A Spread Spectrum Modulation Method for Inverter Based on Secondary Frequency Modulation

Abstract

The present invention provides an inverter spread spectrum modulation method based on secondary frequency modulation. By establishing an equivalent evaluation model of input voltage ripple, output current ripple, and inverter switching loss under the spread spectrum modulation strategy, the influence of different parameters of the spread spectrum modulation strategy on these indicators is analyzed. On this basis, based on the characteristics of the ripple frequency distribution under periodic signal spread spectrum modulation, a "secondary frequency modulation" strategy is proposed to reduce the inverter conducted EMI to a greater extent without increasing the inverter voltage and current ripple and loss, which can achieve the dispersion of EMI peak energy in a specific frequency band.

Description

Technical Field

[0001] The invention belongs to the technical field of power electronic modulation, and in particular relates to an inverter spread spectrum modulation method based on secondary frequency modulation. Background Art

[0002] As a third-generation semiconductor material, SiC (Silicon Carbide) has excellent properties such as high temperature resistance, high voltage resistance, high switching frequency, and low loss, making it used in inverters. However, the high-frequency switching characteristics of SiC MOSFET (Metal Oxide Semiconductor Field Effect Transistor) will deteriorate the electromagnetic compatibility characteristics of the inverter. It is of great research significance to fully utilize the advantages brought by SiC's high-frequency characteristics while suppressing its negative effects to improve the comprehensive performance of the inverter.

[0003] Spread spectrum modulation technology is widely used to reduce the conducted electromagnetic interference (EMI) of motor controllers because of its characteristic of "dispersing energy spikes". Its principle is to change the carrier so that the signal energy originally concentrated in the system is dispersed to a wider frequency range. According to Parseval's theorem, if the energy distribution of the signal in the time domain remains unchanged, the energy of the signal in the frequency domain will also remain unchanged. Therefore, under the premise of ensuring that the signal energy in the frequency domain remains unchanged, spread spectrum modulation achieves the beneficial effect of reducing signal spikes in a wider frequency band by expanding the bandwidth of the signal energy distribution. Common spread spectrum modulation strategies can be divided into two categories: periodic PWM (pulse width modulation) and random PWM. For example, in the literature [Qi Chen, Chen Xiyou, Mou Xianmin. Hybrid spread spectrum modulation technology for PWM inverters [J]. Proceedings of the CSEE, 2012, 32(24): 38-44, 8], they are implemented by injecting periodic signals or random signals into the carrier. Common periodic signals include sinusoidal signals, triangular signals, sawtooth signals, etc. Random signals are usually replaced by chaotic signals. Common chaotic sequences include Logistic sequences, Cubic sequences, etc. Among these two methods, the periodic PWM principle is relatively simple, has little impact on the dynamic performance of the speed regulation system, and is easy to implement in engineering. It has become an important solution to improve the inverter conducted EMI problem.

[0004] Existing research on EMI reduction using spread spectrum modulation technology, such as the literature [Yuan Qingbing. Research on period frequency modulation strategy of permanent magnet synchronous motor system [D]. Harbin Institute of Technology, 2016], mainly focuses on analyzing the impact of different modulation parameters on the spread spectrum effect, such as the impact of different types of random variables, the frequency of periodic signals and the maximum deviation of carrier frequency on the spread spectrum effect. However, in addition to conducted EMI, the indicators affected by the modulation strategy also include input side DC voltage ripple, output current ripple, loss, etc.; for these indicators, there is currently a lack of theoretical derivation of the degree of impact on these indicators under spread spectrum modulation. Existing research usually directly tests the ripple and loss under different spread spectrum modulation strategies through experiments, which is used for comprehensive evaluation and analysis of modulation strategies. Summary of the invention

[0005] In view of the above, the present invention provides an inverter spread spectrum modulation method based on secondary frequency modulation, which can qualitatively analyze the degree to which different indicators are affected by the spread spectrum modulation strategy, and can minimize the conducted EMI spikes without affecting indicators such as ripple and loss.

