JP2010213377A - Power conversion equipment and power conversion method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide power conversion equipment which reduces the peak level of noise sufficiently by reducing irregularities formed in the spectra of higher harmonics of a carrier frequency, in temporally changing the value of the carrier frequency discretely. <P>SOLUTION: In the power conversion equipment, which converts input power into a three-phase AC by a PWM inverter 2, a controller 5, which controls the PWM inverter 2, includes a carrier frequency changer 10, which temporally changes the frequency of carrier waves in PWM control discretely and periodically, and a carrier output part 11, which outputs carriers to PWM comparators 8a, 8b, and 8c. The carrier frequency changer 10 determines the variable range of the carrier frequency so that the frequency spectra of the higher harmonics in each degree of carriers may continue, in a desired frequency band. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a power conversion device and a power conversion method that reduce electromagnetic noise due to switching.

  As a countermeasure against EMI noise in a power converter, a stepping motor drive circuit described in Patent Document 1 is known. This drive circuit forms an H bridge with four FET switches, and controls the magnitude and direction of the current applied from the DC power source to the motor. The FET switch is opened and closed by PWM control, and EMI noise is generated by opening and closing the FET switch. The spectrum of EMI noise shows a peak at the carrier frequency of PWM and its harmonic frequency.

  In this prior art, in order to reduce the EMI noise level, the carrier frequency in PWM control is continuously changed in a sine wave shape with the passage of time. In general, it is considered that the peak level becomes lower when the carrier frequency change range is wider. As an example, there is an article of Non-Patent Document 1. However, this is a case where the clock signal is frequency-modulated in an analog manner.

JP-A-9-99795

“Spread Spectrum Clock Generator” Akihisa Sakurai, Monthly EMC, No.109 (1997.5.5) p.22-p.28

  When the carrier frequency value is discretely changed by digital control, the EMI noise spectrum is obtained by using each discrete value of the carrier frequency and operating under the condition that the carrier frequency is constant. It can be considered almost quasi-static, that is, almost the same as the average value of the EMI spectrum. However, when the value of the carrier frequency is discretely changed with time, irregularities are formed in the harmonic EMI spectrum according to the harmonic distribution of each carrier frequency, and the peak level of noise cannot be sufficiently reduced. There was a problem.

  In order to solve the above problems, the present invention is a power conversion device that converts input power into a desired form by performing PWM control of the switching means, and the frequency of the carrier wave for PWM control is discretely and periodically converted to time. When changing, the variable range of the carrier frequency is determined so that the spread spectrum of the harmonics of each order of the carrier wave is continuous within a desired frequency band.

  According to the present invention, the spread spectrum of the harmonics of the respective orders of the PWM carrier wave in a desired frequency band is continuous, so that the EMI spectrum due to the harmonic waves of the carrier wave can be flattened and the influence of EMI noise can be suppressed. There is.

It is a block diagram which shows the power conversion system provided with the Example of the power converter device which concerns on this invention. It is a time chart which shows the example of the time change of the triangular wave shape of the carrier frequency in digital control. It is a figure which shows the general example of the spread spectrum of the harmonic of the carrier frequency in this invention. It is a figure which shows the spreading | diffusion spectrum example of the harmonic of the carrier frequency in an Example. It is a figure which shows the calculated value of distribution of the harmonic of a carrier frequency in an Example. It is a figure which shows the example of application of this invention to the load drive by a single switch element. It is a figure which shows the example of application of this invention to a DC-DC converter. It is a figure which shows the sawtooth time change example of the carrier frequency in digital control. It is a figure which shows the time change of the periodic waveform of the carrier frequency in digital control. It is a figure which shows the time change of the sine wave shape of the carrier frequency in digital control. It is a figure which shows the example of a spread spectrum of the harmonic of a carrier frequency. It is a figure which shows the calculated value of distribution of the harmonic of a carrier frequency.

