Cascade power grid simulator real-time selective harmonic generation method based on Newton-ant colony algorithm
Through the Newton-ant colony algorithm combined with the SHGPWM model, the problem of high-order harmonic elimination in simulated power grids is solved, efficient grid control is achieved, switching losses are reduced and power efficiency is improved, and it is suitable for adaptability testing of distributed system power grids.
Patent Information
- Application Number
- CN202311707369.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-12
- Publication Date
- 2024-12-20
AI Technical Summary
The prior art is difficult to effectively eliminate high-order harmonics in analog power grids, resulting in increased line loss and reduced operating efficiency of power electronic equipment. Increasing switching frequency of traditional pulse width modulation methods will lead to higher switching losses and low power efficiency.
Using the real-time selective harmonic generation method of cascaded power grid simulator based on Newton-ann colony algorithm, the SHGPWM model of the H-bridge inverter is constructed, and the initial value of the switch angle is solved using the ant colony algorithm, and the final switch is obtained in combination with the Newton iterative algorithm. The angle realizes real-time SHGPWM control of the cascading H bridge, reduces iterative steps and improves convergence speed.
It realizes static difference tracking of specific subharmonics, meets the adaptive testing needs of distributed system power grids, reduces switching losses and improves power efficiency, and provides an effective control strategy.
Smart Images

Figure CN119171441A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technology of cascade power grid simulator, in particular to a real-time selective harmonic generation method of cascade power grid simulator based on Newton-ant colony algorithm. Background Art
[0002] In recent years, renewable energy sources such as wind power and photovoltaics have developed rapidly and are showing a trend of scale. However, renewable energy systems are susceptible to seasonal influences and may experience problems such as voltage drops, frequency changes, and high-order harmonic pollution introduced by power electronic grid-connected equipment. The presence of high-order harmonics in the power grid will increase line losses and reduce the operating efficiency of related power electronic equipment. Therefore, in order to test the comprehensive operating capabilities of new energy and related power electronic equipment under complex power grid conditions, a power grid simulator that can simulate and reproduce complex waveforms of power grid voltage under multiple working conditions has been widely studied by scholars at home and abroad. Scholars have successively proposed single closed-loop control based on proportional integral (PI), proportional resonance (PR), quasi-proportional resonance (QPR) and deadbeat control strategies to improve the dynamic response performance of the system, sliding mode control strategies based on improved exponential reaching laws, and parallel structure fractional-order repetitive controllers.
[0003] Compared with simulating the power grid from the control level, adopting modulation strategies in multilevel converters can bring more freedom to solve problems. However, in practical applications, if traditional pulse width modulation methods are used, the switching frequency must be increased to eliminate low-order harmonics, which will lead to higher switching losses and lower power efficiency. Summary of the invention
[0004] The present invention proposes a real-time selective harmonic generation method for a cascade power grid simulator based on a Newton-ant colony algorithm.
[0005] The technical solution to achieve the purpose of the present invention is: a real-time selective harmonic generation method for a cascade power grid simulator based on Newton-ant colony algorithm, comprising:
[0006] Construct the SHGPWM model of H-bridge inverter;
[0007] The SHGPWM model is solved using the ant colony ACA algorithm to obtain the initial value of the switching angle;
[0008] Substitute the initial value of the switch angle into the Newton iteration NR algorithm to obtain the final switch angle;
[0009] The cascaded H-bridge is controlled in real time by SHGPWM according to the switching angle obtained by Newton algorithm.
[0010] Preferably, the specific method for constructing the SHGPWM model is:
[0011] Set the fundamental and harmonic modulation ratio;
[0012] The output voltage waveform of N cascaded H-bridges is decomposed by Fourier transform to obtain the SHGPWM model of the H-bridge inverter.
[0013] Preferably, the SHGPWM model of the H-bridge inverter is specifically:
[0014]
[0015] In the formula, θ1, θ2, …θ N represents the switching angle, m represents the fundamental modulation ratio, m h3 、m h5 、m h(2N-1) They represent 3rd, 5th, and 2N-1th harmonic modulation ratios respectively, and N is the number of H-bridge inverters.
[0016] Preferably, the SHGPWM model is solved by using the ant colony ACA algorithm, and the specific steps of obtaining the initial value of the switching angle are:
[0017] According to the SHGPWM model construct function y1,y2,y3…y N , specifically:
[0018]
[0019] In the formula, θ1, θ2, …θ N represents the switching angle, m represents the fundamental modulation ratio, m h3 、m h5 、m h(2N-1) They represent the 3rd, 5th, and 2N-1th harmonic modulation ratios respectively, and N is the number of H-bridge inverters;
[0020] The constructed function is transformed into fitness y, specifically:
[0021]
[0022] Use the ant colony ACA algorithm to find the minimum value of fitness y, so that y1, y2, y3…y N All approach 0, the minimum value of y is -1, and the exact initial switch angle group θ0 = [θ 10 ,θ 20 ,θ 30 ,…θ N0 ] T .
