Harmonic suppression method, device and storage medium for silicon carbide semiconductor

Abstract

The application discloses a harmonic suppression method and device for silicon carbide semiconductors and a storage medium, belongs to the technical field of semiconductor design, and comprises the following steps: obtaining random numbers under a target number through a two-state Markov chain random number generator; performing asymmetric processing on the action time of a basic vector pair under the target logarithm through the random numbers in a random pulse position harmonic suppression method; performing adaptive improvement on a triangular carrier signal in a PWM waveform for the basic vector pair after the asymmetric processing; and performing modulation judgment on the opening and closing states of each silicon carbide semiconductor in a three-phase inverter bridge according to the improved triangular carrier signal, and obtaining a required asymmetric pulse signal. The asymmetric random pulse signal disperses high-frequency harmonics concentrated around the switching frequency and multiples to a greater extent, achieves the purposes of random zero vector step and random switching frequency, and reduces harmonics in a low-frequency operation state of a motor and EMI noise in a high-frequency operation state.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of semiconductor technology, and in particular to a harmonic suppression method, device and storage medium for silicon carbide semiconductor. BACKGROUND

[0002] With the continuous development of power electronics technology, SiC MOSFET as a new generation of high-performance power devices has begun to gradually enter various fields due to its high switching speed, low switching loss and other characteristics, especially in high-speed motor drive control. With the continuous development of manufacturing process, the development of traditional Si-based IGBT has been mature, and the problems of low switching speed and large switching loss in Si-based IGBT cannot meet the requirements of high-performance drive systems. Under such strong demand, SiC-based MOSFET is developed for motor drive systems due to its low loss and high-speed switching performance. Although the permanent magnet synchronous motor driver based on SiC MOSFET can achieve high efficiency and power density, due to the characteristics of SiC MOSFET which are different from Si devices, its application in permanent magnet synchronous motor drivers still has some problems. Since SiC MOSFET is often used in high-speed motor drivers, its switching frequency is high, and in addition, the parasitic parameters of SiC MOSFET are small, which may produce large high-frequency harmonics and electromagnetic interference (EMI) during high-speed switching, causing oscillation of motor current and torque, and affecting motor performance. The general motor control strategy adopts a fixed switching frequency PWM control strategy, which will produce high-frequency harmonics near the switching frequency and its multiples, and exacerbate EMI noise. SUMMARY

[0003] In order to reduce the frequency harmonics in the low-frequency state and the EMI noise in the high-frequency state of the motor with SiC MOSFET during operation, the present application improves the harmonic suppression method based on the random pulse position harmonic suppression method SVPWM, and proposes a harmonic suppression method for silicon carbide semiconductor, comprising the steps of:

[0004] S1: based on the uniform distribution random number generation method, a two-state Markov chain random number generator is built under the idea of linear congruential method;

[0005] S2: obtaining random numbers under a target number through the two-state Markov chain random number generator;

[0006] S3: performing asymmetric processing on the action time of the basic vector pair under the target logarithm in the random pulse position harmonic suppression method through the random numbers;

[0007] S4: Adaptive improvement of the triangular carrier signal in the PWM waveform for the asymmetrically processed basic vector pair;

[0008] S5: Modulation determination of the on-off state of each silicon carbide semiconductor in the three-phase inverter bridge according to the improved triangular carrier signal, and obtaining the required asymmetric pulse signal.

[0009] Further, in the S1 step, the idea of linear congruential method is expressed by the following recursive formula:

[0010]

[0011] In the formula, x n+1 is the nth+1 random number generated, x n is the nth random number generated, a is the multiplier, c is the increment, and m is the modulus.

[0012] Further, the target number and the target logarithm have the same numerical value.

[0013] Further, the three-phase inverter bridge contains three pairs of upper and lower bridge arms composed of six silicon carbide semiconductors, and eight kinds of on-off states are obtained according to the on-off state switching of each upper and lower bridge arm pair. The basic vectors represented by these eight kinds of on-off states constitute the first to sixth sectors of the step voltage vector.

[0014] Further, according to the principle of changing only one bridge arm at a time, the action sequence of the basic vectors when synthesizing the reference voltage vector in any sector is as follows,

[0015] V0→V m →V n →V7→V n →V m →V0

[0016] In the formula, V0 and V7 are zero vectors represented by the on-off state of each upper and lower bridge arm pair with only the lower bridge arm or only the upper bridge arm turned on, V m and V n are the basic vectors constituting the current sector in the counterclockwise direction.