  Next, the present invention will be described in detail with reference to the drawings. Prior to the description of the embodiment, the knowledge about the noise spectrum of the power converter by the present inventors will be described. In the PWM-controlled power conversion device, even if the carrier frequency is changed discretely with time, the time average of the number of switching times does not change, so the total energy of the EMI spectrum formed at that time does not change. Therefore, the peak level is reduced as the noise spectrum becomes flatter. Conversely, the peak level increases as the unevenness of the EMI spectrum increases.

  For example, let us consider a case where it is desired to lower the level of the EMI spectrum of the AM radio broadcast band (approximately 500 kHz to 1650 kHz) as the desired frequency band. As shown in FIG. 11, when the range of change of the carrier frequency is 15 kHz ± 0.1 kHz, the 33rd-order harmonic spectrum of the carrier frequency spreads in the range of 491.7 kHz to 498.3 kHz. The spectrum of the 34th harmonic is diffused in the range of 506.6 kHz to 513.4 kHz. However, since the harmonic EMI spectrum does not enter the band between 498.3 kHz and 506.6 kHz between the 33rd-order and 34th-order harmonic spread spectrum, a dent in the spectrum is formed. As described above, even if the carrier frequency is changed with time, the total energy of EMI does not change. Therefore, if the EMI spectrum is diffused in this recess, the level of the EMI spectrum can be further reduced.

  On the other hand, FIG. 12A shows the distribution of harmonics when the carrier frequency change range is 13.39 kHz to 17.05 kHz, and the number of carrier frequency values used at that time is nine. It is. Similarly to the above, the horizontal axis represents the frequency band of AM radio broadcasting, and the vertical axis represents the distribution obtained from the proportion of each harmonic occupying one period of time change of the carrier frequency. Here, the calculation is performed assuming that the resolution bandwidth (RBW) is 9 kHz. From this distribution, it can be seen that there is a high-level band centered around a frequency of about 500 kHz and an integral multiple thereof. This occurs due to the change range of the carrier cycle, which is the reciprocal of the carrier frequency, and the number of values of the carrier frequency (cycle) used for the time change.

In this case, the average interval of the carrier period, which is the reciprocal of the carrier frequency value used for changing the carrier frequency, that is, the difference between the maximum carrier period and the minimum carrier period is the number of carrier frequency values minus 1 (in this case 8) the reciprocal of the number divided by
8 / {1 / (13.39 × 10 ³ ) −1 / (17.05 × 10 ³ )}
≒ 500 × 10 ³ Hz
= 500kHz
A band in which harmonics concentrate is formed in the vicinity of an integral multiple thereof.

  This band is a band with a high level of EMI spectrum. As described above, depending on the range of change of the carrier frequency and the number used for changing the carrier frequency, even if the carrier frequency is changed with time, irregularities are formed in the EMI spectrum, and the level may not be sufficiently reduced.

  12B, as in FIG. 12A, the carrier frequency change range is 13.39 kHz to 17.05 kHz, and the number of carrier frequency values used at that time is nine. Although it is a distribution of harmonics, it is a distribution of harmonics when the value of the carrier frequency is different from the case of FIG. Even in this case, a similar tendency is formed, where a band in which harmonics are concentrated is formed in the vicinity of a frequency of 500 kHz and an integral multiple thereof.

  The present invention has been made on the basis of the above knowledge, and by setting a variable range of the carrier frequency that reduces the unevenness of the EMI spectrum of the harmonic of the carrier frequency, the power conversion with reduced peak level of EMI Providing equipment.

  Next, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a system configuration diagram of a power conversion system 1 for explaining an embodiment of a power conversion device according to the present invention.

  A power conversion system 1 in FIG. 1 includes a PWM inverter 2 that converts direct current into three-phase alternating current by PWM control, a three-phase brushless motor (hereinafter abbreviated as three-phase motor) 3, current sensors 4a, 4b, and 4c. A control device 5 that supplies a PWM control signal to the PWM inverter 2 as main components.