[0023] Preferably, the initial value of the switch angle is brought into the Newton iteration NR algorithm, and the specific method for obtaining the final switch angle is:
[0024] Step 3.1: Transpose the SHGPWM model to obtain:
[0025]
[0026] In the formula, θ1, θ2, …θ N represents the switching angle, m represents the fundamental modulation ratio, m h3 、m h5 、m h(2N-1) They represent the 3rd, 5th, and 2N-1th harmonic modulation ratios respectively, and N is the number of H-bridge inverters;
[0027] The above formula can be simplified as
[0028] F(θ)=0
[0029]
[0030] Step 3.2: Use the switch angle θ0 obtained by the ant colony algorithm as the initial value of the Newton iteration method, substitute it into the above formula, and determine the initial value F0 of the switch angle function
[0031] F(θ0)=F0
[0032] Step 3.3: Linearize the equation in step 3.2 to obtain:
[0033] F(θ)≈F0+F′(θ0)dθ=0
[0034] in:
[0035]
[0036] dθ is the switching angle error;
[0037] Step 3.4: Solve the linearized equations in step 3.2 to obtain the error dθ of the switching angle, which is:
[0038] dθ=-(F′(θ0)) -1 ×F0
[0039] Step 3.5: Update the initial value of the Newton iteration method in step 3.2 to θ n =θ n-1 +dθ, n is the number of iterations, repeat steps 3.2 to 3.4 until the error of the switch angle meets the set conditions
[0040] Compared with the prior art, the present invention has the following significant advantages: first, the present invention establishes a switching angle equation, and uses the characteristics of the ant colony algorithm that it does not rely on initial values, has fast calculation speed, has good optimization ability and robustness, and is easy to combine with other algorithms. The initial value is solved by the ant colony algorithm, and the maximum number of harmonics corresponding to different switching frequencies and calculation steps is obtained by further combining the Newton algorithm, and the modulation ratio-switching angle correspondence is analyzed. Compared with the traditional method, it has the significant advantages of fewer iterative steps and faster convergence speed; the initial switching angle is obtained by a lookup table, and is fed forward to the voltage control loop to achieve zero-static error tracking of specific harmonics, which can meet the distributed system power grid adaptability test requirements and is a very effective control strategy for multi-level converters.
[0041] The present invention is described in further detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Figure 1 It is a topology of cascaded H-bridge multi-level inverter.
[0043] Figure 2 It is the control block diagram of cascaded H-bridge real-time SHGPWM based on QPR control.
[0044] Figure 3 It is the cascade output voltage and its FFT when SHGPWM is implemented with mirror symmetry. DETAILED DESCRIPTION
[0045] A real-time selective harmonic generation method for a cascade power grid simulator based on a Newton-ant colony algorithm, wherein the topological structure of the cascade power grid simulator based on the Newton-ant colony algorithm includes: three phases a, b and c with exactly the same structure, each phase includes n power conversion modules, a DC power supply vs, a three-phase AC power supply, and a three-phase filter impedance. Each power conversion module includes a bidirectional CLLC resonant converter, a CLLC output capacitor, and an H-bridge inverter. One port of the bidirectional CLLC resonant converter is connected to a DC power supply vs, and the other port is connected to a CLLC output capacitor, and the other port of the CLLC output capacitor is connected to a DC side port of an H-bridge inverter. The first branch of the AC side port of the H-bridge inverter is connected to a three-phase filter impedance and a three-phase AC power supply in sequence, and the second branch is cascaded with a second power conversion module of the same phase.
[0046] Specifically, each H-bridge inverter includes: a first switch tube S of the H-bridge inverter an9 , H-bridge inverter No. 2 switch tube S an10 , H-bridge inverter No. 3 switch tube S an11 , H-bridge inverter No. 4 switch tube S an12One of the DC ports of the first H-bridge inverter of phase a is connected to a DC voltage, the first branch of the other AC port is connected to a three-phase filter impedance and a three-phase AC power supply in sequence, and the second branch is cascaded with the AC port of the second H-bridge inverter of the same phase. The second branch of the AC port of the second H-bridge inverter is cascaded with the AC port of the third H-bridge inverter of the same phase, and so on until it is connected to the Nth H-bridge inverter. The second branch of the AC port of the Nth H-bridge inverter is connected to the b-phase H-bridge inverter.