[0006] An inverter spread spectrum modulation method based on secondary frequency modulation comprises the following steps:

[0007] (1) Determine the maximum deviation of the carrier frequency during spread spectrum modulation of a single periodic signal Δ f ;

[0008] (2) Based on the carrier frequency distribution characteristics of different periodic signals (such as sine wave, triangle wave, sawtooth wave) during spread spectrum modulation, the sine signal is selected as the periodic signal during spread spectrum modulation. v m ( t );

[0009] (3) v m ( t ) is a sinusoidal signal, its carrier frequency is dispersed in a specific frequency band and the carrier period is determined;

[0010] (4) Generate a carrier wave and a modulation wave according to the duty cycle and carrier period of the basic voltage vector of each sector, and compare the carrier wave with the modulation wave to generate a PWM signal for the inverter.

[0011] Furthermore, in step (1), the maximum carrier frequency deviation Δ f The value of must meet the following conditions:

[0012]

[0013] in: f s0 is the carrier center frequency (given quantity), fsmin and f smax are the lower and upper limits of the carrier frequency.

[0014] Furthermore, the lower limit of the carrier frequency f smin and upper limit f smax The quantitative relationship between the input voltage ripple, output current ripple, inverter switching loss and each spread spectrum parameter of the inverter under the spread spectrum modulation strategy (conventional) is established, and then calculated and determined according to the limit requirements of input voltage ripple and output current ripple to limit f s1 and f s2 The value of: f smin ≤ f s1 , f s2 ≤ f smax , f s1 and f s2 They are respectively the upper and lower limits of the carrier frequency of the final spread spectrum modulation strategy after secondary frequency modulation (given, that is, finally determined through experimental debugging).

[0015] Furthermore, the specific implementation method of dispersing the carrier frequency in a specific frequency band in step (3) is as follows:

[0016] When 0≤ t <0.15 T m , 0.35 T m ≤ t <0.65 T m , 0.85 T m ≤ t ≤ T m When , the expression of carrier frequency is as follows:

[0017] f s ( t )= f s0 +Δ f m ( t )

[0018] v m ( t)=sin(2π f m t )

[0019] in: f s ( t )express t The carrier frequency at the time, f s0 is the carrier center frequency, f m For periodic signals v m ( t ) frequency (given quantity), T m =1 / f m , t Indicates time;

[0020] When 0.15 T m ≤ t <0.25 T m , 0.25 T m ≤ t <0.35 T m , 0.65 T m ≤ t <0.75 T m , 0.75 T m ≤ t <0.85 T m When the carrier frequency is a linear function, f s ( t )= kt + b , k and b are the slope and intercept of a linear function respectively.

[0021] At 0.15 T m ≤ t <0.25 T m hour, k =( f s1max - f s1min ) / 0.1 T m , b=2.5 f s1min -1.5 f s1max

[0022] At 0.25 T m ≤ t <0.35 T m hour, k =( f s1min - f s1max ) / 0.1 T m , b =3.5 f s1max -2.5 f s1min

[0023] At 0.65 T m ≤ t <0.75 T m hour, k =( f s2min - f s2max ) / 0.1 T m , b =7.5 f s2max -6.5 f s2min

[0024] At 0.75 T m ≤ t <0.85 T m hour, k =( f s2max - f s2min ) / 0.1 T m , b =8.5 f s2min -7.5 f s2max

[0025] in: f s1min = f s0 +0.8Δ f ,f s2max = f s0 -0.8Δ f , f s1max = f s1 , f s2min = f s2 , f s1 and f s2 They are respectively the upper and lower limits of the carrier frequency of the final spread spectrum modulation strategy after secondary frequency modulation.

[0026] Furthermore, in step (3), for the next carrier cycle, it is equal to the inverse of the carrier frequency at the last moment of the previous carrier cycle.

[0027] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor is used to execute the computer program to implement the inverter spread spectrum modulation method based on secondary frequency modulation.

[0028] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the inverter spread spectrum modulation method based on secondary frequency modulation is implemented.