[0017] Further, in the S3 step, in the random number pair random pulse position harmonic suppression method, the action time of the basic vector pair under the target number is symmetrically and equally arranged before asymmetric processing. The asymmetric processing is specifically as follows:

[0018] Under the condition of keeping the overall PWM period unchanged, the symmetrically arranged basic vector pair is multiplied by the random number.

[0019] Further, in the S4 step, the adaptive improvement of the triangular carrier signal specifically includes:

[0020] The same isosceles right triangle wave as the PWM cycle is used as a carrier wave, and a square wave with the same cycle as the triangle wave is introduced, and the signal amplitude of the square wave is greater than the maximum signal amplitude of the triangle wave.

[0021] Further, the modulation judgment in the S5 step specifically comprises the following judgment steps:

[0022] A0: whether the current square wave amplitude is greater than the triangle wave amplitude, if yes, go to step A1, if no, go to step A2;

[0023] A1: whether the normalized triangle wave amplitude is less than the low level time of the upper half cycle of the triangle wave, if yes, control the upper and lower bridge arms corresponding to the current sector to keep the first opening and closing state and go to step A3, if no, control the upper and lower bridge arms corresponding to the current sector to keep the second opening and closing state and go to step A3;

[0024] A2: whether the normalized triangle wave amplitude is less than the low level time of the lower half cycle of the triangle wave, if yes, control the upper and lower bridge arms corresponding to the current sector to keep the first opening and closing state and go to step A3, if no, control the upper and lower bridge arms corresponding to the current sector to keep the second opening and closing state and go to step A3;

[0025] A3: whether the sampling time reaches a PWM cycle, if yes, end, if no, go to the next sampling cycle and return to step A0.

[0026] Further, the application also provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to realize the steps of the harmonic suppression method.

[0027] Further, the application also provides a data processing device, which comprises:

[0028] a storage which stores a computer program;

[0029] a processor which is used to execute the computer program in the storage to realize the steps of the harmonic suppression method.

[0030] Compared with the prior art, the application has at least the following beneficial effects:

[0031] (1) The harmonic suppression method, device and storage medium for silicon carbide semiconductor can control the action time of the symmetric basic vector in a PWM cycle through a random variable, generate a non-symmetric random pulse signal, and more widely disperse the high frequency harmonics concentrated around the switching frequency and multiples, so as to achieve the purposes of random zero vector step and random switching frequency, and further reduce the harmonics in the low frequency running state of the motor and the EMI noise in the high frequency running state.

[0032] (2) In the idea of linear congruential method, the Markov chain random number generator is built, so that the generated random numbers are more evenly distributed on both sides of the expected value;

[0033] (3) The signal is improved by taking the isosceles right triangle wave as a carrier and incorporating the square wave, so that the waveform conforms to the asymmetric pulse modulation logic without increasing the additional time for amplitude setting and calculation time. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 It is a step diagram of a harmonic suppression method for a silicon carbide semiconductor;

[0035] Figure 2 It is a basic vector distribution and voltage vector synthesis schematic diagram;

[0036] Figure 3 It is a pulse signal and vector action time schematic diagram of traditional SVPWM and RPPSVPWM in a PWM period;

[0037] Figure 4 It is an adaptive improvement schematic diagram of a triangular carrier signal;

[0038] Figure 5 It is an on-off state judgment schematic diagram;

[0039] Figure 6 It is a two-state Markov chain principle schematic diagram; DETAILED DESCRIPTION

[0040] The following is a specific embodiment of the present application and further describes the technical solutions of the present application in conjunction with the drawings, but the present application is not limited to these embodiments.

[0041] Example 1

[0042] Pulse width modulation (PWM) is a main technology of electric energy conversion, and two-level three-phase space voltage vector pulse width modulation (SVPWM) has the advantages of simple structure, easy implementation, etc., and is widely used in motor control. There are many pulse width modulation methods derived from SVPWM, including but not limited to: discontinuous pulse width modulation (DPWM), zero vector pulse width modulation (NZPWM), random PWM (RPWM), etc. In pulse width modulation, zero vectors can be used to maintain the original state, and although the lack of zero vectors can suppress the common-mode voltage on the output side, it will make the inverter load current output value more deviate from the reference value, and the waveform quality will decrease and the harmonics will increase.