[0047] The method of the present invention is mainly applied to cascaded H-bridge inverters, a real-time selective harmonic generation method of a cascaded power grid simulator based on Newton-ant colony algorithm, and the specific steps are:
[0048] Step 1: Construct the SHGPWM (real-time selective harmonic generation pulse width modulation technology) model. The specific steps are as follows:
[0049] Step 1.1: First, set the fundamental and harmonic modulation ratios. The output voltage of each H-bridge inverter has a switching angle in the first 1 / 4 cycle. N H-bridge inverters have a total of N switching angles, which can control 3, 5, and 2N-1 harmonics. The fundamental and harmonic modulation ratios are m, m respectively. h3 、m h5 、m h(2N-1) .
[0050] Step 1.2 Perform Fourier decomposition on the output voltage waveforms of the N cascaded H-bridges to obtain the expression of the SHGPWM model of the H-bridge as shown in equation (1), θ1, θ2, …θ N Indicates the switching angle.
[0051]
[0052] Step 2: Use the ant colony ACA algorithm to solve the SHGPWM model and obtain the initial value of the switching angle. The specific steps are as follows:
[0053] Step 2.1 Transpose the terms of the equation system of formula (1), subtract the left-hand equation from the right-hand equation, and construct the function y1, y2, y3…y N The formula (2) is as follows:
[0054]
[0055] Step 2.2 transforms the constructor into a formula for fitness y:
[0056]
[0057] Step 2.3 Use the ACA algorithm to find the minimum value of fitness y, so that y1, y2, y3…y N All approach 0, the minimum value of y is -1, and the precise initial switching angle group is recorded as θ0.
[0058] θ0=[θ 10 ,θ 20 ,θ 30 ,…θ N0 ] T (4)
[0059] Step 3: Newton Iteration NR algorithm to calculate the final switch angle
[0060] Step 3.1: Rearrange the terms of equation (1) to obtain the following equation:
[0061]
[0062] Can be abbreviated as
[0063] F(θ)=0 (6)
[0064]
[0065] Step 3.2: Use the switch angle θ0 obtained by the ant colony algorithm as the initial value of the Newton iteration method and substitute it into equation (7) to determine the initial value of the switch angle function F0.
[0066] F(θ0)=F0 (8)
[0067] Step 3.3 Linearize formula (8) to obtain the following:
[0068] F(θ)≈F0+F′(θ0)dθ=0 (9)
[0069] in
[0070]
[0071] Step 3.4: Solve the linear equation group (9) to obtain the error dθ of the switching angle.
[0072] dθ=-(F′(θ0)) -1 ×F0 (11)
[0073] In step 3.5, replace the initial value θ0 with θ1=θ0+dθ. Repeat steps 3.2 to 3.4 above until the switch angle that meets the accuracy requirements is solved. The switch angle error dθ can be set as follows.
[0074] -10 -6 ≤dθ≤10 -6 (12)
[0075] Step 4: Perform real-time SHGPWM control on the cascaded H-bridge according to the switch angle obtained by the Newton algorithm. The specific steps are as follows:
[0076] Step 4.1 Put the final switching angle obtained by the NR-ACA algorithm into the lookup table LUT as the initial switching angle θ for real-time SHGPWM control 10 ,θ 20 ,θ 30 …θ N0 , calculated from the initial switching angle θ 10 ,θ 20 ,θ 30 …θ N0 The matrix T l and stored in a lookup table, where T l The calculation formula is:
[0077]
[0078] Step 4.2: Set the given fundamental, 3rd, 5th, and 2N-1th harmonic amplitudes V h1 * 、V h3 * 、V h5 * …V h(2N-1) * , multiplied by the unit sinusoidal signal sin(ωt) in phase with the AC power grid and the actual cascaded H-bridge output voltage U o The difference is controlled by quasi-proportional QPR to obtain the fundamental wave, 3rd harmonic, 5th harmonic, ..., 2N-1th harmonic amplitude V of the output voltage after QPR control. h1QPR 、V h3QPR 、V h5QPR …V h(2N-1)QPR . submit V h1QPR 、submit V h3QPR 、submit V h5QPR …submitV h(2N-1)QPR It represents the extraction of the fundamental wave and each harmonic amplitude of the output voltage after QPR control.
[0079] Step 4.3: Multiply the fundamental wave and each harmonic amplitude by the DC side voltage NE after N H bridges are cascaded and then add it to the output value V controlled by SHGPWM h1 、V h3 、V h5 …V h(2N-1) Subtract, and get through the proportional controller K and the integral controller 1 / s Then, the switching angle disturbance value is obtained from formula (14):
[0080]
[0081] Step 4.4: Set the switch angle disturbance value The actual switching angles θ1, θ2, …θ are added to the switching angles calculated by the NR-ACA algorithm and stored in the lookup table to obtain the actual switching angles θ1, θ2, …θ N , one is used as the switching signal of the cascaded H-bridge CMI to achieve multi-level output, and the other is passed through the transient observer given by formula (15) to obtain the output of the overall closed-loop control V h1 、V h3 、V h5 …V h(2N-1) , and the input quantity forms a closed-loop control.
[0082]
[0083] The embodiments are described in more detail.
