400 is realized by a processor 91, a memory 92, and peripheral devices 93.

[0210] Each function of the control unit 400 is realized by software, firmware, or a combination of software and firmware. The software or firmware is described as a control program and stored in the memory 92. In the processing circuit that realizes the control unit 400, the processor 91 reads and executes the control program stored in the memory 92 to realize each function. This control program may be provided by a computer-readable recording medium on which the control program is recorded, or may be provided by other means such as a communication medium. It can also be said that the control program is a program that causes the control unit 400 to execute the processing of steps S10 to S80 in FIG. 24.

[0211] The processor 91 is a CPU (Central Processing Unit, also known as a central processing unit, processing unit, arithmetic unit, microprocessor, microcomputer, processor, or DSP (Digital Signal Processor)), or a system LSI (Large Scale Integration).

[0212] The memory 92 can exemplify non-volatile or volatile semiconductor memories such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), and EEPROM (registered trademark) (Electrically Erasable Programmable Read Only Memory). Further, the memory 92 is not limited thereto and may be a magnetic disk, optical disk, compact disk, mini disk, or DVD (Digital Versatile Disc).

[0213] The peripheral device 93 is, for example, a PWM pulse generation circuit, an analog-digital conversion circuit, an encoder counter, etc. The PWM pulse generation circuit is arranged in the modulation unit 13. The PWM pulse generation circuit is used for driving the inverter 11 and the AC-DC converter.

[0214] The analog-digital conversion circuit is arranged in the modulation unit 13, the coordinate conversion unit 17, etc. The analog-digital conversion circuit is used for, for example, detecting the DC bus voltage V DC and the phase current of the AC motor 1.

[0215] The encoder counter is arranged in the rotor position calculation unit 14 when the position of the rotor is detected by a position sensor arranged in the compressor 3. The encoder counter is used for, for example, acquiring rotor position information.

[0216] Thus, according to Embodiment 1, the control unit 400 of the drive device 4 calculates the operation amount θ of the voltage phase to minimize the pulsation component, so that even in the voltage saturation region of the inverter voltage where the amplitude of the voltage command cannot be operated, the pulsation can be reduced as intended. b Also, since the control unit 400 reduces the limit value of the beatless control when detecting LC resonance, even in the case where the LC resonance angular frequency ω LC is close to the disturbance angular frequency ω dis , the influence of LC resonance can be reduced, and a good current waveform can be obtained in various power supply environments. Thereby, the control unit 400 can effectively suppress the beat vibration caused by the pulsation of the DC bus voltage V DC without performing complicated control adjustment in various power supply environments.

[0217] Also, since the control unit 400 can effectively suppress the beat vibration, the DC reactor 8 and the capacitor 9 can be reduced in size and capacitance. Thereby, the manufacturing cost of the drive device 4 can be reduced, and the energy saving performance can be improved. LC is close to the disturbance angular frequency ω dis , the influence of LC resonance can be reduced, and a good current waveform can be obtained in various power supply environments. Thereby, the control unit 400 can effectively suppress the beat vibration caused by the pulsation of the DC bus voltage V DC without performing complicated control adjustment in various power supply environments. dis is close to the disturbance angular frequency ω dis , the influence of LC resonance can be reduced, and a good current waveform can be obtained in various power supply environments. Thereby, the control unit 400 can effectively suppress the beat vibration caused by the pulsation of the DC bus voltage V DC without performing complicated control adjustment in various power supply environments. DC is close to the disturbance angular frequency ω dis , the influence of LC resonance can be reduced, and a good current waveform can be obtained in various power supply environments. Thereby, the control unit 400 can effectively suppress the beat vibration caused by the pulsation of the DC bus voltage V DC without performing complicated control adjustment in various power supply environments.

[0218] Also, since the control unit 400 can effectively suppress the beat vibration, the DC reactor 8 and the capacitor 9 can be reduced in size and capacitance. Thereby, the manufacturing cost of the drive device 4 can be reduced, and the energy saving performance can be improved.

[0219] Also, the control unit 400 can prevent deterioration of motor efficiency due to an increase in the current peak value, a decrease in the maximum output of the motor due to constraints on overcurrent protection, an increase in vibration and noise of the AC motor 1, etc. by suppressing the beat vibration.