  The control device 5 includes a current command value generation unit 6, PID control units 7a, 7b, and 7c, PWM comparison units 8a, 8b, and 8c, signal inversion units 9a, 9b, and 9c, a carrier frequency change unit 10, and a carrier output unit 11. Prepare.

  The PWM inverter 2 includes six switching elements Tu +, Tu−, Tv +, Tv−, Tw +, and Tw− formed of semiconductor elements such as IGBTs (Insulated Gate Bipolar Transistors). The PWM inverter 2 selects the positive or negative electrode of the DC power source composed of the battery B and the capacitor C in accordance with the control of the PWM comparators 8a, 8b, and 8c, and selects the selected electrodes as the U phase, V phase, and W of the three-phase motor 3. Connect to each electrode of the phase.

  Current sensors 4a, 4b, and 4c detect U-phase, V-phase, and W-phase current values supplied from PWM inverter 2 to three-phase motor 3, respectively. The current command value generation unit 6 generates a sinusoidal current command value so that the detection values of the current sensors 4a, 4b, and 4c are converted into a sinusoidal alternating current. The PID control units 7a, 7b, and 7c detect the current command values and the current sensors 4a, 4b, and 4c so that the detection values of the current sensors 4a, 4b, and 4c follow the current command values generated by the current command value generation unit 6. Based on the value, the level command value corresponding to the duty command value output to the PWM comparators 8a, 8b, 8c is PID controlled. The carrier output unit 11 generates a triangular carrier signal for each phase and outputs it to the PWM comparison units 8a, 8b, and 8c.

  The PWM comparison units 8a, 8b, and 8c compare the magnitude relationship between the level command value output from the PID control units 7a, 7b, and 7c and the triangular wave carrier signal, and the switching element of the PWM inverter 2 according to the magnitude relationship A signal for controlling ON / OFF of Tu +, Tv +, and Tw + is output to the PWM inverter 2. The signal inverters 9a, 9b, and 9c invert the outputs of the PWM comparators 8a, 8b, and 8c, and output signals to the PWM inverter 2 that control ON / OFF of the switching elements Tu−, Tv−, and Tw−. .

  Here, the operation of the PWM comparison unit 8a will be specifically described by taking control of the U-phase switching elements Tu + and Tu− as an example. When the output value of the PID control unit 7a is larger than the triangular wave carrier signal, the PWM comparison unit 8a controls the switching elements Tu + and Tu− to be in the on state and the off state, respectively, thereby changing the positive voltage to the U phase of the motor. Apply. On the contrary, when the output value of the PID control unit 7a is smaller than the triangular wave carrier signal, the negative voltage is reduced to U of the three-phase motor 3 by controlling the switching elements Tu + and Tu− respectively to the off state and the on state. Apply to phase.

  In this embodiment, it is assumed that the control device 5 is composed of a microcomputer. The microcomputer includes, for example, a CPU that is an arithmetic control unit, a ROM that stores a program and a control map, a working RAM, and an input / output interface.

  The carrier frequency changing unit 10 discretely changes the frequency fc of the triangular wave carrier signal output from the carrier output unit 11 with time. For example, when the carrier frequency is changed to a triangular wave shape, the carrier frequency changes as shown in FIG.

  Next, a method for setting the carrier frequency change width for flattening the EMI spectrum and reducing the level by discretely changing the carrier frequency over time will be described.

  First, a method for determining the minimum value of the carrier frequency change width so that the spread spectrum of the harmonics of the carrier frequency is continuous in a desired band will be described.

