Power conversion device

Abstract

In this power conversion device for converting a DC voltage to an AC voltage by turning on / off a semiconductor switching element (Q), a drive controller (10) that generates a gate signal for on / off driving the semiconductor switching element (Q) is configured so as to: generate a first command that is a voltage command for a target voltage of the AC voltage; generate a triangular wave carrier signal, the frequency of which is an even multiple of the fundamental wave component of the first command; and generate the gate signal from a second command and the carrier signal, said second command being constituted by the first command and a harmonic signal including at least one of odd multiple harmonic signals, the frequencies of which are odd multiples of the fundamental wave component of the first command, and one of even multiple harmonic signals, the frequencies of which are even multiples of the fundamental wave component of the first command.

Description

Power Conversion Device

[0001] The present disclosure relates to a power conversion device.

[0002] Inverters are widely used as power conversion devices that convert DC power to AC power. Inverters apply a desired AC voltage to a load, such as a rotating machine, by comparing a sinusoidal modulation signal with a triangular carrier signal and using pulse width modulation (PWM) through the switching operation of semiconductor switching elements. (Hereinafter, PWM based on the comparison with a triangular carrier signal will be referred to as triangular-wave comparison PWM.) However, PWM generates harmonics in the voltage applied to the load, which also generates harmonics in the current flowing through the load. These harmonics increase noise and losses in the load. In the case of a rotating machine, they can cause not only increased noise and losses but also torque pulsation and vibration. Therefore, suppression of the harmonics generated by PWM is desirable.

[0003] As a method for suppressing harmonics, there is a method called low-order harmonic elimination PWM or optimal pulse pattern, which designs the switching phase so as to suppress harmonic components of a predetermined order relative to the fundamental frequency of the load (see, for example, Patent Document 1). However, these methods cannot be implemented in a general-purpose controller IC because they require issuing a command to perform switching operation at a predetermined phase.

[0004] Therefore, an implementation method has been disclosed that can achieve the same effect as low-order harmonic removal PWM or optimal pulse pattern by using a triangular wave comparison type PWM mounted on a general-purpose controller IC (for example, Patent Document 2).

[0005] JP 2013-215041 A JP 2020-171086 A

[0006] In the power conversion device of Patent Document 2, a PWM command is generated by comparing a triangular wave, which has one electrical cycle, with a DC modulation signal, which has symmetrical positive and negative polarities, so the number of PWM pulses (the number of on or off switching operations) per electrical cycle is always odd. Therefore, when the number of pulses is even, there is a problem in that the effect of suppressing harmonics cannot be obtained.

[0007] The present disclosure has been made to solve the above-mentioned problems, and aims to control load noise and loss with a simple configuration even when the number of pulses is an even number, and to control the noise, torque pulsation, vibration, and loss of the rotating machine when the load is a rotating machine.

[0008] The power conversion device disclosed herein is a power conversion device that converts DC voltage into AC voltage by turning semiconductor switching elements on and off, and a drive controller that generates a gate signal for driving the semiconductor switching elements on and off is configured to generate a first command that is a voltage command for a target voltage of the AC voltage, generate a triangular wave carrier signal whose frequency is an even multiple of the frequency of the fundamental wave component of the first command, and generate the gate signal from the carrier signal and a second command that is composed of a harmonic signal that includes at least one odd-multiple harmonic signal whose frequency is an odd-multiple and one even-multiple harmonic signal whose frequency is an even multiple of the frequency of the fundamental wave component of the first command, and the first command.

[0009] According to the present disclosure, even when the number of pulses is an even number, a simple configuration using triangular wave comparison PWM control mounted on a general-purpose controller IC can be used to control load noise and loss, and when the load is a rotating machine, the noise, torque pulsation, vibration, and loss of the rotating machine can be controlled.

[0010] 7A is a diagram showing an example of a second command value Mu of a conventional power conversion device, and FIG. 7B is a diagram showing an output voltage waveform of the power conversion device controlled by the second command value Mu of FIG. 7A. FIG. 7B is a diagram showing harmonics included in the output voltage waveform output by the conventional power conversion device shown in FIG. 7B. FIG. 7C is a diagram showing an example of an output voltage waveform output by the power conversion device according to embodiment 1. FIG. 9 is a diagram showing harmonics included in the output voltage waveform output by the power conversion device according to embodiment 1. FIG. 11A is a diagram showing an example of a second command value Mu of the power conversion device according to embodiment 1, and FIG. 11B is a diagram showing an output voltage waveform of the power conversion device according to embodiment 1 controlled by the second command value Mu of FIG. 11A. FIG. 12A is a diagram showing the magnitude of harmonic components included in the second command value Mu of FIG. 11A, and FIG. 12B is a diagram showing the phase of the harmonic components included in the second command value Mu of FIG. 11A. FIG. 12B is a block diagram showing the configuration of a power conversion device according to embodiment 2. FIG. 12A is a block diagram showing the configuration of a carrier signal generating means of the power conversion device according to embodiment 2. FIG. 12B is a diagram showing an example of a carrier signal generated by a carrier signal calculation unit of the power conversion device according to embodiment 2. FIG. 17A is a diagram showing an example of a second command value Mu of a conventional power conversion device, and FIG. 17B is a diagram showing the output voltage waveform of the power conversion device controlled by the second command value Mu of FIG. 17A. FIG. 17B is a diagram showing harmonics included in the output voltage waveform output by the conventional power conversion device shown in FIG. 19 is a diagram showing an example of an output voltage waveform output by the power conversion device according to embodiment 2. FIG. 20 is a diagram showing harmonics included in the output voltage waveform output by the power conversion device according to embodiment 1 shown in FIG.Fig. 21A is a diagram showing an example of a second command value Mu of a power conversion device according to embodiment 2, and Fig. 21B is a diagram showing an output voltage waveform of the power conversion device according to embodiment 2 controlled by the second command value Mu of Fig. 21A. Fig. 22A is a diagram showing the magnitude of harmonic components included in the second command value Mu of Fig. 21A, and Fig. 22B is a diagram showing the phase of the harmonic components included in the second command value Mu of Fig. 21A. Fig. 22B is a block diagram showing the configuration of a power conversion device according to embodiment 3. Fig. 22C is a block diagram showing the configuration of modulated voltage generating means of a power conversion device according to embodiment 3. Fig. 22D is a block diagram showing an example of a hardware configuration of a power conversion device according to the present disclosure. Fig. 22E is a block diagram showing another example of the hardware configuration of a power conversion device according to the present disclosure.

[0011] Embodiment 1. Figure 1 is a block diagram showing the configuration of a power conversion device according to embodiment 1. The power conversion device 1A includes an inverter circuit 3 connected to a DC power supply 2 and a rotating machine 4, which is a load, and a drive controller 10 that constitutes voltage command generation means 5, carrier signal generation means 6A, modulated voltage generation means 7A, and gate signal generation means 8. The drive controller 10 controls the on / off of a semiconductor switching element Q, which will be described next. The DC power supply 2 applies a DC voltage vdc to the inverter circuit 3. The DC power supply 2 may be any device that outputs a DC voltage Vdc, such as a battery, a DC-DC converter, a diode rectifier, or a PWM rectifier.