[0220] Also, the control unit 400 calculates the operation amount θ of the voltage phase to minimize the extracted pulsation component, and controls the inverter 11 using this operation amount θ, so that the reduction effect of the current pulsation can be easily stabilized regardless of the operating conditions of the drive device 4 and the installation conditions of the drive device 4. Therefore, the control unit 400 can effectively suppress the beat vibration of the current even if the operating conditions and installation conditions of the drive device 4 change greatly. b is calculated, and the inverter 11 is controlled using this operation amount θ b , so that the reduction effect of the current pulsation can be easily stabilized regardless of the operating conditions of the drive device 4 and the installation conditions of the drive device 4. Therefore, the control unit 400 can effectively suppress the beat vibration of the current even if the operating conditions and installation conditions of the drive device 4 change greatly. b is calculated, and the inverter 11 is controlled using this operation amount θ b , so that the reduction effect of the current pulsation can be easily stabilized regardless of the operating conditions of the drive device 4 and the installation conditions of the drive device 4. Therefore, the control unit 400 can effectively suppress the beat vibration of the current even if the operating conditions and installation conditions of the drive device 4 change greatly.

[0221] Embodiment 2. Next, Embodiment 2 will be described with reference to FIGS. 29 to 33. In Embodiment 1, it was described that the optimum value of the rotation amount θ changes due to the change in the power supply impedance. R The optimum value of the rotation amount θ R When the rotation amount θ R is out of the optimum value, the time for searching for the optimum value of the rotation amount θ R becomes long, so the control response of the beatless control may become slow. Therefore, in Embodiment 2, in order to shorten the search time of the rotation amount θ R when LC resonance is detected, the rotation amount θ

[0222] FIG. 29 is a diagram showing the configuration of a beatless control unit provided in the drive device according to Embodiment 2. Among the components in FIG. 29, the components that achieve the same functions as those of the beatless control unit 18 in Embodiment 1 shown in FIG. 5 are denoted by the same reference numerals, and redundant explanations are omitted.

[0223] The beatless control unit 18A according to Embodiment 2 has a rotation amount correction value calculation unit 106 in addition to the components included in the beatless control unit 18 according to Embodiment 1. Further, the beatless control unit 18A has a pulsation suppression unit 104A instead of the pulsation suppression unit 104 as compared with the beatless control unit 18.

[0224] In the beatless control unit 18A, the rotation amount correction value calculation unit 106 and the limit value determination unit 105 receive a resonance detection signal from the resonance detection unit 20. When receiving the resonance detection signal, the rotation amount correction value calculation unit 106 calculates a rotation amount correction value. The rotation amount correction value is a value for correcting the rotation amount θ R is a value for correcting

[0225] The rotation amount correction value calculation unit 106 stores correction value information indicating the correspondence between the resonance detection signal and the rotation amount correction value in advance. The rotation amount correction value calculation unit 106 calculates the rotation amount correction value based on the correction value information and the resonance detection signal received from the resonance detection unit 20.

[0226] The rotation amount correction value is zero by default, but when the rotation amount correction value calculation unit 106 receives a resonance detection signal from the resonance detection unit 20, it becomes a non-zero value (a value other than zero). That is, the rotation amount correction value becomes a non-zero value during LC resonance detection.

[0227] The rotation amount correction value calculation unit 106 may change the magnitude of the rotation amount correction value according to at least one of the risk of occurrence and the degree of occurrence of LC resonance. For example, when the resonance detection signal includes the pulsation amplitude of the DC bus voltage V DC |V dcDis |, the rotation amount correction value calculation unit 106 may increase the rotation amount correction value as |V dcDis | becomes larger. Also, the rotation amount correction value calculation unit 106 may calculate the frequency of receiving the resonance detection signal, and increase the rotation amount correction value as the frequency increases.

[0228] When the rotation amount correction value calculation unit 106 changes the rotation amount correction value, it outputs the changed rotation amount correction value to the pulsation suppression unit 104A. Note that when the rotation amount correction value calculation unit 106 does not change the rotation amount correction value, it may output the unchanged rotation amount correction value (zero, which is the default value) to the pulsation suppression unit 104A.

[0229] When the pulsation suppression unit 104A receives the rotation amount correction value, it corrects the rotation amount θ R by the rotation amount correction value. Hereinafter, the case where the correction value information is a graph showing the correspondence between the resonance detection signal and the rotation amount correction value will be described, but the correction value information may be a table showing the correspondence between the resonance detection signal and the rotation amount correction value.