  As shown in FIG. 3, the center value of the carrier frequency is fcc, and the variable width that can be changed up and down from the center value fcc is ± Δfc. The variable range of the carrier frequency is from fcc-Δfc to fcc + Δfc, and in this case, the spread spectrum of the nth-order harmonic spreads in a band from about n (fcc-Δfc) to n (fcc + Δfc). Further, the spread spectrum of the (n + 1) -order harmonic spreads in a band from approximately (n + 1) (fcc-Δfc) to (n + 1) (fcc + Δfc). Fmin is the lowest frequency of the desired band in which the level of the EMI spectrum is to be reduced, and n is the maximum integer that satisfies n ≦ Fmin / fc.

The requirement now is that the spread spectrum of the harmonics is continuous in the band higher than Fmin. For that purpose, the lowest frequency of the spread spectrum of the (n + 1) th order harmonics only needs to be equal to or lower than the maximum frequency of the spread spectrum of the nth order harmonics. That is, Δfc may be determined so as to satisfy the inequality shown in Expression (1).
(n + 1) (fcc-Δfc) ≦ n (fcc + Δfc) (1)
When formula (1) is arranged, formula (2) is obtained.
fcc / (2n + 1) ≦ Δfc (2)
That is, Δfc may be set to a value equal to or greater than fcc / (2n + 1), and the carrier frequency may be changed over time within a variable range (fcc−Δfc to fcc + Δfc). The carrier frequency may be defined in a variable range by system requirements such as heat generation and loss due to switching and generation of acoustic noise. In that case, the carrier frequency variable range is determined so as to satisfy Expression (2) within the variable range determined by the system requirements.

  As an example using specific values, FIG. 4 schematically shows an EMI spectrum when fcc is 15 kHz, Δfc is 0.5 kHz, and Fmin is 500 kHz. In this case, Expression (2) is satisfied, and the diffusion spectrum of the 33rd-order harmonic and the diffusion spectrum of the 34th-order harmonic, which are the lowest orders in the band of Fmin or higher, are continuous. In the band beyond this, the spread spectrum is always continuous.

  In the PWM control, for example, there is a case where the drive is performed with a duty of 50%. In this case, only the odd-order harmonics exist in the output waveform of the PWM inverter 2, and there are no even-order harmonics. In such a case, in order for the spread spectrum of adjacent harmonics to be continuous, the spread spectrum of a certain odd-order harmonic and the spread spectrum of the next odd-order harmonic need to be continuous. .

In such a case, if the change width of the carrier frequency is set to be twice the range of the equation (2), the spread spectrum of harmonics is continuous. That is, the change width of the carrier frequency is determined so as to satisfy Expression (3).
2fcc / (2n + 1) ≦ Δfc (3)

  Next, a method for defining the maximum value of the carrier frequency variation will be described. In this case, the band where the harmonics of the carrier frequency formed due to the change range of the carrier frequency and the number of carrier frequency values used for the change, that is, the band where the level of the EMI spectrum is high is the desired band. This is a method for defining the maximum value of the change width of the carrier frequency in order to make the band outside the above range, particularly higher than the desired band.

Here, the band in which the harmonics of the carrier frequency formed due to the change range of the carrier frequency and the number of carrier frequency values used for the change are concentrated is the average of the carrier period, which is the reciprocal of the carrier frequency used. , In other words, a frequency band that is the reciprocal of a value obtained by dividing the difference between the maximum value and the minimum value of the carrier period by the number of carrier frequency values minus 1 and a frequency that is an integer multiple of the frequency. What is necessary is just to prescribe | regulate the range of change of a carrier frequency so that this band may not enter into a desired band. Expressed by the equation, if the maximum and minimum values of the carrier period are Tcmax and Tcmin, respectively, and the number of carrier frequency values is N,
(N-1) / (Tcmin-Tcmax) (4)
And the harmonic distribution of the carrier frequency tends to concentrate in a band around the frequency that is an integral multiple of the frequency. Therefore, if Fmax is the maximum frequency of the desired frequency band,
Fmax ≦ (N-1) / (Tcmax-Tcmin) (5)
By selecting the carrier frequency so as to satisfy the above, and using the value to change the carrier frequency, the influence of EMI can be suppressed.