[0012] The inverter circuit 3 converts DC power from the DC power source 2 into AC power and supplies the power to the rotating machine 4. Note that, in this example, the inverter circuit 3 has three phases, designated as u-phase, v-phase, and w-phase. However, the number of phases of the inverter circuit 3 does not have to be three, as long as it is the same as the number of phases of the rotating machine 4. The inverter circuit 3 is configured such that legs, each of which has two reversely conductive semiconductor switching elements Q connected in series between the positive and negative terminals of the DC power source 2, are connected in parallel in the same number as the number of phases. In this example, since the inverter circuit 3 has three phases, the inverter circuit 3 has six semiconductor switching elements Q, and as shown in FIG. 1 , the six semiconductor switching elements Q are referred to as semiconductor switching elements Qup, Qun, Qvp, Qvn, Qwp, and Qwn, respectively. Furthermore, a series connection of a positive-side semiconductor switching element Qup and a negative-side semiconductor switching element Qun corresponding to the u-phase is referred to as leg 31u, a series connection of a positive-side semiconductor switching element Qvp and a negative-side semiconductor switching element Qvn corresponding to the v-phase is referred to as leg 31v, and a series connection of a positive-side semiconductor switching element Qwp and a negative-side semiconductor switching element Qwn corresponding to the w-phase is referred to as leg 31w. The intermediate terminals of each leg are connected to the respective phases of the rotating machine 4.

[0013] Here, each semiconductor switching element Q is configured with an insulated gate bipolar transistor (IGBT) and an anti-parallel diode. Furthermore, if a metal-oxide-semiconductor field-effect transistor (MOSFET), a reverse conducting (RC) IGBT, or the like is used instead of the IGBT, the anti-parallel diode may be omitted. Furthermore, each semiconductor switching element may be configured with a wide bandgap semiconductor such as SiC or GaN. In the case of a wide bandgap semiconductor, in addition to being able to reduce losses in the semiconductor switching element, high-speed switching operation can shorten dead time, thereby achieving the effect of reducing errors between the switching operation command and the actual switching operation of the present disclosure.

[0014] The rotating machine 4 has a stator consisting of three-phase stator windings, u-phase, v-phase, and w-phase, and a rotor that rotates due to a rotating magnetic field generated by an AC current flowing through the stator winding. The number of phases of the stator winding does not have to be three, and the stator winding may be configured with two or more sets. The load does not have to be a rotating machine, and may be a load configured with a resistor, a coil, or the like.

[0015] For example, when the load is a rotating machine, the voltage command generating means 5 calculates first commands vu*, vv*, vw*, which are phase voltage commands that are sinusoidal, i.e., mainly composed of fundamental sinusoidal components, as the voltage to be supplied to the rotating machine 4 as shown in equation (1) based on one or more operation commands of the torque, rotational speed, position, current amplitude, phase, and frequency of the rotating machine 4, and outputs the calculated vu*, vv*, vw* to the carrier signal generating means 6A and the modulated voltage generating means 7A, respectively. In equation (1), vphp represents the amplitude of the first commands vu*, vv*, vw*, and θv represents the fundamental wave phase of the u phase.

[0016] The voltage command generating means 5 may be configured to feed back to the voltage command generating means 5 one or more detected or estimated values ​​of the torque, rotational speed, position, current amplitude, phase, and frequency of the rotating machine 4, and calculate the first commands vu*, vv*, and vw* by feedback control based on the operation command and the detected or estimated values. For example, the voltage command generating means 5 may be configured to calculate the first commands vu*, vv*, and vw* by known current feedback control, speed feedback control, or position feedback control. The first commands vu*, vv*, and vw* are not limited to sine waves and may have waveforms containing harmonic components in a sine wave. Furthermore, if the load is not a rotating machine, there is no torque parameter. Therefore, the voltage command generating means 5 may calculate the first commands vu*, vv*, and vw*, which are sinusoidal phase voltage commands as voltages to be supplied to the load, from one or more operation commands of the voltage, current amplitude, phase, and frequency.

[0017] The carrier signal generating means 6A generates a triangular wave carrier signal c based on the first commands vu*, vv*, and vw*, and outputs the generated carrier signal c to the gate signal generating means 8.

[0018] The modulation voltage generating means 7A generates second commands mu, mv, and mw, which are phase voltage commands for modulation, based on the first commands vu*, vv*, and vw*, and outputs the generated second commands mu, mv, and mw to the gate signal generating means 8. Note that, since the amplitudes of the second commands mu, mv, and mw need to be changed according to the magnitude of the DC voltage vdc, the DC voltage vdc may be detected by a voltage sensor, and the first commands vu*, vv*, and vw* may be normalized by dividing them by half of the detected DC voltage vdc.

[0019] The gate signal generating means 8 compares the magnitude of the second commands mu, mv, mw with the magnitude of the carrier signal c to generate gate signals gup, gun, gvp, gvn, gwp, gwn (sometimes collectively referred to as gate signals g) that control the on / off of the semiconductor switching elements Q of the inverter circuit 3. The gate signals gup, gun, gvp, gvn, gwp, gwn correspond to the semiconductor switching elements Qup, Qun, Qvp, Qvn, Qwp, Qwn, respectively, and turn the corresponding semiconductor switching elements Q on or off. The gate signals gup, gun, gvp, gvn, gwp, gwn each take on a high "H" or low "L" value. When the value of the gate signal g is "H", the semiconductor switching element Q corresponding to the gate signal g is controlled to be ON, and when the value of the gate signal g is "L", the semiconductor switching element Q corresponding to that gate signal g is controlled to be OFF.

[0020] FIG. 2 is a diagram illustrating the operation of the gate signal generating means 8 shown in FIG. 1 . FIG. 2 shows one phase, the u phase. The gate signal generating means 8 compares the second command mu with the carrier signal c to generate the gate signals gup and gun. Specifically, when the second command mu is greater than the carrier signal c, the value of the gate signal gup is set to "H" and the value of the gate signal gun is set to "L." When the second command mu is smaller than the carrier signal c, the value of the gate signal gup is set to "L" and the value of the gate signal gun is set to "H." The gate signals gup and gun are complementary to each other. In other words, when the gate signal gup is "H," the gate signal gun is set to "L," and when the gate signal gup is "L," the gate signal gun is set to "H." When the gate signal gup is "H", the positive side semiconductor switching element Qup turns on and the output phase voltage vu becomes vdc, and when the gate signal gun is "H", the negative side semiconductor switching element Qun turns on and the output phase voltage vu becomes 0.

[0021] 3 is a block diagram showing the configuration of the carrier signal generating means 6A shown in FIG. The carrier signal generating means 6A includes a three-phase to two-phase conversion unit 601, a phase calculation unit 602, and a carrier signal calculation unit 603A. The three-phase to two-phase conversion unit 601 performs three-phase to two-phase conversion from first commands vu*, vv*, and vw* on a three-phase coordinate system to first commands vα* and vβ* on a two-phase coordinate system. Specifically, the three-phase to two-phase conversion unit 601 can perform the three-phase to two-phase conversion using equation (2).

[0022] The phase calculation unit 602 calculates the fundamental wave phase θv of the u-phase of the first commands vα* and vβ* on the two-phase coordinate system. Specifically, as shown in equation (3), the arctangent calculation is performed on the first commands vα* and vβ* on the two-phase coordinate system to obtain the phase of vα*. This phase is the same as the phase of vu*.

[0023] Note that the description here is based on the assumption that the first commands vu*, vv*, vw* and the first commands vα*, vβ* have waveforms with sufficiently few harmonic components and are the same as the fundamental wave components. If the first commands vu*, vv*, vw* and the first commands vα*, vβ* contain many harmonic components, the fundamental wave components can be extracted by passing them through a low-pass filter, for example.

[0024] The carrier signal calculation unit 603A calculates a carrier signal c whose frequency is an even number Kc times the frequency of the fundamental wave components of the first commands vu*, vv*, and vw*. Specifically, as shown in equation (4), the phase θc of the carrier signal is generated by multiplying the fundamental wave phase θv by the even number Kc.