[0029] Compared with the prior art, the present invention has the following beneficial technical effects:

[0030] 1. Existing research on the impact of the distribution characteristics of periodic signals on EMI spectrum is still insufficient. This invention designs a modulation strategy for specific periodic signals based on their distribution characteristics, which can reduce the conducted EMI spikes to a greater extent within a limited carrier frequency deviation range.

[0031] 2. In addition to the beneficial effect of reducing conducted EMI spikes, existing studies lack theoretical analysis of other indicators (ripple, loss, etc.) under the spread spectrum modulation strategy. The present invention establishes an equivalent evaluation model for these indicators under the spread spectrum modulation strategy, which can theoretically evaluate the degree to which these indicators are affected by the modulation strategy.

[0032] 3. Most of the existing periodic spread spectrum modulation strategies study the spread spectrum effect of a single periodic signal. The present invention makes specific improvements on periodic signals and can disperse the EMI peak energy in a specific frequency band. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a schematic diagram of the implementation process of traditional periodic spread spectrum modulation, where u aref ,u bref , u cref For a given three-phase reference voltage, u αref , u βref is the reference voltage after Clarke transformation, U 0~ U 7 are the eight basic vectors of SVPWM, among which U 1~ U 6 is six non-zero basis vectors with magnitude 2 U dc / 3, U dc is the inverter DC bus voltage, I~VI are the numbers of the six sectors, U ref is the reference voltage vector, θ for U ref The angle with the starting coordinate axis, T 0. T 4. T 6 is the action time of the corresponding basic vector, T s1 , T s2 , T s3 For different carrier period values; s a , s b , s c is the switching state of the three-phase bridge arm of the inverter, and its value is 1 or 0; t 0~ t 8 is a carrier cycle T s Different moments within.

[0034] Figure 2 It is a two-level inverter topology diagram containing LISN (Line Impedance Stabilization Network), where v EMI For the test results of conducted EMI, i L From the power supply end to the DC support capacitor C dc The current in the branch, i Cap is the current flowing through the capacitor, u Cap is the capacitor voltage, idc is the input current of the inverter DC side, S1~S6 are the six switch tubes of the inverter, i a , i b , i c is the phase current on the load side, Z L is the load impedance.

[0035] Figure 3 is a schematic diagram of the voltage error vector of the Ith sector, where θ 1 and θ 2 are the reference vectors U ref The angle between two adjacent non-zero basis vectors, U 4,err and U 6,err are non-zero vectors U 4 and U 6 When in action U ref The error vector, U 0,7,err Zero vector U 0 and U 7 When acting U ref The error vector between the dq coordinate axis U ref The direction coordinate axis is the positive direction of the d-axis, and the direction 90° ahead of the d-axis is the positive direction of the q-axis.

[0036] Figure 4 The duration of different frequency bands under the same frequency band width and the carrier frequency distribution characteristic diagram corresponding to different periodic signals, in which (a) is v m ( t ) is the duration of the sinusoidal signal spread spectrum modulation, (b) is the carrier frequency distribution characteristics corresponding to the sinusoidal wave signal, (c) is the carrier frequency distribution characteristics corresponding to the triangle wave signal, and (d) is the carrier frequency distribution characteristics corresponding to the sawtooth wave signal.

[0037] Figure 5 This is a schematic diagram of the secondary frequency modulation of the present invention, in which (a) is the spectrum dispersion change of the secondary frequency modulation strategy near one-fold frequency, (b) is the carrier frequency change law of the secondary frequency modulation, and (c) is the distribution change of each frequency band of the carrier frequency of the secondary frequency modulation based on a single periodic signal.

[0038] Figure 6 The EMI spectrum results under different spread spectrum modulation strategies. The corresponding test conditions in Figure (a) are fs0 =10kHz fixed carrier frequency, (b) corresponding test conditions are v m ( t ) Take a sinusoidal signal and Δ f =1kHz, (c) corresponds to the test condition: v m ( t ) Take the triangular signal and Δ f =1kHz, (d) corresponds to the test condition: v m ( t ) Take the sawtooth signal and Δ f =1kHz, (e) corresponds to the test condition of secondary frequency modulation and f s1 =13.9kHz, f s2 =6.5kHz.