[0230] FIG. 30 is a diagram for explaining a first example of correction value information used by the beatless control unit according to Embodiment 2. The horizontal axis of the graph shown in FIG. 30 is the pulsation amplitude (|V dcDis |), and the vertical axis is the rotation amount correction value.

[0231] The rotation amount correction value calculation unit 106 of the beatless control unit 18A stores in advance correction value information corresponding to the graph shown in FIG. 30. For example, when the resonance detection signal includes the pulsating amplitude of the DC bus voltage V DC |V dcDis |, the rotation amount correction value calculation unit 106 determines whether LC resonance is excited based on the pulsating amplitude and the correction value information.

[0232] When |V dcDis | becomes larger than a specific threshold value, the rotation amount correction value calculation unit 106 can determine that the beatless control is exciting LC resonance. For example, when the correction value information is the information shown in FIG. 30, the rotation amount correction value calculation unit 106 sets the rotation amount correction value to a non-zero value when |V dcDis | becomes larger than the threshold value p1. In this case, the rotation amount correction value calculation unit 106 calculates, based on the correction value information, a value corresponding to |V dcDis | as the rotation amount correction value. Note that if the search time of the rotation amount θ R can be shortened, the rotation amount correction value can be freely determined.

[0233] Also, when a current sensor is attached to any power line connected to the diode rectifier 7 and the resonance detection unit 20 determines the presence or absence of LC resonance from the current detected by this current sensor, in principle, the LC resonance angular frequency itself can be calculated. In this case, the resonance detection unit 20 outputs the LC resonance angular frequency to the rotation amount correction value calculation unit 106 included in the resonance detection signal.

[0234] The rotation amount correction value calculation unit 106 stores correction value information (second example) in which the rotation amount correction value is a function of the estimated value of the LC resonance angular frequency. FIG. 31 is a diagram for explaining a second example of the correction value information used by the beatless control unit according to the second embodiment.

[0235] The horizontal axis of the graph shown in FIG. 31 is the estimated value of the LC resonance angular frequency, ω^ LC , and the vertical axis is the rotation amount correction value. FIG. 31 shows the correction value information when the rotation amount correction value is a function of the estimated value of the LC resonance angular frequency.

[0236] The rotation amount correction value calculation unit 106 of the beatless control unit 18A stores in advance correction value information corresponding to the graph shown in FIG. 31. For example, when the resonance detection signal includes ω^ LC (the estimated value of the LC resonance angular frequency), the rotation amount correction value calculation unit 106 determines whether the beatless control is exciting the LC resonance based on the estimated value of the LC resonance angular frequency and the correction value information.

[0237] ω^ LC (the estimated value of the LC resonance angular frequency) approaches ω dis (f dis (the conversion result of converting to the angular frequency)), the DC bus voltage V DC becomes unstable. Therefore, as ω^ LC approaches ω dis , that is, as ω^ LC becomes smaller, the rotation amount correction value calculation unit 106 increases the rotation amount correction value.

[0238] FIG. 32 is a diagram showing the configuration of the pulsation suppression unit included in the beatless control unit according to Embodiment 2. The pulsation suppression unit 104A of Embodiment 2 has a rotation amount adjustment unit 206 instead of the rotation amount adjustment unit 205 as compared with the pulsation suppression unit 104 of Embodiment 1.

[0239] The rotation amount adjustment unit 206 determines the rotation amount θ R using the rotation amount correction value sent from the rotation amount correction value calculation unit 106. In this way, the rotation amount adjustment unit 206 uses the rotation amount correction value for the search of the rotation amount θ R . The rotation amount adjustment unit 206 outputs the rotation amount θ R to the rotation operation unit 202.

[0240] FIG. 33 is a diagram showing the configuration of the rotation amount adjustment unit included in the pulsation suppression unit according to Embodiment 2. The rotation amount adjustment unit 206 according to Embodiment 2 includes the same components as the rotation amount adjustment unit 205 according to Embodiment 1. The rotation amount adjustment unit 206 according to Embodiment 2 differs from the rotation amount adjustment unit 205 according to Embodiment 1 in the process executed by the adder 309. Specifically, in the rotation amount adjustment unit 206 according to Embodiment 2, the adder 309 adds the rotation amount θ R sent from the PID control unit 306 to the default value stored in advance and the rotation amount correction value sent from the rotation amount correction value calculation unit 106 to correct the rotation amount θ R . The adder 309 outputs the rotation amount θ R which is the addition result to the rotation operation unit 202.