[0025] The carrier signal calculation unit 603A then generates a triangular wave carrier signal c as shown in FIG. 4. FIG. 4 is a diagram showing the carrier signal generated by the carrier signal calculation unit 603A shown in FIG. 3. The carrier signal c is a triangular wave whose cycle is 1 / Kc of the fundamental wave components of the first commands vu*, vv*, and vw*, and has a maximum value when the phase θc of the carrier signal is 0° and 360°, a minimum value when it is 180°, and a median value when it is 90° and 270°. In the example shown in FIG. 3, the maximum value of the triangular wave is 1, the minimum value is -1, and the median value is zero. The carrier signal calculation unit 603A outputs the carrier signal c to the gate signal generation means 8.

[0026] In the first embodiment, as an example, the even number Kc is set to 6, and the phase of the carrier signal c is synchronized to 90°, i.e., the median of the triangular wave, with the phase of the fundamental wave component of the first command vu* at 0°. In this case, the carrier signal is as shown in FIG. 5 . FIG. 5 is a diagram illustrating an example of the carrier signal c generated by the carrier signal calculation unit 603A shown in FIG. 3 . Furthermore, in the first embodiment, the carrier signal c is synchronized with the fundamental wave phase θv of the first command vu*, and a common carrier signal c is used for the three phases. In this case, since the phase difference between the phases is 120°, in order to generate similar output phase voltage waveforms for each phase, the even number Kc, which determines the frequency of the carrier signal c, can be set to a multiple of 6. Note that when separate carrier signals c are used for the three phases, the even number Kc, which determines the frequency of the carrier signal c, does not need to be set to a multiple of 6; Kc may be set to any even number for each phase.

[0027] Fig. 6 is a diagram showing the configuration of the modulated voltage generating means 7A shown in Fig. 1. The modulated voltage generating means 7A includes a three-phase to two-phase conversion unit 701, an amplitude calculation unit 702, a phase calculation unit 703, a fundamental wave signal calculation unit 704, an odd-numbered harmonic signal calculation unit 705, and an even-numbered harmonic signal calculation unit 706A.

[0028] The three-phase to two-phase conversion unit 701 performs three-phase to two-phase conversion from first commands vu*, vv*, vw*, which are sinusoidal phase voltage commands on the three-phase coordinate system, to first commands vα*, vβ* on the two-phase coordinate system, using processing similar to that of the three-phase to two-phase conversion unit 601, and outputs the first commands vα*, vβ* on the two-phase coordinate system to the amplitude calculation unit 702 and the phase calculation unit 703.

[0029] The amplitude calculation unit 702 first calculates the amplitudes vphp of the first commands vu*, vv*, and vw* from the first commands vα* and vβ* on the two-phase coordinates. For example, the amplitude calculation unit 702 can calculate the amplitudes vphp of the first commands vu*, vv*, and vw* using the following equation (5).

[0030] Next, the amplitude calculation unit 702 divides the calculated amplitude vphp of the first commands vu*, vv*, and vw* by half the DC voltage vdc, as shown in the following equation (6), to convert it into a normalized amplitude M of the first commands vu*, vv*, and vw*, and outputs the calculated amplitude M to the fundamental wave signal calculation unit 704, the odd-multiple harmonic signal calculation unit 705, and the even-multiple harmonic signal calculation unit 706A.

[0031] The phase calculation unit 703 calculates the fundamental wave phase θv of the u-phase of the first commands vα* and vβ* on the two-phase coordinate system using processing similar to that of the phase calculation unit 602, and outputs the calculated fundamental wave phase θv to the fundamental wave signal calculation unit 704, the odd-multiple harmonic signal calculation unit 705, and the even-multiple harmonic signal calculation unit 706A.

[0032] The fundamental wave signal calculation unit 704 calculates fundamental wave signals mu_1, mv_1, mw_1, which are the fundamental wave components of the first commands vu*, vv*, vw*, based on the amplitude M of the normalized first commands vu*, vv*, vw* and the fundamental wave phase θv of the u phase.

[0033] The odd-multiple harmonic signal calculation unit 705 calculates odd-multiple harmonic signals mu_odd, mv_odd, mw_odd whose frequencies are odd multiples of the fundamental wave components of the first commands vu*, vv*, vw* based on the amplitude M of the normalized first commands vu*, vv*, vw* and the fundamental wave phase θv of the u phase.

[0034] The even-multiple harmonic signal calculation unit 706A calculates even-multiple harmonic signals mu_even, mv_even, mw_even whose frequencies are even multiples of the fundamental wave components of the first commands vu*, vv*, vw* based on the amplitude M of the normalized first commands vu*, vv*, vw* and the fundamental wave phase θv of the u phase.

[0035] Then, as shown in equation (7), the modulated voltage generating means 7A adds together the fundamental signals mu_1, mv_1, mw_1, odd-multiple harmonic signals mu_odd, mv_odd, mw_odd, and even-multiple harmonic signals mu_even, mv_even, mw_even calculated by the fundamental signal calculation unit 704, odd-multiple harmonic signal calculation unit 705, and even-multiple harmonic signal calculation unit 706A, respectively, to generate second commands mu, mv, and mw, and outputs the generated second commands mu, mv, and mw to the gate signal generating means 8.

[0036] Hereinafter, we will explain in order the fundamental signals mu_1, mv_1, mw_1, odd harmonic signals mu_odd, mv_odd, mw_odd, and even harmonic signals mu_even, mv_even, mw_even calculated by the fundamental signal calculation unit 704, odd harmonic signal calculation unit 705, and even harmonic signal calculation unit 706A of the modulated voltage generating means 7A.

[0037] 7A and 7B are diagrams showing an example of a phase voltage (FIG. 7B) output by a conventional power conversion device as a comparative example when the sinusoidal first command vu* is directly provided to the gate signal generating means 8 as the second command mu (FIG. 7A). In FIGS. 7A and 7B, the amplitude M of the normalized first commands vu*, vv*, and vw* is 0.6. The waveforms of the phase voltages output by the inverter circuit 3 are shown for the u-phase component when the gate signal g is generated from the second commands mu, mv, and mw using the carrier signal c and the gate signal generating means 8. FIG. 7A shows the sinusoidal second command mu and the triangular wave carrier signal c, while FIG. 7B shows the output phase voltage vu output based on the sinusoidal second command mu and the triangular wave carrier signal c. Here, α1 to α11 in FIG. 7B represent switching phases. When using the carrier signal c of this embodiment, switching occurs 11 times within one cycle of the second command mu. Of these, when the fundamental wave phase θv is 180°, the second command mu and carrier signal c always pass through 0, so the switching phase α6 is always 180°. Furthermore, since the output phase voltage vu is symmetrical in both positive and negative directions between phases 0° and 180° and between phases 180° and 360°, once the switching phases α1 to α5 are determined, the switching phases α7 to α11 are also determined. Therefore, to obtain an output phase voltage vu that reduces the amplitude of the specified frequency components of the voltage, current, torque, loss, and vibration of the rotating machine 4, as well as the voltage and current of the DC power supply, as described below, it is sufficient to appropriately adjust the switching phases α1 to α5. Note that, when the load is other than the rotating machine 4, there is no torque parameter, so it is sufficient to appropriately adjust the switching phases α1 to α5 to reduce the amplitude of the specified frequency components of the load voltage, current, loss, and voltage and current of the DC power supply.