[0039] Figure 7 The input DC voltage waveform, output AC current waveform, and carrier frequency variation diagram under the spread spectrum modulation strategy are shown in Figure 2. u cap represents the DC voltage, i a Indicates the A-phase AC current.

[0040] Figure 8 For Figure 7 The local enlarged diagram of the selected AC current peak, the carrier frequency of the local enlarged part in the figure is 7.1kHz, and the corresponding current ripple is 6.2A. DETAILED DESCRIPTION

[0041] In order to describe the present invention more specifically, the technical solution of the present invention is described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0042] The inverter spread spectrum modulation method based on secondary frequency modulation of the present invention comprises the following steps:

[0043] (1) Establish a quantitative model of input voltage ripple, output current ripple, and switching loss under the spread spectrum modulation strategy to determine Δ f , v m ( t ) on these three indicators.

[0044] 1.1 The calculation formula for input voltage ripple in each time period is as follows:

[0045]

[0046] in: ~ Indicates the input voltage ripple at different times, namely Figure 2 middle u cap The fluctuation amount, t 0~ t 8 time segments such as Figure 1 As shown, One carrier cycle T s DC input current within i dc The average value of ; i 1. i 2 is the DC side current corresponding to the non-zero basic vector action i dc , different sectors i 1 and i 2 As shown in Table 1, i a , i b , i c is the load side current.

[0047] Table 1: Different sectors i 1 and i 2

[0048]

[0049] Based on the above formula, different f s Next, a carrier cycle T s The maximum value of the input voltage fluctuation within is recorded as According to the above derivation process, under the premise that other parameters are fixed, Mainly affected by T s ) max The influence of is a monotonically increasing relationship, corresponding to f s ( t ) f s ( t ) min Decision. f s ( t )= f s0 +Δ f m ( t ), f s (t ) min = f s0 -Δ f , and f s ( t ) min For a monotonically decreasing relationship, determine the limit Under the premise of f s ( t ) f smin1 , while calculating different v m ( t )right degree of influence, choose the appropriate v m ( t ).

[0050] 1.2 Based on the stator flux fluctuation ( ψ rip_rms ) max As the output current ripple Δ i The equivalent evaluation index of the stator flux fluctuation in each time period is calculated as follows. The product of different error vectors and time is decomposed into the d and q axes, such as Figure 3 As shown, there are:

[0051]

[0052] Among them, let:

[0053]

[0054] The stator flux fluctuation components of the d and q axes can be calculated as:

[0055]

[0056] in: D , Q They represent the d-axis component and q-axis component of the stator flux respectively; taking the I sector as an example, D and Q The expressions in different time periods are d 0. d 1. d 2 and q 1. q 2. The remaining sectors D and Q The expression can be obtained in the same way, and the effective value of the magnetic flux fluctuation can be obtained .

[0057] Based on the above, ( ψ rip_rms ) max Also mainly by T smax Decide, that is, by f s ( t ) min (Right now f s0 -Δ f ) is determined; therefore, by fitting the simulation test data, the output current ripple Δ i and f s ( t )、( ψ rip_rms ) max The corresponding relationship table of f s ( t ) min and v m ( t ) for current ripple Δ i The influence of the current limit Δ i max Under the premise of v m ( t )and f smin2 .

[0058] 1.3 The relationship between single switching loss and current is fitted with a quadratic function. The number of carriers contained in unit time of different modulation strategies is used to equivalently evaluate the number of switches, thereby evaluating the fluctuation level of switching loss. The single switching loss of the inverter switch tube is:

[0059]

[0060] Among them: coefficient A 1. B 1. C 1 can be obtained by fitting the double pulse test of the switch tube of the inverter; in this embodiment A 1=1×10 -9 , B 1=1×10 -4 , C 1=0.0022.