[0043] The main idea of SVPWM is to change the opening and closing mode of the inverter to make the actual flux vector formed by the inverter track the ideal stator flux circle in the three-phase permanent magnet synchronous motor. Because SVPWM modulation strategy only needs to switch one device each time, the switching loss can be reduced. The core of SVPWM modulation is to switch the opening and closing mode of the inverter, and the upper and lower bridge arms of the three-phase inverter bridge are SiC MOSFET modules, which cannot be turned on at the same time, otherwise a short circuit will occur, causing damage to the components. Then, by arranging and combining the opening and closing control of the six power devices of the three-phase inverter bridge, eight different opening and closing states (divided into U0-U7) can be obtained, that is, 100, 110, 001, 010, 011, 101, 111, 000 (wherein, the upper bridge arm is turned on and the lower bridge arm is turned off as 1; the upper bridge arm is turned off and the lower bridge arm is turned on as 0, and 111 is an example that three pairs of upper and lower bridge arms are in the state of upper bridge arm turned on and lower bridge arm turned off). Among them, the first six represent the basic quality of the effective state (hereinafter referred to as V1 to V6), and the last two are zero vector states (hereinafter referred to as V0 and V7), and the basic vector distribution and voltage vector synthesis principle are shown in the sector diagram (including I sector to VI sector) as shown in Figure 2 According to the area equivalence principle, when the output voltage vector V ref is located in a certain sector, the output voltage vector value in a sampling period can be synthesized by the two adjacent non-zero voltage vectors and two zero vectors of this sector, wherein the action time of the two non-zero voltage vectors is T1 and T2 respectively, and the action time of the two zero vectors is T0.

[0044] Since the harmonic signal is generally generated by pulse width modulation, Fourier transform of the harmonic current can be performed to see that the fundamental current and the harmonic current are closely related to the switching frequency of the system. The traditional SVPWM control strategy is to synthesize the reference voltage vector with the basic vector in a switching period, and according to the volt-second balance principle, the action time of the vector participating in the synthesis is:

[0045]

[0046] In the formula, V ref is the reference voltage vector, V dc is the bus voltage, T s is the switching period, is the deflection angle of the reference voltage vector.

[0047] From the above calculation, it can be seen that when the bus voltage is constant, the vector action time is related to the switching period. The switching frequency of the traditional SVPWM is a fixed value, so the harmonics will be concentrated at the switching frequency and its multiples, resulting in an increase in harmonic content and the generation of EMI.

[0048] For the problems of the SiC MOSFET applied in the pulse width modulation process, the application proposes a harmonic suppression method for silicon carbide semiconductors, which is improved on the basis of the random pulse position harmonic suppression method (i.e. SVPWM), and specifically includes the following steps: Figure 1

[0049] S1: based on the uniform distribution random number generation method, a two-state Markov chain random number generator is built under the idea of linear congruential method;

[0050] S2: the random number under the target quantity is obtained through the two-state Markov chain random number generator;

[0051] S3: the asymmetric processing of the action time of the basic vector pair under the target logarithm in the random pulse position harmonic suppression method is carried out through the random number;

[0052] S4: the adaptive improvement of the triangular carrier signal in the PWM waveform is carried out for the basic vector pair after the asymmetric processing;

[0053] S5: according to the improved triangular carrier signal, the modulation judgment of the on-off state of each silicon carbide semiconductor in the three-phase inverter bridge is carried out, and the required asymmetric pulse signal is obtained.

[0054] The random pulse position harmonic suppression method refers to randomly changing a certain quantity in a switching cycle to change the PWM waveform, and then improving the quality of the output current. According to the different random change quantities, it can be divided into random switching pulse width modulation, random carrier pulse width modulation and random pulse position pulse width modulation. The random switching pulse width modulation method has heavy distortion of the modulation signal when the modulation degree is low; the random carrier pulse width modulation method makes the sampling frequency and the switching frequency out of synchronization, thereby affecting the PID regulation effect; and the random pulse position pulse width modulation (hereinafter referred to as RPPSVPWM) is simple to operate, has strong practicability, and can solve the problem of fixed switching frequency.