[0038] The technology disclosed herein allows for flexible control of harmonic components generated by switching. Therefore, if for some reason you want to increase load noise or loss due to harmonic components, or if the load is a rotating machine, you want to increase the noise, torque pulsation, vibration, and loss of the rotating machine due to harmonic components. Similarly to reducing these, the technology disclosed herein can be applied. If you want to increase noise, loss, or the like, you simply need to appropriately adjust the switching phases α1 to α5 to obtain an output phase voltage vu that increases the amplitude of a specified frequency component. That is, compared to the case where a gate signal is generated by comparing the carrier signal with a sine wave modulated signal of the fundamental wave component of the first command to operate the power conversion device, if the load is a rotating machine, the drive controller can be configured to generate the second command so that the amplitude of a predetermined frequency component included in at least one parameter of the rotating machine's voltage, current, torque, loss, vibration, or DC voltage or DC current flowing on the DC side is reduced or increased; and if the load is other than a rotating machine, the amplitude of a predetermined frequency component included in at least one parameter of the load's voltage, current, loss, DC voltage, or DC current flowing on the DC side is reduced or increased.

[0039] FIG. 8 is a diagram showing the magnitude of harmonic components contained in the output phase voltage vu shown in FIG. 7B. The horizontal axis of FIG. 8 represents the order relative to the fundamental wave, and the vertical axis represents the magnitude of the harmonic components. From FIG. 8, it can be seen that the output phase voltage vu shown in FIG. 7B contains odd- and even-multiple harmonic components. Furthermore, excluding the integer multiples of three that do not appear in the harmonic components of the phase current flowing through the rotating machine 4, it can be seen that the 11th- and 13th-order harmonics are particularly large. Therefore, in this embodiment, as an example, it is assumed that the 11th- and 13th-order harmonic components relative to the fundamental wave component of the output phase voltage vu excite mechanical resonance in the rotating machine 4, causing loud noise or torque pulsation and increasing losses in the rotating machine 4. The switching phases α1 to α5 are adjusted to reduce the 11th- and 13th-order harmonic components of the output phase voltage vu.

[0040] To find the switching phases α1 to α5 that reduce the 11th and 13th harmonic components of the output phase voltage vu, it is sufficient to solve an optimization problem so as to minimize the objective function fobj(α) expressed by, for example, equation (8). Here, the objective function fobj(α) is set to minimize the square root of the sum of squares of the nth-order harmonic voltage vn in the range from second to pth order divided by the fundamental voltage v1, i.e., the total harmonic distortion. Furthermore, kn is a weighting coefficient, and by increasing the weighting coefficient kn for a specific order, it is possible to preferentially reduce harmonics of a specific order. Note that the switching phases α1 to α5 are subject to constraints as shown in Equation (9), and the optimization problem is solved while observing these constraints.

[0041] In this embodiment, p in equation (8) is set to 50, and the objective function fobj(α) is set so as to minimize the total harmonic distortion for the second to fiftieth harmonic voltages. In addition, by increasing the weighting coefficients of the 11th and 13th orders as shown in equation (10), the 11th and 13th harmonic components are preferentially reduced.

[0042] FIG. 9 shows an example of the output phase voltage vu obtained by adjusting the switching phases α1 to α5 to minimize the 11th- and 13th-order harmonic components. FIG. 10 shows the magnitude of the harmonic components contained in the output phase voltage vu shown in FIG. 9 . The horizontal axis of FIG. 10 represents the order relative to the fundamental wave, and the vertical axis represents the magnitude of the harmonic components. As shown in FIG. 9 , by solving the optimization problem to minimize the objective function fobj(α) expressed by equation (8), it can be seen that the switching phases α1 to α5 of the output phase voltage vu shown in FIG. 7B are changed. Furthermore, as shown in FIG. 10 , it can be seen that the 11th- and 13th-order harmonic components are reduced compared to the magnitude of the harmonic components shown in FIG. 8 . Therefore, the second command mu can be generated using a triangular-wave comparison PWM to output the output phase voltage vu shown in FIG. 9 .

[0043] 11A and 11B are diagrams showing an example of the phase voltage ( FIG. 11B ) output by the power conversion device 1A when the gate signal generating means 8 is provided with the second command mu ( FIG. 11A ) generated so as to output the output phase voltage vu shown in FIG. 9 . FIGS. 11A and 11B show the waveform of the phase voltage output for the u-phase component when the inverter circuit 3 is operated by generating a gate signal g from the second commands mu, mv, and mw using the carrier signal c and the gate signal generating means 8, when the amplitude M of the normalized first commands vu*, vv*, and vw* is 0.6. FIG. 11A shows the second command mu and the triangular wave carrier signal c, and FIG. 11B shows the output phase voltage vu output based on the second command mu and the triangular wave carrier signal c. By performing triangular wave comparison PWM using the second command mu shown in FIG. 11A , it is possible to output a waveform similar to the output phase voltage vu shown in FIG. 9 .

[0044] In the above, an example of the waveform of the second command mu has been described in which the value of the second command mu remains constant throughout the period in which the carrier signal c changes from its minimum value to its maximum value and the period in which the carrier signal c changes from its maximum value to its minimum value, i.e., the waveform of the second command mu has a stepped shape. However, as long as the output phase voltage vu shown in Fig. 9 can be output, that is, as long as the phase at which the second command mu crosses the carrier signal c coincides with each switching phase α in Fig. 9, the second command mu does not need to maintain a constant value over a half cycle of the carrier signal c and may have a waveform that changes smoothly.

[0045] 12A and 12B are diagrams showing the magnitude and phase of the harmonic components of the second command mu shown in FIG. 11A. The horizontal axes of FIGS. 12A and 12B represent the order relative to the fundamental wave. The vertical axes of FIG. 12A represent the magnitude of the harmonic components, and the vertical axes of FIG. 12B represent the phase of the harmonic components relative to a sine wave. From FIG. 12A , it can be seen that the second command mu is composed of a fundamental wave component, odd-multiple harmonic components, and even-multiple harmonic components. From FIG. 12B , it can be seen that the fundamental wave component, odd-multiple harmonic components, and even-multiple harmonic components of the second command mu are all composed of sine wave components (sine waves in which the phase 0° of each harmonic component is synchronized with the phase 0° of the fundamental wave component).

[0046] Therefore, the fundamental signal calculation unit 704, odd-multiple harmonic signal calculation unit 705, and even-multiple harmonic signal calculation unit 706A of the modulation voltage generation means 7A only need to calculate the fundamental signal mu_1, odd-multiple harmonic signal mu_odd, and even-multiple harmonic signal mu_even, respectively, as shown in equation (11), where q is a value set so that 2q+1 is the maximum order considered, M1 is the amplitude of the fundamental signal mu_1, M2n+1 is the amplitude of the odd-multiple harmonic signal mu_odd at the 2n+1th order, and M2n is the amplitude of the even-multiple harmonic signal mu_even at the 2nth order.

[0047] As described above, in this embodiment, the harmonic signals of each order of the odd-multiple harmonic signal mu_odd and the even-multiple harmonic signal mu_even are preferably sine waves, and the phase zero of each harmonic signal may be synchronized with the phase zero of the fundamental wave component of the first command vu*. Furthermore, the harmonic signals of each order of the odd-multiple harmonic signal mu_odd and the even-multiple harmonic signal mu_even are not limited to sine waves and may have distorted waveforms, as long as they have the same value at the beginning and end of one cycle of the fundamental wave component of the first command and have periodicity in which they cross zero at least once per cycle.