[0061] Considering a carrier cycle T s In one carrier cycle, each of the six switches of the inverter is switched on once. Ts The loss inside can be expressed as:

[0062]

[0063] Substitute the coefficients A 1. B 1. C 1, we can get:

[0064]

[0065] The number of carriers in any modulation strategy within 1s is calculated through a computer program, so as to obtain the total number of switching times of the inverter switch tube and the total loss of the inverter. The difference in the number of carriers contained in 1s can be used to equivalently evaluate the difference in switching loss of different modulation strategies. The switching loss is mainly affected by f s ( t )The influence of mid-high frequency band, f s ( t ) has more high-frequency bands, more switching times per unit time, and greater losses; while in periodic signal spread spectrum modulation, f s ( t ) is continuously changing; based on this, in limiting the loss increase Δ E sw,loss Under the premise of the upper limit of f s ( t ) f smax .

[0066] (2) Based on the frequency distribution characteristics of the conducted EMI peak and the switching frequency f s exist[ f s0 -Δ f , f s0 +Δ f ]The distribution characteristics within the frequency band are highly similar, which can be used according to different types v m ( t ) The carrier frequency distribution characteristics corresponding to the spread spectrum modulation, select one of the periodic signals, and select the appropriate Δ f , and the single periodic signal spread spectrum modulation strategy is used as the basis of "secondary frequency modulation".

[0067] (3) Different analysis v m ( t ) The carrier frequency distribution state corresponding to the spread spectrum modulation, under the same Δf Under the condition of , the uneven frequency band under sinusoidal signal modulation is mainly concentrated on both sides of the entire frequency band, such as Figure 4 As shown in (a) and (b) in Figure 2. Figure 4 In (b), (c), and (d), except for the high-frequency and low-frequency bands on both sides, v m ( t ) is a sine signal, and the proportion of the remaining frequency bands is significantly lower than that of the triangular signal and the sawtooth signal; specifically, its spectrum peaks are mainly distributed in [ f s0 -Δ f , f s0 -0.8Δ f ]and[ f s0 +0.8Δ f , f s0 +Δ f ] frequency band.

[0068] (4) Considering the above analysis v m ( t ) will not have a significant impact on indicators such as ripple and loss. Therefore, under this premise, based on the study of the distribution characteristics of the carrier frequency under periodic signal modulation, sinusoidal signal modulation is selected as the basis of "secondary frequency modulation", and based on its distribution characteristics, its concentrated uneven frequency band is dispersed to the remaining frequency bands, such as Figure 5 As shown in (a) in .

[0069] (5) In order to make the spectrum diffusion as sufficient as possible and avoid overlapping with the original frequency band, the present invention chooses to f s0 -Δ f , f s0 -0.8Δ f ] frequency components in the frequency band are evenly spread to [ f s2 , f s0 -0.8Δ f ] frequency band, f s0 +0.8Δ f , f s0 +Δ f ]The frequency components in [ f s0 +0.8Δ f , f s1 ] frequency band, the distribution changes as Figure 5 As shown in (b) and (c) in the figure, the conducted EMI peak in the entire frequency band is reduced by dispersing specific frequency bands in a targeted manner.

[0070] (6) According to step (1) and step (4), this embodiment selects v m ( t ) is a sinusoidal signal; determine the carrier frequency according to steps 1.1 and 1.2 f s ( t ) f smin =max{ f smin1 , f smin2},Right now f smin1 and f smin2 According to step 1.3, determine the carrier frequency f s ( t ) f smax . It is necessary to ensure that during the entire modulation process f smin < f s ( t )< f smax , similarly Δ f Need to meet Δ f ≤min[ f smax - f s0 , f s0 - f smin ],Right now f smax - f s0 and f s0 - f smin The smaller of the two. Determine the parameter Δ f As the basis of single-cycle signal spread spectrum modulation, and on the basis of single-cycle signal spread spectrum modulation, a spread spectrum modulation strategy based on secondary frequency modulation is designed; in this embodiment, f smin =5kHz, f smax =15kHz.