[0241] In this way, since the drive device 4 changes the rotation amount correction value according to at least one of the risk of occurrence and the degree of occurrence of LC resonance, the time for searching for the optimum value of the rotation amount θ R can be shortened. As a result, the drive device 4 can suppress the beat vibration in a short time.

[0242] As described above, according to Embodiment 2, since the control unit 400 of the drive device 4 changes the rotation amount correction value according to at least one of the risk of occurrence and the degree of occurrence of LC resonance, regardless of the power impedance, the time for searching for the optimum value of the rotation amount θ R can be shortened. As a result, the drive device 4 can effectively suppress the beat vibration caused by the pulsation of the DC bus voltage V DC in a short time.

[0243] Embodiment 3. Next, Embodiment 3 will be described with reference to FIG. 34. In Embodiment 3, the drive device 4 described in Embodiments 1 and 2 is applied to a refrigeration cycle device.

[0244] FIG. 34 is a diagram showing the configuration of the refrigeration cycle apparatus according to Embodiment 3. Components having the same functions as the drive device 4 and the compressor 3 of Embodiment 1 shown in FIG. 1 among the components of FIG. 34 are denoted by the same reference numerals, and redundant descriptions are omitted.

[0245] The refrigeration cycle apparatus 900 according to Embodiment 3 includes a compressor drive system 800. The compressor drive system 800 includes a drive device 4 having a control unit 400 and a compressor 3 incorporating the AC motor 1 in Embodiment 1. Further, the refrigeration cycle apparatus 900 includes a four-way valve 902, an indoor heat exchanger 906, an expansion valve 908, an outdoor heat exchanger 910, and a refrigerant pipe 912.

[0246] The refrigeration cycle apparatus 900, which is a refrigeration cycle applied device, can be applied to products equipped with a refrigeration cycle such as an air conditioner, a refrigerator, a freezer, and a heat pump water heater. In the compressor drive system 800, the compressor 3, the drive device 4, the four-way valve 902, the indoor heat exchanger 906, the expansion valve 908, and the outdoor heat exchanger 910 are connected via the refrigerant pipe 912.

[0247] Inside the compressor 3, a compression mechanism 904 for compressing the refrigerant and an AC motor 1 for operating the compression mechanism 904 are provided. The compression mechanism 904 corresponds to the mechanical device 2 described in Embodiment 1. The refrigeration cycle apparatus 900 can perform a heating operation or a cooling operation by switching the operation of the four-way valve 902. The compression mechanism 904 is driven by an AC motor 1 that is variably speed-controlled.

[0248] During the heating operation, as shown by the solid-line arrow, the refrigerant is pressurized and sent out by the compression mechanism 904, passes through the four-way valve 902, the indoor heat exchanger 906, the expansion valve 908, the outdoor heat exchanger 910, and the four-way valve 902, and returns to the compression mechanism 904.

[0249] During the cooling operation, as indicated by the dashed arrow, the refrigerant is pressurized and sent out by the compression mechanism 904, and returns to the compression mechanism 904 through the four-way valve 902, the outdoor heat exchanger 910, the expansion valve 908, the indoor heat exchanger 906, and the four-way valve 902.

[0250] During the heating operation, the indoor heat exchanger 906 acts as a condenser to release heat, and the outdoor heat exchanger 910 acts as an evaporator to absorb heat. During the cooling operation, the outdoor heat exchanger 910 acts as a condenser to release heat, the indoor heat exchanger 906 acts as an evaporator, and absorbs heat. The expansion valve 908 reduces the pressure of the refrigerant and expands it.

[0251] Thus, according to the third embodiment, even in the voltage saturation region of the inverter voltage, the pulsation can be reduced as intended. Therefore, the refrigeration cycle device 900 can prevent the deterioration of the motor efficiency due to the increase in the current peak value, the decrease in the maximum output of the motor due to the restriction of overcurrent protection, the vibration and noise increase of the AC motor 1, etc. by suppressing the beat vibration.

[0252] In addition, since the control unit 400 can effectively suppress the beat vibration, the DC reactor 8 and the capacitor 9 can be miniaturized and downsized. Thereby, the manufacturing cost of the refrigeration cycle device 900 can be reduced, and the energy-saving performance can be improved.