[0048] The odd-multiple harmonic signal mu_odd and the even-multiple harmonic signal mu_even that constitute the second command preferably use a relatively large value for q in equation (11), i.e., are composed of a plurality of odd-multiple harmonic signals and a plurality of even-multiple harmonic signals. However, if the odd-multiple harmonic signal mu_odd and the even-multiple harmonic signal mu_even each contain a harmonic signal of at least one order, the switching phase can be controlled to control the harmonic components generated by switching, as compared to conventional synchronous PWM, thereby achieving the effect of the present disclosure of increasing or decreasing load noise, loss, etc.

[0049] In this embodiment, only the second command mu of the u phase has been described, but since the waveforms of the second commands mu, mv, and mw of the three phases are waveforms in which the phases of the fundamental wave components are shifted by 120° from each other, the second commands mv and mw of the remaining two phases can also be generated in the same way as the second command mu.

[0050] As described above, according to the power conversion device of the first embodiment, even when the triangular wave carrier signal has a frequency that is an even multiple of the fundamental wave frequency, i.e., when the number of pulses of the output voltage per cycle of the fundamental wave is even, it is possible to control harmonics of a predetermined order of the voltage or current generated by PWM with a relatively simple configuration by using a triangular wave comparison type PWM mounted on a general-purpose controller IC. This makes it possible to control noise and loss in the load, and when the load is a rotating machine, it is possible to control noise, torque pulsation, vibration, and loss of the rotating machine.

[0051] 13 is a diagram showing the configuration of a power conversion device 1B according to embodiment 2. The power conversion device 1B includes an inverter circuit 3 connected to a DC power supply 2 and a rotating machine 4, which is a load, a voltage command generating means 5, a carrier signal generating means 6B, a modulated voltage generating means 7B, and a gate signal generating means 8.

[0052] Power conversion device 1B has carrier signal generating means 6B having a different function from carrier signal generating means 6A of power conversion device 1A according to embodiment 1, and has modulated voltage generating means 7B having a different function from modulated voltage generating means 7A. Below, a description of the parts common to power conversion device 1A according to embodiment 1 will be omitted, and the following description will mainly focus on the parts that are different from power conversion device 1A.

[0053] Fig. 14 is a block diagram showing the configuration of the carrier signal generating means 6B shown in Fig. 13. The carrier signal generating means 6B includes a three-phase to two-phase conversion unit 601, a phase calculation unit 602, and a carrier signal calculation unit 603B. The carrier signal generating means 6B has a slightly different function from the carrier signal calculation unit 603A of the carrier signal generating means 6A according to the first embodiment.

[0054] The function of the carrier signal calculation unit 603B is basically the same as that of the carrier signal calculation unit 603A, except that the even number Kc, which determines the frequency of the carrier signal, is 6. In the carrier signal calculation unit 603A, the phase of the carrier signal c (90°, i.e., the median value of the triangular wave) is synchronized with the phase of the fundamental wave component of the first command vu* (0°). However, in the carrier signal calculation unit 603B according to the second embodiment, the phase of the carrier signal c (180°, i.e., the minimum value of the triangular wave) is synchronized with the phase of the fundamental wave component of the first command vu* (0°). In this case, the carrier signal becomes as shown in FIG. 15 . FIG. 15 is a diagram illustrating an example of the carrier signal c generated by the carrier signal calculation unit 603B shown in FIG. 14 . Also, in the second embodiment, as in the first embodiment, the carrier signal c is synchronized with the fundamental wave phase θv of the first command vu*, and a common carrier signal c is used for all three phases. In this case, since the phase difference between the phases is 120°, in order to generate similar output phase voltage waveforms for each phase, the even number Kc that determines the frequency of carrier signal c should be set to a multiple of 6. Note that when separate carrier signals c are used for the three phases, it is not necessary to set the even number Kc that determines the frequency of carrier signal c to a multiple of 6.

[0055] 16 is a diagram showing the configuration of the modulated voltage generating means 7B shown in FIG. 13. The modulated voltage generating means 7B includes a three-phase to two-phase conversion unit 701, an amplitude calculation unit 702, a phase calculation unit 703, a fundamental wave signal calculation unit 704, an odd-multiple harmonic signal calculation unit 705, and an even-multiple harmonic signal calculation unit 706B. The even-multiple harmonic signal calculation unit 706B has an even-multiple harmonic signal calculation unit 706B whose function is slightly different from that of the even-multiple harmonic signal calculation unit 706A according to the first embodiment. The function of the even-multiple harmonic signal calculation unit 706B is basically the same as that of the even-multiple harmonic signal calculation unit 706A described in the first embodiment, but the even-multiple harmonic signals mu_even, mv_even, and mw_even that it calculates are different.

[0056] 17A and 17B are diagrams showing an example of the phase voltage (FIG. 17B) output by the power conversion device 1B when the sinusoidal first command vu* is directly provided to the gate signal generating means 8 as the second command mu (FIG. 17A). FIG. 17B shows the waveform of the phase voltage output for the u-phase component when the amplitude M of the normalized first commands vu*, vv*, and vw* is 0.6 and the gate signal generating means 8 is used to generate the gate signal g from the second commands mu, mv, and mw to operate the inverter circuit 3. FIG. 17A shows the sinusoidal second command mu and the triangular wave carrier signal c, while FIG. 17B shows the output phase voltage vu output based on the sinusoidal second command mu and the triangular wave carrier signal c. Here, α1 to α12 in FIG. 17B represent switching phases. When the carrier signal c of the second embodiment is used, switching is performed 12 times within one period of the second command mu. The output phase voltage vu is inversion symmetric between phases 0° to 90° and between phases 90° to 180°, and between phases 180° to 270° and between phases 270° to 360°. Therefore, once the switching phases α1 to α3 and α7 to α9 are determined, the switching phases α4 to α6 and α10 to α12 are also determined. Therefore, in order to obtain an output phase voltage vu that reduces or increases the amplitude of predetermined frequency components of the voltage, current, torque, loss, and vibration of the rotating machine 4, and the voltage and current of the DC power supply, the switching phases α1 to α3 and α7 to α9 can be appropriately adjusted. In addition, if the load is other than the rotating machine 4, the switching phases α1 to α3 and α7 to α9 can be appropriately adjusted so that the amplitude of the specified frequency components of the voltage, current, and loss of the load, and the voltage and current of the DC power supply become smaller or larger.

[0057] FIG. 18 is a diagram showing the magnitude of harmonic components contained in the output phase voltage vu shown in FIG. 17B. The horizontal axis of FIG. 18 represents the order relative to the fundamental wave, and the vertical axis represents the magnitude of the harmonic components. From FIG. 18, it can be seen that the output phase voltage vu shown in FIG. 17B contains odd- and even-multiple harmonic components. Furthermore, excluding the integer multiples of three that do not appear in the harmonic components of the phase current flowing through the rotating machine 4, the 11th- and 13th-order harmonics are particularly large. However, since the first embodiment has already described the reduction of the 11th- and 13th-order harmonic components, in this embodiment, as an example, it is assumed that the 5th- and 7th-order harmonic components relative to the fundamental wave component of the output phase voltage vu excite mechanical resonance in the rotating machine 4, causing loud noise or torque pulsation and increasing losses in the rotating machine 4. The switching phases α1 to α3 and α7 to α9 are adjusted to reduce the 5th- and 7th-order harmonic components of the output phase voltage vu.

[0058] To find the switching phases α1 to α3 and α7 to α9 that reduce the fifth and seventh harmonic components of the output phase voltage vu, it is sufficient to solve an optimization problem so as to minimize the objective function fobj(α) expressed by, for example, equation (12).