[0071] (7) Determination vm ( t ) type and frequency band dispersion method, the remaining parameters to be determined are Δ f , test the sinusoidal signal modulation at different Δ f Under the EMI spectrum, choose the appropriate Δ f As the basis for secondary frequency modulation, in this embodiment, Δ f =2kHz.

[0072] (8) Test different upper frequency limits under the spread spectrum modulation strategy based on secondary frequency modulation f s1 , lower frequency f s2 The two frequency parameters have different combinations to reduce the EMI spikes and determine the optimal parameters. f s1 , f s2 Restrictions must be met: f smin ≤ f s1 , f s2 ≤ f smax ; In this embodiment, the final selection f s1 =13.9kHz, f s2 =6.5kHz.

[0073] To verify the beneficial effects of the present invention, we built a three-phase two-level inverter experimental platform, using resistor-inductor instead of motor as inductive load, variable carrier frequency space vector pulse width modulation (SVPWM) as modulation algorithm, and the load connected to the inverter Z L =2.75Ω+1mH, experimental research is carried out on single period signal spread spectrum modulation and secondary frequency modulation strategy respectively.

[0074] The experiment verifies the superior effect of secondary frequency modulation in reducing conducted EMI compared with traditional single-cycle signal spread spectrum modulation and the degree of impact on other indicators.

[0075] We first test v m ( t ) is a sine signal, a triangle signal, or a sawtooth signal, Δ f =The conducted EMI peak value under 1~4kHz conditions, the EMI spectrum under single signal spread spectrum modulation is as follows Figure 6As shown in (a), (b), (c), and (d) in the figure, the lowest EMI peak value and conditions under single-cycle signal spread spectrum modulation are finally obtained as shown in Table 2. Under single-cycle signal modulation, strategy 3 can minimize the conducted EMI peak value.

[0076] Table 2: Minimum EMI peak value and conditions under single-cycle signal spread spectrum modulation

[0077]

[0078] The experiment verifies the secondary frequency modulation strategy, and sets the frequency band of the sine signal to [8.4kHz, 11.6kHz], and separates the [8kHz, 8.4kHz] frequency band into [ f s2 , 8.4kHz], and disperse the [11.6kHz, 12kHz] band into [11.6kHz, f s1 ] frequency band, the results of the conducted EMI spectrum of the secondary frequency modulation are as follows Figure 6 As shown in (e), the EMI spectrum peak is 86.04dBuV. The comparison target of the secondary frequency modulation method is strategy 3 in Table 2. In terms of conducted EMI, it is reduced by an additional 1.5dBuV compared to 87.54dBuV in strategy 3.

[0079] Modulation strategy ripple and loss index evaluation: test the input voltage of fixed carrier frequency, three periodic signals under different modulation strategies and output current ripple under different carrier frequencies, such as Figure 7 and Figure 8 The results are shown in Table 3 and Table 4. f smin =6.5kHz, Strategy 3 f smin =6.6kHz, referring to Table 3, the frequency deviation of 0.1kHz can be considered to have no effect on the input voltage ripple.

[0080] Table 3: Output phase current ripple corresponding to different carrier frequencies in experimental tests

[0081]

[0082] Table 4: Input voltage ripple under different modulation strategies tested in the experiment

[0083]

[0084] Experimental results f s =10kHz fixed carrier frequency and different v m ( t ) in Δf =1~4kHz condition, the inverter efficiency is 99.46%~99.49% when performing single-cycle signal spread spectrum modulation. The measured power loss is about 260W, and the theoretical power loss is about 280W. The two are close, and the loss fitting is accurate. In this verification example, the inverter efficiency under the spread spectrum modulation strategy based on secondary frequency modulation is 99.47%, and no additional loss is added, indicating that the spread spectrum modulation strategy under this condition has no effect on the loss.

[0085] The above results show that the secondary frequency modulation strategy can reduce the conducted EMI of the inverter to a greater extent than the single-cycle signal spread spectrum modulation without increasing the voltage and current ripple and loss.