For example, let us consider that the above formula (4) and an integral multiple thereof do not fall within the AM radio band (approximately 500 kHz to 1650 kHz). Assume that the number of carrier frequency values used for changing the carrier frequency is nine, and the maximum and minimum carrier frequency values are 15.23 kHz and 14.78 kHz, respectively. In this case, when the value of Expression (4) is obtained, Expression (6) is obtained.
(9-1) / {1 / 15.23 × 10 ³ ) -1 / (14.78 × 10 ³ )} ≒ 4 × 10 ⁶ Hz = 4MHz (6)
In other words, in this case, the lowest frequency in the band where the distribution of harmonics of the carrier frequency is likely to be concentrated is around 4 MHz, which is higher than the AM radio band, so that the EMI spectrum in the AM radio band can be flattened. become.

In the above equation, Tcmax (= 1 / (fcc-Δfc)) and Tcmin (= 1 / (fcc + Δfc)) are rewritten using approximation, and the equation (7) is obtained. It can be seen that the equation defines the maximum value.
Δfc ≦ N · fcc ² / (2Fmax) (7)
N may be limited by the CPU clock used. For example, when the clock frequency is 1 MHz, the reciprocal clock period is 1 μsec. Since the carrier period is limited to a value that is an integral multiple of the clock period, for example, if the carrier period is in the range of 100 μsec to 110 μsec, only 11 elements can be used at most. The above is a description of a method for defining the range of the carrier frequency when the number of N that can be used for the time change of the carrier frequency is limited.

On the other hand, there are cases where the change range of the carrier frequency is limited due to system requirements such as heat generation and sound vibration (that is, Δfc has an upper limit). In this case, to prevent the band where the harmonics of the carrier frequency formed due to the carrier frequency change range and the number of carrier frequency values used for the change from entering the desired band, What is necessary is just to determine the number of frequencies appropriately. It can be defined by the following formula (8) based on the above-mentioned concept.
Fmax (Tcmax-Tcmin) + 1 ≦ N (8)
If this is similarly organized using approximation, Equation (9) is obtained.
2FmaxΔfc / fcc ² ≦ N (9)
The power converter that drives a three-phase brushless motor by PWM control has been described as an example. However, the present invention is not limited to the configuration of the power converter, and changes the form of power by opening and closing the switch. This defines a discrete carrier frequency value used when changing the carrier frequency of the power conversion device capable of performing the above.

  For example, a configuration in which a motor is driven by an H bridge shown in Patent Document 1 of the prior art, a configuration in which some load is opened and closed by a single switch as shown in FIG. 6, and a DC-DC converter as shown in FIG. Etc. are also applicable.

  Further, the present invention is not limited to the time change pattern of a specific carrier frequency. For example, the present invention can be applied to time-varying patterns of various carrier frequencies such as a sawtooth wave change as shown in FIG. 8, a periodic wave shape change as shown in FIG. 9, and a sine wave change as shown in FIG. In that case, what is necessary is just to determine a frequency value by the method demonstrated above as a carrier frequency to be used.

  According to the present embodiment described above, the frequency of the carrier wave used for PWM control can be varied within the desired frequency band so that the spread spectrum of the harmonics of each order of the carrier wave is continuous. By determining the above, there is an effect that the EMI spectrum in a desired frequency band can be flattened and the influence of EMI noise can be suppressed.

  Further, according to the present embodiment, in a desired frequency band, the harmonic frequency of an arbitrary order of the upper limit frequency of the carrier wave is equal to or lower than the harmonic frequency of an order higher than the order of the lower limit frequency of the carrier wave. Thus, by determining the variable range of the carrier frequency, the harmonic spectrum of each order of the carrier in the desired frequency band is continuous, the EMI spectrum can be flattened, and the influence of EMI noise can be suppressed. effective.