[0059] Here, the objective function fobj(α) is set to minimize the square root of the sum of squares of the nth-order harmonic voltage vn in the range from second to pth order divided by the fundamental voltage v1, i.e., the total harmonic distortion. Furthermore, kn is a weighting coefficient, and by increasing the weighting coefficient kn for a specific order, it is possible to preferentially reduce harmonics of a specific order. Note that the switching phases α1 to α3 and α7 to α9 are constrained by the existence of a phase where the second command and the carrier signal cross each other every half cycle of the carrier signal, as shown in Equation (13). Therefore, the optimization problem is solved while observing these constraints.

[0060] In the second embodiment, the objective function fobj(α) is set so as to minimize the total harmonic distortion for the second to fiftieth harmonic voltages. In addition, by increasing the weighting coefficients for the fifth and seventh harmonic components as shown in equation (14), the fifth and seventh harmonic components are preferentially reduced.

[0061] FIG. 19 shows an example of the output phase voltage vu obtained by adjusting the switching phases α1 to α3 and α7 to α9 to minimize the fifth- and seventh-order harmonic components. FIG. 20 shows the magnitude of the harmonic components contained in the output phase voltage vu shown in FIG. 18 . The horizontal axis of FIG. 20 represents the order relative to the fundamental wave, and the vertical axis represents the magnitude of the harmonic components. By solving the optimization problem to minimize the objective function fobj(α) expressed by equation (12), as shown in FIG. 19 , the switching phases α1 to α3 and α7 to α9 of the output phase voltage vu shown in FIG. 17B are changed. FIG. 20 also shows that the fifth- and seventh-order harmonic components are reduced compared to the magnitude of the harmonic components shown in FIG. 18 . Therefore, the second command mu can be generated using a triangular-wave comparison PWM to output the output phase voltage vu shown in FIG. 19 .

[0062] 21A and 21B are diagrams showing an example of the phase voltage ( FIG. 21B ) output by the power conversion device 1B when the second command mu ( FIG. 21A ) generated so as to output the output phase voltage vu shown in FIG. 19 is provided to the gate signal generating means 8. FIG. 21B shows the waveform of the phase voltage output for the u-phase component when the gate signal g is generated from the second commands mu, mv, and mw using the carrier signal c and the gate signal generating means 8 to operate the inverter circuit 3, in the case where the amplitude M of the normalized first commands vu*, vv*, and vw* is 0.6. FIG. 21A shows the second command mu and the triangular wave carrier signal c, and FIG. 21B shows the output phase voltage vu output based on the second command mu and the triangular wave carrier signal c. It can be seen that the triangular wave comparison PWM using the second command mu and triangular wave carrier signal c shown in FIG. 21A can output a waveform similar to the output phase voltage vu shown in FIG. 19.

[0063] Note that, as an example of the waveform of the second command, the second command mu has been described above as a waveform in which the value of the second command mu remains constant throughout the period in which the carrier signal c changes from its minimum value to its maximum value and the period in which the carrier signal c changes from its maximum value to its minimum value, i.e., the waveform of the second command mu has a stepped shape. However, as long as the output phase voltage shown in Fig. 19 can be output, that is, as long as the second command mu is such that the phase at which it crosses the carrier signal c coincides with each switching phase α in Fig. 19, it goes without saying that the second command mu does not need to maintain a constant value over a half cycle of the carrier signal c, and may have a waveform that changes smoothly.

[0064] 22A and 22B are diagrams showing the magnitude (FIG. 22A) and phase (FIG. 22B) of the harmonic components of the second command mu shown in FIG. 21A. The horizontal axes in FIGS. 22A and 22B represent the order relative to the fundamental wave. The vertical axes in FIG. 21A represent the magnitude of the harmonic components, and the vertical axes in FIG. 22B represent the phase of the harmonic components relative to a sine wave. As shown in FIG. 22A, the second command mu is composed of a fundamental wave component, odd-multiple harmonic components, and even-multiple harmonic components. As shown in FIG. 22B, the fundamental wave component and odd-multiple harmonic components of the second command mu are composed of sine wave components synchronized at a phase of 0°, and the even-multiple harmonic components are composed of sine wave components synchronized at a phase minus 90° to the phase of the fundamental wave component.

[0065] Therefore, fundamental signal calculation unit 704, odd-multiple harmonic signal calculation unit 705, and even-multiple harmonic signal calculation unit 706B of modulated voltage generation means 7B only need to calculate fundamental signal mu_1, odd-multiple harmonic signal mu_odd, and even-multiple harmonic signal mu_even as shown in equation (15), where q is a value set so that 2q+1 is the maximum order considered, M1 is the amplitude of fundamental signal mu_1, M2n+1 is the amplitude of odd-multiple harmonic signal mu_odd at the 2n+1th order, and M2n is the amplitude of even-multiple harmonic signal mu_even at the 2nth order.

[0066] In this embodiment, the harmonic signals of each order of the odd-multiple harmonic signal mu_odd are sine waves, and their phases can be synchronized with the zero phase of the fundamental wave component of the first command vu*. Furthermore, if the harmonic signals of each order of the even-multiple harmonic signal mu_even, which are expressed as cosine waves in equation (15), are sine waves, then the minus 90° phase of the sine wave of the harmonic signals of each order can be synchronized with the zero phase of the fundamental wave component of the first command vu*. Furthermore, the harmonic signals of each order of the odd-multiple harmonic signal mu_odd and the even-multiple harmonic signal mu_even are not limited to sine waves and can have distorted waveforms. They can be periodic signals in which the beginning and end of one cycle of the fundamental wave component of the first command have the same value and cross zero at least once per cycle. Furthermore, if the odd-multiple harmonic signal mu_odd and the even-multiple harmonic signal mu_even each contain at least one harmonic signal of one order, the switching phase can be controlled to control the harmonic components generated by switching, compared to conventional synchronous PWM, thereby achieving the effect of the present disclosure of increasing or decreasing load noise, loss, etc.

[0067] In this second embodiment, only the second command mu has been described, but since the waveforms of the three-phase second commands mu, mv, and mw are waveforms in which the phases of the fundamental wave components are shifted by 120° from each other, the second commands mv and mw of the remaining two phases can also be generated in the same way as the second command mu.

[0068] As described above, in the power conversion device 1B according to the second embodiment, the phase of the carrier signal c is shifted by 90° compared to that of the power conversion device 1A according to the first embodiment. However, even in such a case, when the number of pulses is even, it is possible to control harmonics of a predetermined order of the voltage or current generated by PWM with a relatively simple configuration using a triangular wave comparison type PWM mounted on a general-purpose controller IC. This has the effect of controlling the noise, torque pulsation, vibration, and loss of the rotating machine when the load is a rotating machine.

[0069] 23 is a diagram showing the configuration of a power conversion device 1C according to embodiment 3. The power conversion device 1C includes an inverter circuit 3 connected to a DC power supply 2 and a rotating machine 4 as a load, a voltage command generating means 5, a carrier signal generating means 6A, a modulated voltage generating means 7C, and a gate signal generating means 8.

[0070] The power conversion device 1C has a modulated voltage generating means 7C whose function is different from that of the modulated voltage generating means 7A of the power conversion device 1A according to embodiment 1. Below, a description of the parts common to the power conversion device 1A according to embodiment 1 will be omitted, and the following description will mainly focus on the parts that are different from the power conversion device 1A.

[0071] Fig. 24 is a block diagram showing the configuration of the modulated voltage generating means 7C shown in Fig. 23. The modulated voltage generating means 7C includes a three-phase to two-phase conversion unit 701, an amplitude calculation unit 702, a phase calculation unit 703, a storage unit 707, and a modulated voltage calculation unit 708.