[0253] The configurations shown in the above embodiments are examples, and it is possible to combine them with other known technologies, or to combine the embodiments with each other, or to omit or change a part of the configuration without departing from the gist.

Explanation of Reference Numerals

[0254] 1 AC motor, 2 mechanical device, 3 compressor, 4 drive device, 5 AC power supply, 6 power supply inductance, 7 diode rectifier, 8 DC reactor, 9 capacitor, 10 DC bus voltage detection unit, 11 inverter, 12 current detection unit, 13 modulation unit, 14 rotor position calculation unit, 15 voltage command determination unit, 16, 17 coordinate conversion unit, 18, 18A beatless control unit, 19, 309 adder, 20 resonance detection unit, 91 processor, 92 memory, 93 peripheral device, 101 norm calculation unit, 102 weight coefficient setting unit, 103 pulsation extraction unit, 104, 104A pulsation suppression unit, 105 limit value determination unit, 106 rotation amount correction value calculation unit, 201A, 201B, 304, 701 subtractor, 202 rotation calculation unit, 203A, 203B integral control unit, 204 AC restoration unit, 205, 206 rotation amount adjustment unit, 300 outer product calculation unit, 302A, 302B pseudo differentiator, 303A, 303B multiplier, 305 dead zone, 306 PID control unit, 400 control unit, 401 pulsation amplitude calculation unit, 402 comparator, 702 PID controller, 800 compressor drive system, 900 refrigeration cycle device, 902 four-way valve, 904 compression mechanism, 906 indoor heat exchanger, 908 expansion valve, 910 outdoor heat exchanger, 912 refrigerant pipe, E, ER deviation vector, Et, ERt deviation vector locus, Ita, Itb current locus, Pc phase change amount, VD disturbance voltage, Vt voltage locus.

Claims

1. A drive device that drives an AC motor using an inverter, A current detection unit for detecting the current flowing through the AC motor, A rotor position calculation unit calculates rotor position information, which is information about the position of the rotor of the AC motor, A voltage command determination unit determines a voltage command based on the current vector of the current and the rotor position information, A beatless control unit that suppresses the current pulsation caused by periodic pulsation of the DC bus voltage by manipulating the voltage phase of the voltage command, A resonance detection unit for detecting inductor-capacitor resonance between the AC power supply that provides power to the inverter and the inverter, Equipped with, The beatless control unit reduces the amount of voltage phase manipulation when the inductor-capacitor resonance is excited. A norm calculation unit that calculates the norm of the current vector, A pulsation extraction unit for extracting the pulsation component of the norm, A pulsation suppression unit that determines the amount of manipulation at which the pulsation component is reduced, A limit value determination unit calculates a limit value to limit the manipulated amount based on the inductor-capacitor resonance detected by the resonance detection unit, It has, The pulsation suppression unit determines the manipulated amount at which the pulsation component is reduced within the range of the limit value. A drive device characterized by the following features.

2. The pulsation suppression unit is, A rotation amount adjustment unit that determines the amount of rotation when performing rotation calculations on the cosine and sine components of the pulsating component, A rotation calculation unit that performs rotation calculations on the cosine component and the sine component using the rotation amount, A first integral control unit that integrates the rotated cosine component, A second integral control unit that integrates the rotated sine component, An AC restoration unit calculates the manipulated variable based on the integrated cosine component, the integrated sine component, and the disturbance frequency which is the pulsation frequency of the DC bus voltage. Equipped with, The drive device according to feature 1.

3. The rotation amount adjustment unit is, When an increase in the cosine component and the sine component is detected, the amount of rotation is corrected. The drive device according to feature 2.

4. The aforementioned beatless control unit is The system further includes a rotation amount correction value calculation unit that calculates a rotation amount correction value, which is a correction value for the rotation amount, based on the inductor-capacitor resonance detected by the resonance detection unit. The rotation amount adjustment unit corrects the rotation amount using the rotation amount correction value, The rotation calculation unit performs the rotation calculation using the corrected amount of rotation. The drive device according to feature 2.

5. A drive device according to any one of claims 1 to 4, A compressor driven by the aforementioned AC motor, Equipped with, A compressor drive system characterized by the following:

6. The compressor drive system is provided according to claim 5. A refrigeration cycle device characterized by the following features.

Images