[0055] ​The traditional SVPWM algorithm adopts the method of "back-to-back" zero vector to arrange the basic vectors in order to ensure that only one bridge arm changes at each vector change. For example, in the first sector (which can also be other sectors, and here only the first sector is taken as an example), the action order of the basic vectors is "V0→V4→V6→V7→V6→V4→V0" when the synthesized reference voltage vector is generated, and the corresponding vector action time is symmetrical and equal. This embodiment continues to take the random number with a target number of 1 as an example (in fact, at most three random numbers can be set for pulse adjustment of V0, V4 and V6 according to the action order of the basic vectors, and this embodiment only takes V0 as an example for explanation and description), and adds the random number δ to the action time of the first V0 (that is, T0, and the action times of the other two basic vectors are T a and T b respectively) in the RPPSVPWM algorithm adopted by the present application, so that the pulse conduction time changes randomly, that is, the pulse position changes randomly in a PWM cycle, and then the switching frequency is randomized. Figure 3 Fig. 1 is a schematic diagram of the pulse signals and vector action time of the traditional SVPWM and RPPSVPWM in a PWM cycle.

[0056] Considering that the carrier wave used in general PWM technology is a triangular wave, and because the pulse signal in the traditional SVPWM is symmetrical, an isosceles right triangle carrier wave with the same PWM cycle is set, so that the amplitude of the triangular wave is equal to the time increment in the first half of the PWM cycle. Thus, the switching state can be changed by comparing the amplitude of the triangular wave and the on-off time of the switch, and the switching state in the second half of the PWM cycle is controlled according to the left-right symmetry. Then, it can be known from Figure 3 that when the random number δ is added, the pulse signal is in an asymmetrical form in a PWM cycle. In this case, the method of controlling the on-off time of the switch by left-right symmetry of the triangular wave signal in the traditional SVPWM cannot be applied to the generation of the asymmetrical PWM wave under the RPPSVPWM algorithm. In order to solve this problem, a method of using a sawtooth wave instead of a triangular wave as a carrier wave is proposed in the prior art, which changes the triangular wave that increases and decreases with time into a sawtooth wave that only increases with time, and then judges the switching state according to the traditional comparison logic. Although this method conforms to the logic of asymmetrical pulse modulation, it is found in actual experimental simulation that the amplitude setting of the sawtooth wave signal generated by the system is complex, which increases the calculation time and reduces the running efficiency of the system. As can be seen from the above analysis, the key to solving the problem of asymmetrical modulation is to avoid the conflict of the on-off control of the switch in the first and second halves of the PWM cycle. Therefore, the present application proposes a method of combining square waves and triangular waves to solve this problem. For example, Figure 4As shown, the present application continues to adopt the same isosceles right triangle wave as a carrier with a PWM cycle, and introduces a square wave signal with the same cycle as the triangle wave, and the amplitude of the square wave signal needs to be greater than the maximum amplitude of the triangle wave, so as to divide the triangle wave into two parts with equal time on the left and right. Since there is only one random variable, the on-off time of the switch must exist on both sides of the half cycle, so the triangle wave amplitude and the on-off time of the switch on both sides are compared respectively to determine the change of the open-close state, and finally generate the required asymmetric pulse signal. The specific determination steps are as follows (such as Figure 5 ):

[0057] A0: Determine whether the current square wave amplitude is greater than the triangle wave amplitude, if yes, go to step A1, if no, go to step A2;

[0058] A1: Determine whether the normalized triangle wave amplitude is less than the low level time of the upper half cycle of the triangle wave, if yes, control the upper and lower bridge arms corresponding to the current sector to keep the first open-close state and go to step A3, if no, control the upper and lower bridge arms corresponding to the current sector to keep the second open-close state and go to step A3;

[0059] A2: Determine whether the normalized triangle wave amplitude is less than the low level time of the lower half cycle of the triangle wave, if yes, control the upper and lower bridge arms corresponding to the current sector to keep the first open-close state and go to step A3, if no, control the upper and lower bridge arms corresponding to the current sector to keep the second open-close state and go to step A3;

[0060] A3: Determine whether the sampling time reaches a PWM cycle, if yes, end, if no, go to the next sampling cycle and return to step A0.

[0061] Wherein, since the principle of changing only one bridge arm at a time, the open-close state actually includes two kinds of upper bridge arm conduction, lower bridge arm off, and upper bridge arm off, lower bridge arm conduction.