[0072] The memory unit 707 stores the waveforms of the second commands mu, mv, mw based on the first commands vu*, vv*, vw* or the signals constituting the second commands mu, mv, mw, and outputs the stored waveforms of the second commands mu, mv, mw or the signals constituting the second commands mu, mv, mw to the modulation voltage calculation unit 708.

[0073] The waveforms of the second commands mu, mv, and mw can be determined by setting the amplitude M of the normalized first commands vu*, vv*, and vw*, the u-phase fundamental wave phase θv, and the harmonic order to be reduced. Therefore, for example, for each combination of harmonic orders to be reduced, the waveforms of the second commands mu, mv, and mw are determined in advance as a function of M and θv, or for each M and θv. The storage unit 707 then stores the waveforms of the second commands mu, mv, and mw based on the amplitude M of the normalized first commands vu*, vv*, and vw* and the u-phase fundamental wave phase θv. Alternatively, the storage unit 707 stores the waveforms of the fundamental signals mu_1, mv_1, mw_1, odd-multiple harmonic signals mu_odd, mv_odd, mw_odd, and even-multiple harmonic signals mu_even, mv_even, mw_even that constitute the second commands mu, mv, and mw, which are based on the amplitudes M of the normalized first commands vu*, vv*, and vw* and the fundamental phase θv of the u phase. Alternatively, the storage unit 707 stores the amplitudes of the fundamental signals mu_1, mv_1, mw_1, odd-multiple harmonic signals mu_odd, mv_odd, mw_odd, and even-multiple harmonic signals mu_even, mv_even, and mw_even that constitute the second commands mu, mv, and mw, which are based on the amplitudes M of the normalized first commands vu*, vv*, and vw*.

[0074] In the third embodiment, the second commands mu, mv, and mw have positive and negative symmetry between the phases of 0° to 180° and the phases of 180° to 360°, and therefore only the waveforms of the second commands mu, mv, and mw corresponding to half the fundamental cycle of the first commands vu*, vv*, and vw* may be stored. Alternatively, only the waveforms of the fundamental signals mu_1, mv_1, and mw_1, the odd-multiple harmonic signals mu_odd, mv_odd, and mw_odd, and the even-multiple harmonic signals mu_even, mv_even, and mw_even corresponding to half the fundamental cycle of the first commands vu*, vv*, and vw* may be stored.

[0075] Based on the first commands vu*, vv*, vw*, the modulation voltage calculation unit 708 refers to the waveforms of the second commands mu, mv, mw or the signals constituting the second commands mu, mv, mw stored in the memory unit 707, and calculates the second commands mu, mv, mw from the waveforms of the second commands mu, mv, mw or the signals constituting the second commands mu, mv, mw that have been referred to.

[0076] Specifically, as shown in equation (16), the modulation voltage calculation unit 708 refers to the waveforms of the second commands mu, mv, and mw stored in the memory unit 707 based on the amplitude M of the normalized first commands vu*, vv*, and vw* and the fundamental wave phase θv of the u phase, i.e., as a function of M and θv or for each value of M and θv, and calculates mu, mv, and mw from the waveforms of mu, mv, and mw that have been referenced.

[0077] Alternatively, as shown in equation (17), the modulation voltage calculation unit 708 references the waveforms of the fundamental signals mu_1, mv_1, mw_1, odd-multiple harmonic signals mu_odd, mv_odd, mw_odd, and even-multiple harmonic signals mu_even, mv_even, mw_even stored in the storage unit 707 based on the amplitudes M of the normalized first commands vu*, vv*, vw* and the fundamental wave phase θv of the u phase, that is, as a function of M and θv or for each value of M and θv, and calculates the second commands mu, mv, mw from the waveforms of the fundamental signals mu_1, mv_1, mw_1, odd-multiple harmonic signals mu_odd, mv_odd, mw_odd, and even-multiple harmonic signals mu_even, mv_even, mw_even that have been referenced.

[0078] Alternatively, as shown in equation (18), the modulation voltage calculation unit 708 references the amplitudes of each order of the fundamental signals mu_1, mv_1, mw_1, odd-multiple harmonic signals mu_odd, mv_odd, mw_odd, and even-multiple harmonic signals mu_even, mv_even, mw_even stored in the storage unit 707 based on the amplitudes M of the normalized first commands vu*, vv*, vw*, that is, as a function of M or for each value of M, and calculates the second commands mu, mv, mw from the referenced amplitudes of each order of the fundamental signals mu_1, mv_1, mw_1, odd-multiple harmonic signals mu_odd, mv_odd, mw_odd, and even-multiple harmonic signals mu_even, mv_even, mw_even and the fundamental phase θv of the u phase.

[0079] Here, if the memory unit 707 only stores the waveforms of the second commands mu, mv, and mw corresponding to half a cycle of the fundamental wave, the second commands mu, mv, and mw are symmetrical in positive and negative between the phases of 0° to 180° and the phases of 180° to 360°, and by taking advantage of this, it is possible to reproduce the second commands mu, mv, and mw for one cycle of the fundamental wave by referring to the waveforms of the second commands mu, mv, and mw corresponding to half a cycle of the fundamental wave.

[0080] Alternatively, if the memory unit 707 only stores the waveforms of the fundamental wave signals mu_1, mv_1, mw_1, odd-multiple harmonic signals mu_odd, mv_odd, mw_odd, and even-multiple harmonic signals mu_even, mv_even, mw_even, which correspond to half a cycle of the fundamental wave, then the second commands mu, mv, and mw for one cycle of the fundamental wave can be reproduced by referring to the waveforms of the fundamental wave signals mu_1, mv_1, mw_1, odd-multiple harmonic signals mu_odd, mv_odd, mw_odd, and even-multiple harmonic signals mu_even, mv_even, mw_even, which correspond to half a cycle of the fundamental wave.

[0081] Since the waveforms of the second commands mu, mv, and mw are waveforms in which the phases of each phase are shifted by 120° from each other, the modulation voltage calculation unit 708 may generate the second command mu based on the fundamental wave phase θv of the u phase, and then generate the second commands mu, mv, and mw for three phases by shifting the phases.

[0082] As described above, according to the power conversion device 1C of the third embodiment, even when the number of pulses is even, it is possible to control harmonics of a predetermined order of the voltage or current generated by PWM using a triangular wave comparison PWM mounted on a general-purpose controller IC with a relatively simple configuration. This provides the effect of being able to control the noise, torque pulsation, vibration, and loss of the rotating machine. At the same time, because the data for generating the second commands mu, mv, and mw described in the first embodiment is stored in advance in the storage unit 707, it is possible to reduce the calculation load when calculating the second commands mu, mv, and mw.

[0083] Here, the hardware configuration of power conversion devices 1A, 1B, and 1C according to embodiments 1 to 3 will be described. Here, power conversion devices 1A, 1B, and 1C are collectively referred to as power conversion device 1. The functions of drive controller 10 in power conversion device 1, i.e., voltage command generation means 5, carrier signal generation means 6A and 6B, modulated voltage generation means 7A, 7B, and 7C, and gate signal generation means 8, etc., can be realized using a processing circuit. The processing circuit may be dedicated hardware such as a dedicated processing circuit, or may be a processor and storage means.

[0084] 25 is a diagram showing a configuration example of the power conversion device 1 when dedicated hardware is used. When dedicated hardware is used, the drive controller 10 is configured with a dedicated processing circuit. The dedicated processing circuit may be a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a combination thereof. Each of the multiple functions of the power conversion device 1 described above may be realized by a different dedicated processing circuit, or multiple functions of the power conversion device 1 may be realized collectively by a dedicated processing circuit.