Further, according to the present embodiment, the variable range of the carrier frequency is fcc-Δfc to fcc + Δfc, n is the maximum integer satisfying n ≦ Fmin / fcc, and Fmin is the lowest frequency of the desired frequency band. The variable range of the carrier frequency is
fcc / (2n + 1) ≦ Δfc
Therefore, there is an effect that the variable range of the carrier frequency that can suppress the influence of EMI noise can be calculated easily and accurately.

  Further, according to the present embodiment, in a desired frequency band, the harmonic frequency of an arbitrary order of the upper limit frequency of the carrier wave is equal to or lower than the harmonic frequency of the second order higher than the order of the lower limit frequency of the carrier wave. As described above, by determining the variable range of the carrier frequency, even when the carrier includes only odd-order or even-order harmonics, the EMI spectrum in the desired frequency band is flattened and the influence of EMI noise is suppressed. There is an effect that can be.

Further, according to the present embodiment, the variable range of the carrier frequency is fcc-Δfc to fcc + Δfc, n is the maximum integer satisfying n ≦ Fmin / fcc, and Fmin is the lowest frequency of the desired frequency band. The variable range of the carrier frequency is
2fcc / (2n + 1) ≤Δfc
Therefore, the variable range of the carrier frequency that can suppress the influence of EMI noise can be calculated easily and accurately, even when the carrier wave contains only odd-order or even-order harmonics. There is an effect that can be done.

  Further, according to the present embodiment, the variable range of the carrier frequency has a band in which harmonics of the carrier frequency due to the range of the carrier frequency change when the carrier frequency is changed over time and the number of carrier frequencies used are concentrated. Since it is determined not to fall within the desired frequency band, the EMI spectrum in the desired frequency band can be flattened and the effect of EMI noise can be suppressed.

  Further, according to the present embodiment, the variable range of the carrier frequency has a band in which harmonics of the carrier frequency due to the range of the carrier frequency change when the carrier frequency is changed over time and the number of carrier frequencies used are concentrated. Since it is determined to be larger than the maximum frequency of the desired frequency band, there is an effect that the influence of EMI noise can be suppressed by flattening the EMI spectrum in the desired frequency band.

Further, according to the present embodiment, the change range of the carrier cycle, which is the reciprocal of the carrier frequency, is Tcmin to Tcmax, the number of carrier frequency values used for the change of the carrier frequency is N, and the maximum frequency of the desired frequency band is If Fmax, the variable range of the carrier frequency is
Fmax ≦ (N-1) / (Tcmax-Tcmin)
Therefore, it is possible to set a variable range of the carrier frequency in consideration of the number of carrier frequency values, flatten the EMI spectrum in a desired frequency band, and suppress the influence of EMI noise. There is an effect that can be.

  In addition, according to the present embodiment, the number of carrier frequencies used when changing the carrier frequency depends on the range of changes in the carrier frequency when changing the carrier frequency over time and the number of carrier frequencies used. Since the band in which the harmonics of the frequency are concentrated is determined so as not to fall within the desired frequency band, the EMI spectrum in the desired frequency band can be flattened and the effect of EMI noise can be suppressed. .

Further, according to the present embodiment, when the change range of the carrier cycle, which is the reciprocal of the carrier frequency, is Tcmin to Tcmax, and the maximum frequency of the desired band is Fmax, the number N of carrier frequency values used for the change of the carrier frequency is ,
Fmax (Tcmax-Tcmin) +1 ≦ N
Therefore, the number of carrier frequency values that can effectively suppress the influence of EMI noise can be calculated.