[0062] Further, considering that the quality of random numbers is one of the important factors to determine the pros and cons of random PWM modulation strategy, since the generation of true random numbers is relatively complex, pseudo random numbers can be used instead of real random numbers. The generation methods of random numbers mainly include product taking method, linear congruential method, logic method, table lookup method, etc. Among them, the value generated by the table lookup method is set in advance, the random number generated by the logic method is related to the initial value, and the product taking method will cause the result to deviate. Therefore, the linear congruential method is selected as the random number generation scheme in this embodiment, which is relatively simple in operation and small in error, and its recursive formula is as follows:

[0063]

[0064] In the formula, x n+1 is the nth+1 random number generated, xn For the generated nth random number, a is a multiplier, c is an increment, and m is a modulus.

[0065] However, the Markov chain is a discrete random process, which represents the probability of the next state given the current state only related to the current state, and has nothing to do with the past state, that is, the Markov chain has no aftereffect characteristics. This characteristic can make the generated random number more uniformly distributed on both sides of the expected value, so that the harmonic energy on the current voltage spectrum is more evenly dispersed. In order to ensure the quality of the generated random number and reduce the calculation pressure of the system, the two-state Markov chain is also used to provide random numbers in the application. The basic principle is as shown in the figure, wherein state 1 and state 2 respectively represent that the sampling period is greater than or less than the expected period, P represents the probability in the state, and the randomization process is to constantly transfer between state 1 and state 2, thereby generating more uniform random numbers. Figure 6

[0066] The application combines linear congruence and Markov chain to generate uniformly distributed random numbers. Compared with ordinary uniformly distributed random numbers, which are mostly distributed on the same side of the expected value, the random numbers generated based on the Markov chain can be more uniformly distributed on both sides of the expected value, which is conducive to reducing harmonic distortion and thus strengthening the suppression of harmonics by the random pulse modulation strategy.

[0067] At the same time, according to the modulation strategy of the traditional SVPWM, the size of the reference voltage vector is determined by the action time of the non-zero vector involved in the synthesis. Therefore, in order to avoid the problem that the value of δ is too large to cause the action time of the non-zero vector to be insufficient within a PWM period, it is necessary to determine the value range of δ. At the same time, in order to ensure that the action time of the non-zero vector is completely within a PWM period, it is necessary to satisfy:

[0068]

[0069] From the above formula, it can be seen that δ∈[0,1], when δ=0.5, the pulse width modulation becomes the traditional symmetrical pulse width modulation. In the random modulation strategy, the generation range of the random number also affects the harmonic suppression, so the harmonic content under different random ranges is also different, which needs to be set according to the actual demand.

[0070] In summary, the harmonic suppression method, device and storage medium for silicon carbide semiconductor described in the application control the action time of the symmetrical basic vector within a PWM period through a random variable, generate an asymmetric random pulse signal, and disperse the high-frequency harmonics concentrated around the switching frequency and multiples to the greatest extent, so as to achieve the purpose of random zero vector step and random switching frequency, thereby reducing the harmonics in the low-frequency running state of the motor and the EMI noise in the high-frequency running state. ​

[0071] The Markov chain random number generator is built in the idea of linear congruential method, so that the generated random numbers are more uniformly distributed on both sides of the expected value.

[0072] The signal is improved by taking the isosceles right-angle triangular wave as a carrier and incorporating the square wave, so that the waveform conforms to the asymmetric pulse modulation logic without increasing additional time for amplitude setting and calculation time.

[0073] It should be noted that all directional indications (such as upper, lower, left, right, front, rear, etc.) in the embodiments of the present application are only used to explain the relative position relationship, movement condition, etc. between components in a certain specific posture (as shown in the drawings), and if the specific posture changes, the directional indications will also change accordingly.

[0074] In addition, the description such as "first", "second", "one" and the like in the present application is only for the purpose of description, and cannot be understood as indicating or implying the relative importance of the technical features indicated or implicitly indicating the number of technical features. Therefore, the features defined as "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise specifically limited.

[0075] In the present application, unless otherwise specifically defined and limited, the terms "connection", "fixing" and the like should be understood in a broad sense, for example, "fixing" can be fixed connection, or detachable connection, or integral; can be mechanical connection, or electrical connection; can be directly connected, or indirectly connected through intermediate medium; can be internal communication of two elements or interaction relationship between two elements, unless otherwise specifically limited. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0076] In addition, the technical solutions of each embodiment of the present application can be combined with each other, but it must be based on the realization of the ordinary skilled in the art, when the combination of technical solutions appears contradictory or unachievable, it should be considered that the combination of technical solutions does not exist, nor within the scope of protection required by the present application.