[0086] The above description of the embodiments is to facilitate the understanding and application of the present invention by those skilled in the art. It is obvious that those skilled in the art can easily make various modifications to the above embodiments and apply the general principles described herein to other embodiments without creative work. Therefore, the present invention is not limited to the above embodiments. Improvements and modifications made by those skilled in the art to the present invention based on the disclosure of the present invention should be within the protection scope of the present invention.

Claims

1. A spread spectrum modulation method for an inverter based on secondary frequency modulation, comprising the following steps: (1) Determine the maximum deviation Δf of the carrier frequency during spread spectrum modulation of a single periodic signal; (2) Based on the carrier frequency distribution characteristics of different periodic signals during spread spectrum modulation, a sinusoidal signal is selected as the periodic signal v during spread spectrum modulation. m (t); (3) In v m (t) is a sinusoidal signal, and its carrier frequency is dispersed in a specific frequency band, and the carrier period is determined. The specific implementation method is as follows: When 0≤t<0.15T m , 0.35T m ≤t<0.65T m , 0.85T m ≤t≤T m When , the expression of carrier frequency is as follows: f s (t)=f s0 +Δfv m (t) v m (t)=sin(2πf m t) in: f s (t) represents the carrier frequency at time t, f s0 is the carrier center frequency, f m is a periodic signal v m (t) frequency, T m =1 / f m , t represents time; When 0.15T m ≤t<0.25T m , 0.25T m ≤t<0.35T m , 0.65T m ≤t<0.75T m , 0.75T m ≤t<0.85T m When the carrier frequency is a linear function, that is, f s (t) = kt + b, k and b are the slope and intercept of the linear function respectively; At 0.15T m ≤t<0.25T m When k=(f s1max -f s1min ) / 0.1T m , b=2.5f s1min -1.5f s1max At 0.25T m ≤t<0.35T m When k=(f s1min -f s1max ) / 0.1T m , b=3.5f s1max -2.5f s1min At 0.65T m ≤t<0.75T m When k=(f s2min -f s2max ) / 0.1T m , b=7.5f s2max -6.5f s2min At 0.75T m ≤t<0.85T m When k=(f s2max -f s2min ) / 0.1T m , b=8.5f s2min -7.5f s2max Where: f s1min =f s0 +0.8Δf,f s2max =f s0 -0.8Δf,f s1max =f s1 , f s2min =f s2 , f s1 and f s2 They are respectively the upper and lower limits of the carrier frequency of the final spread spectrum modulation strategy after secondary frequency modulation; For the next carrier cycle, it is equal to the inverse of the carrier frequency at the last moment of the previous carrier cycle; (4) Generate a carrier wave and a modulation wave according to the duty cycle and carrier period of the basic voltage vector of each sector, and compare the carrier wave with the modulation wave to generate a PWM signal for the inverter.

2. The inverter spread spectrum modulation method based on secondary frequency modulation according to claim 1, characterized in that: The value of the maximum carrier frequency deviation Δf in step (1) must meet the following conditions: 0.1f s0 ≤Δf≤min[f smax -f s0 ,f s0 -f smin ] Where: f smin and f smax are the lower and upper limits of the carrier frequency.

3. The inverter spread spectrum modulation method based on secondary frequency modulation according to claim 2, characterized in that: The lower limit value f of the carrier frequency smin and the upper limit f smax The quantitative relationship between the input voltage ripple, output current ripple, inverter switching loss and various spread spectrum parameters of the inverter under the spread spectrum modulation strategy is established, and then the input voltage ripple and output current ripple are calculated and determined according to the limit requirements to limit f s1 and f s2 The value of: f smin ≤f s1 , f s2 ≤f smax .

4. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: The processor is used to execute the computer program to implement the inverter spread spectrum modulation method based on secondary frequency modulation as claimed in any one of claims 1 to 3.

5. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the computer program implements an inverter spread spectrum modulation method based on secondary frequency modulation as claimed in any one of claims 1 to 3.

Images