[0085] FIG. 26 is a diagram showing an example of the configuration of the power conversion device 1 when a processor and storage means are used. When the processor 15 and storage means 16 are used as the drive controller 10, the functions of the power conversion device 1 described above are realized by software, firmware, or a combination of these. The software or firmware is written as a program, and the processor 15 reads and executes the program stored in the storage means 16. It can also be said that these programs cause a computer to execute the procedure or method of each function of the power conversion device 1. The processor 15 is a CPU (Central Processing Unit) and is also called a processing device, arithmetic unit, microprocessor, microcomputer, DSP (Digital Signal Processor), etc. The storage means 16 is, for example, a non-volatile or volatile semiconductor memory such as a read-only memory (ROM), an erasable programmable read-only memory (EPROM), or an electrically programmable programmable read-only memory (EEPROM), or a flexible disk, an optical disk, a compact disk, a digital versatile disk (DVD), etc. Also, some of the functions of the power conversion device 1 may be realized by dedicated hardware, and some may be realized by software or firmware.

[0086] Although various exemplary embodiments and examples are described in this disclosure, the various features, aspects, and functions described in one or more embodiments are not limited to the application of a particular embodiment, but may be applied to the embodiments alone or in various combinations. Therefore, countless modifications not illustrated are contemplated within the scope of the technology of this disclosure. For example, this includes cases where at least one component is modified, added, or omitted, and even cases where at least one component is extracted and combined with components of another embodiment.

[0087] For example, in the above embodiment, the inverter circuit 3 is a three-phase inverter circuit, but it may be an inverter circuit with another number of phases, and various inverter circuits can be used, such as a multilevel inverter such as a three-level inverter or a five-level inverter.

[0088] Furthermore, in the first and third embodiments, when the carrier signal generating means 6A synchronizes the median of the triangular wave that is the carrier signal c with the phase 0° of the first commands vu*, vv*, and vw*, the median of the decreasing side of the triangular wave, i.e., a phase of 90° in the example of FIG. 4, is used. However, the carrier signal generating means 6A may also synchronize the median of the increasing side of the triangular wave, i.e., a phase of 270° in the example of FIG. 4, with the phase 0° of the first commands vu*, vv*, and vw*. In this case, as described in the first embodiment, when the odd-multiple harmonic signals and even-multiple harmonic signals that constitute the second commands are sine waves, the zero phase of each harmonic signal is formed by a sine wave synchronized with the zero phase of the fundamental wave component of the first command.

[0089] Furthermore, in the second embodiment, the carrier signal generating means 6B synchronizes the minimum value of the triangular wave that is the carrier signal c, which is a phase of 180° in the example of FIG. 4, with the phase of 0° of the first commands vu*, vv*, and vw*. However, the carrier signal generating means 6B may also synchronize the maximum value of the triangular wave, which is a phase of 0° or 360° in the example of FIG. 4, with the phase of 0° of the first commands vu*, vv*, and vw*. In this case, as described in the second embodiment, when the odd-multiple harmonic signal and the even-multiple harmonic signal that constitute the second command mu are sine waves, the odd-multiple harmonic component is composed of a sine wave synchronized with the phase of 0° of the fundamental wave component, and the even-multiple harmonic component is composed of a sine wave whose phase -90° is synchronized with the phase of 0° of the fundamental wave component.

[0090] 10 Drive controller, c Carrier signal, mu, mv, mw Second command, Q, Qup, Qun, Qvp, Qvn, Qwp, Qwn Semiconductor switching element, vu*, vv*, vw* First command, g, gup, gun, gvp, gvn, gwp, gwn Gate signal

Claims

1. A power conversion device that converts DC voltage to AC voltage by turning semiconductor switching elements on and off, wherein a drive controller that generates gate signals for driving the semiconductor switching elements on and off is configured to: generate a first command that is a voltage command for a target voltage of the AC voltage; generate a triangular wave carrier signal whose frequency is an even multiple of the frequency of the fundamental wave component of the first command; and generate the gate signal from the carrier signal and a second command that is composed of a harmonic signal that includes at least one odd-multiple harmonic signal whose frequency is an odd-multiple and one even-multiple harmonic signal whose frequency is an even multiple of the frequency of the fundamental wave component of the first command, and the first command.

2. The power conversion device according to claim 1, wherein the odd-multiple harmonic signal and the even-multiple harmonic signal each have the same value at the beginning and end of one cycle of the fundamental wave component of the first command, and are periodic signals with at least one zero crossing point within one cycle.

3. A power conversion device according to claim 1 or 2, wherein the carrier signal has a triangular wave whose median, minimum or maximum value is synchronized with the zero phase of the fundamental wave component of the first command.

4. A power conversion device as described in claim 1, wherein the carrier signal has a triangular wave whose median value is synchronized with the phase zero of the fundamental wave component of the first command, and the odd-multiple harmonic signal and the even-multiple harmonic signal constituting the second command are sine waves, and the phase zero of the sine wave is synchronized with the phase zero of the fundamental wave component of the first command.

5. A power conversion device as described in claim 1, wherein the minimum or maximum value of the triangular wave of the carrier signal is synchronized with phase zero of the fundamental wave component of the first command, the odd-multiple harmonic signal and the even-multiple harmonic signal constituting the second command are sine waves, phase zero of the odd-multiple harmonic signal is synchronized with phase zero of the fundamental wave component of the first command, and phase minus 90° of the even-multiple harmonic signal is synchronized with phase zero of the fundamental wave component of the first command.

6. A power conversion device according to any one of claims 1 to 5, wherein the waveform of the second command is a waveform that does not change and remains constant during both the period in which the carrier signal changes from its maximum value to its minimum value and the period in which the carrier signal changes from its minimum value to its maximum value.

7. A power conversion device according to any one of claims 1 to 6, wherein the drive controller references the generated first command to call up a second command candidate stored with the amplitude and phase of a voltage command for a target voltage within a predetermined range as parameters, and sets the second command as the second command.

8. The power conversion device according to claim 7, wherein the drive controller stores the second command candidates as waveforms of one cycle or half cycle of the fundamental wave component for each amplitude and phase of the voltage command.

9. The power conversion device according to claim 7, wherein the drive controller stores the second command candidates as a function of the amplitude and phase of the voltage command.

10. A power conversion device according to any one of claims 1 to 9, wherein the AC voltage is supplied to a load, and the drive controller generates the first command based on one or more of the torque, rotational speed, position, current amplitude, phase, and frequency of the rotating machine when the load is a rotating machine, and generates the first command based on one or more of the amplitude, phase, and frequency of the current of the load when the load is other than a rotating machine.

11. A power conversion device according to any one of claims 1 to 10, wherein the AC voltage is supplied to a load, and the drive controller generates the second command so that, when the load is a rotating machine, the amplitude of a predetermined frequency component included in at least one parameter of the voltage, current, torque, loss, vibration of the rotating machine, the DC voltage, and the DC current flowing on the DC side of the power conversion device is decreased or increased, compared to when the AC voltage is supplied to a load and the drive controller generates the gate signal by comparing the carrier signal with a sine wave modulated signal of the fundamental wave component of the first command to operate the power conversion device; or, when the load is other than a rotating machine, the amplitude of a predetermined frequency component included in at least one parameter of the voltage, current, loss, DC voltage, and the DC current flowing on the DC side of the power conversion device is decreased or increased.

12. A power conversion device according to any one of claims 1 to 11, comprising an inverter circuit in which a DC voltage is applied to both ends and a plurality of legs, each leg having two reverse-conducting semiconductor switching elements connected in series, are connected in parallel.

13. A power conversion device according to any one of claims 1 to 12, wherein the semiconductor switching elements are formed from wide bandgap semiconductors.

Images