Claims

1. A method for harmonic suppression of a silicon carbide semiconductor, characterized by, The improvement is specifically comprised of steps of: S1: based on the uniform distribution random number generation method, a two-state Markov chain random number generator is built under the idea of linear congruential method; S2: random numbers in the target quantity are obtained through the two-state Markov chain random number generator; S3: the random numbers are used to perform asymmetric processing on the action time of the basic vector pair in the target logarithm in the random pulse position harmonic suppression method; S4: the asymmetrically processed basic vector pair is subjected to adaptive improvement of the triangular carrier signal in the PWM waveform; S5: modulation judgment of the on-off state of each silicon carbide semiconductor in the three-phase inverter bridge is performed according to the improved triangular carrier signal, and the required asymmetric pulse signal is obtained. In the S4 step, the adaptive improvement of the triangular carrier signal specifically comprises: The same isosceles right triangle wave as the PWM period is used as the carrier, and a square wave with the same period as the triangular wave is introduced, and the signal amplitude of the square wave is greater than the maximum signal amplitude of the triangular wave. In the S5 step, the modulation judgment specifically comprises the following judgment steps: A0: whether the current square wave amplitude is greater than the triangular wave amplitude is judged, if yes, step A1 is entered, if not, step A2 is entered; A1: whether the standardized triangular wave amplitude is less than the low level time of the upper half cycle of the triangular wave is judged, if yes, the first on-off state of the upper and lower bridge arms corresponding to the current sector is maintained and step A3 is entered, if not, the second on-off state of the upper and lower bridge arms corresponding to the current sector is maintained and step A3 is entered; A2: whether the standardized triangular wave amplitude is less than the low level time of the lower half cycle of the triangular wave is judged, if yes, the first on-off state of the upper and lower bridge arms corresponding to the current sector is maintained and step A3 is entered, if not, the second on-off state of the upper and lower bridge arms corresponding to the current sector is maintained and step A3 is entered; A3: whether the sampling time reaches a PWM period is judged, if yes, the process is ended, if not, the next sampling period is entered and step A0 is returned.

2. A method for harmonic suppression of a silicon carbide semiconductor as defined in claim 1, wherein In the S1 step, the idea of linear congruential method is expressed by the following recursive formula: ； In the formula is the (n+1)th random number generated, is the nth random number generated, is the multiplier, c is the increment, m is the modulus, and mod is the modulus operation.

3. A method for harmonic suppression of a silicon carbide semiconductor as defined in claim 1, wherein The target quantity and the target logarithm have the same value.

4. The method of claim 1, wherein the harmonic suppression method is used for a silicon carbide semiconductor. The three-phase inverter bridge contains three pairs of upper and lower bridge arms composed of six silicon carbide semiconductors, and eight kinds of on-off states are obtained according to the on-off state switching of each pair of upper and lower bridge arms, and the basic vectors represented by the eight kinds of on-off states constitute the first sector to the sixth sector of the step voltage vector.

5. A method for harmonic suppression of a silicon carbide semiconductor as defined in claim 4, wherein According to the principle that each vector change only changes one bridge arm, the action order of the basic vectors is as follows when synthesizing the reference voltage vector in any sector. ； In the formula, and are zero vectors respectively represented in the on-off state of each upper and lower bridge arm pair only turning on the lower bridge arm and only turning on the upper bridge arm, and are basic vectors constituting the current sector in the counterclockwise direction.

6. A method for harmonic suppression of a silicon carbide semiconductor as defined in claim 5, wherein In the S3 step, the action time of the basic vector pair in the target logarithm in the random pulse position harmonic suppression method is symmetrically arranged before the asymmetric processing, and the asymmetric processing is specifically as follows: Under the condition that the overall PWM period remains unchanged, the symmetrically arranged basic vector pair is multiplied by the random number.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that The program is executed by the processor to implement the steps of the harmonic suppression method in any one of claims 1 to 6.

8. An apparatus for processing data, the apparatus comprising: It comprises: a storage having a computer program stored thereon; a processor for executing a computer program in said storage to implement the steps of the harmonic suppression method of any one of claims 1 to 6.

Images