1 Power conversion system 2 PWM inverter (opening / closing means)
3 Three-phase brushless motor 4 Current sensor 5 Control device (control means)
6 Current command value generation unit 7a, 7b, 7c PID control unit 8a, 8b, 8c PWM comparison unit (PWM comparison means)
9a, 9b, 9c Signal inversion unit 10 Carrier frequency changing unit (carrier frequency changing means)
11 Carrier output section (carrier output means)

Claims (11)

A power conversion device comprising an opening / closing means for converting input power into a desired form, and a control means for performing PWM control of the opening / closing operation of the opening / closing means,
The control means includes
Carrier wave frequency changing means for discretely and periodically changing the frequency of the carrier wave used for the PWM control;
Carrier wave output means for outputting the carrier wave to the PWM comparison means;
With
The carrier frequency changing means determines a variable range of the carrier frequency so that a spread spectrum of harmonics of each order of the carrier wave is continuous within a desired frequency band.   The carrier frequency changing means is configured such that, in a desired frequency band, the harmonic frequency of an arbitrary order of the upper limit frequency of the carrier wave is equal to or lower than the harmonic frequency of an order higher than the order of the lower limit frequency of the carrier wave. The power conversion device according to claim 1, wherein a variable range of the carrier frequency is determined. The variable range of the carrier frequency is fcc-Δfc to fcc + Δfc, n is the maximum integer satisfying n ≦ Fmin / fcc, and Fmin is the lowest frequency of the desired frequency band, the variable range of the carrier frequency is
fcc / (2n + 1) ≦ Δfc
The power conversion device according to claim 2, wherein the relationship is satisfied.   In the desired frequency band, the carrier frequency changing means is configured such that the harmonic frequency of an arbitrary order of the upper limit frequency of the carrier wave is equal to or lower than the harmonic frequency of the second order higher than the order of the lower limit frequency of the carrier wave. The power conversion device according to claim 1, wherein a variable range of the carrier frequency is determined. The variable range of the carrier frequency is fcc-Δfc to fcc + Δfc, n is the maximum integer satisfying n ≦ Fmin / fcc, and Fmin is the lowest frequency of the desired frequency band, the variable range of the carrier frequency is
2fcc / (2n + 1) ≤Δfc
The power conversion device according to claim 4, wherein the relationship is satisfied.   The variable range of the carrier frequency includes a range in which the carrier frequency changes when the carrier frequency is changed over time and a band in which harmonics of the carrier frequency due to the number of carrier frequencies used are concentrated in a desired frequency band. The power conversion device according to claim 1, wherein the power conversion device is determined so as not to exist.   The variable range of the carrier frequency is the maximum frequency of the desired frequency band, where the carrier frequency change range when the carrier frequency is changed over time and the band where the harmonics of the carrier frequency due to the number of carrier frequencies used are concentrated. The power conversion device according to claim 6, wherein the power conversion device is determined so as to be larger. When the carrier frequency change range, which is the reciprocal of the carrier frequency, is Tcmin to Tcmax, the number of carrier frequency values used to change the carrier frequency is N, and the maximum frequency of the desired frequency band is Fmax, the carrier frequency The variable range is
Fmax ≦ (N-1) / (Tcmax-Tcmin)
The power conversion device according to claim 7, wherein the relationship is satisfied.   The number of carrier frequencies used when changing the carrier frequency is such that the range of changes in the carrier frequency when changing the carrier frequency over time and the band where the harmonics of the carrier frequency due to the number of carrier frequencies used are concentrated. The power conversion device according to claim 1, wherein the power conversion device is determined so as not to fall within a desired frequency band. When the carrier frequency change range, which is the reciprocal of the carrier frequency, is Tcmin to Tcmax, and the maximum frequency of the desired band is Fmax, the number N of carrier frequency values used for the change of the carrier frequency is:
Fmax (Tcmax-Tcmin) +1 ≦ N
The power conversion device according to claim 9, wherein: A power conversion method for converting input power into a desired form by opening and closing operations,
Within the desired frequency band, determine the frequency variable range of the carrier so that the spread spectrum of the harmonics of each order of the carrier is continuous,
Discretely and periodically changing the frequency of the carrier wave,
Generating a control signal for controlling the opening and closing operation based on the carrier wave;
A power conversion method, wherein the opening / closing operation is controlled based on the control signal.

Images