[0084] A real-time selective harmonic generation method for a cascade power grid simulator based on Newton-ant colony algorithm, the specific implementation method is as follows:
[0085] Construct the SHGPWM model of H-bridge inverter;
[0086] Cascade output voltage v o The waveform is 1 / 4 cycle symmetrical, and uses a mirror-symmetrical switch angle arrangement. Taking a two-way cascaded five-level inverter as an example, a half-cycle modulation method is used, that is, each switch tube works for half a cycle. For the first H bridge: the output voltage v H1 There are N switching angles in the first 1 / 4 cycle; N pulses are generated in the first half cycle; the waveform of the entire cycle is symmetrical about the point (π, 0); the output voltage v H1 There are 3 levels: E, 0, -E. For the second H bridge: output voltage v H2 There are N+1 switching angles in the first 1 / 4 cycle; N+1 pulses are generated in the first half cycle; the waveform of the entire cycle is symmetrical about the point (π, 0); v H2 There are 3 levels: E, 0, -E. H1 and v H2 The five-level cascade output voltage v is obtained by superposition. o :2E, E, 0, -E, -2E.
[0087] The driving pulses of the upper and lower bridge arms are complementary and opposite, that is, S an9 and S an10 , S an11 and S an12 The driving signals of the two H bridges are opposite. The two H bridges satisfy the following conditions. h1.H1 、m h1.H2 , m are the fundamental modulation ratios of the output voltages of the first H-bridge, the second H-bridge, and the cascaded H-bridge, respectively; m h3.H1 、m h3.H2 、m h3 is the third harmonic modulation ratio; m h5.H1 、mh5.H2 、m h5 is the 5th harmonic modulation ratio; m h7.H1 、m h7.H2 、m h7 It is the 7th harmonic modulation ratio.
[0088]
[0089] For the first H-bridge: the number of switching angles in the first T / 4 is N1=3, which are θ1, θ2, and θ3 respectively. Given m, m h3 、m h5 , thereby controlling the fundamental wave, 3rd and 5th harmonics. The equations that need to be solved are as follows:
[0090]
[0091] For the second H-bridge: the number of switch angles in the first T / 4 is N2=N1+1=4, which are θ4, θ5, θ6, θ7 respectively. Given m, m h3 、m h5 , thereby controlling the fundamental wave, 3rd and 5th harmonics. In order to control the 7th harmonic of the cascade output voltage, the equations that need to be solved are as follows:
[0092]
[0093] The group ACA algorithm is used to solve the SHGPWM model and obtain the initial value of the switching angle;
[0094] For the cascaded multilevel inverter SHGPWM model, the nonlinear equations that need to be solved to control the 3rd and 5th harmonics are as follows:
[0095]
[0096] Move the terms of the nonlinear equations, subtract the left and right equations, and construct the functions y1, y2, and y3 to obtain the following formula:
[0097]
[0098] Transformed to find the minimum value of fitness y:
[0099]
[0100] The ACA algorithm is used to make y1, y2, and y3 approach 0, the minimum value of y is -1, and the precise initial switching angle is obtained.
[0101] Substitute the initial value of the switch angle into the Newton iteration NR algorithm to obtain the final switch angle;
[0102] The iteration process is as follows:
[0103] 1) Substitute the initial value of the switching angle θ0 into formula (9), denoted as F0
[0104] F(θ0)=F0
[0105] 2) Linearize formula (11) to get the following:
[0106] F(θ)≈F0+F′(θ0)dθ=0
[0107]
[0108] 3) Solve the linear equations to obtain the error dθ of the switching angle.
[0109] dθ=-(F′(θ0)) -1 ×F0
[0110] 4) Replace the initial value θ0 with θ1=θ0+dθ. Repeat steps 1) to 3) above until the switch angle that meets the accuracy requirements is solved. The switch angle error dθ can be set as follows.
[0111] -10 -6 ≤dθ≤10 -6
[0112] The control block diagram of cascade H-bridge real-time SHGPWM is as follows: Figure 2 As shown, the cascaded H-bridge is controlled in real time by the switching angle obtained by the NR-ACA algorithm.
[0113] 1. Use the switching angle calculated by the NR-ACA algorithm as the initial switching angle θ 10 ,θ 20 ,θ 30 , the matrix T formed by the initial switching angle l It is given by a formula and stored in a lookup table LUT.
[0114]
[0115] 2. Given fundamental wave, 5th and 7th harmonics and output voltage U o After the difference is made, the quasi-proportional control is performed to output the fundamental wave of the voltage, the amplitude of the 5th harmonic and the 7th harmonic; the proportional coefficient of the controller is set to K1=K2=K3=1000; the switching angle disturbance value is obtained according to the formula through the 1 / s integral controller
[0116]
[0117] 3. One path of θ1, θ2, and θ3 is used as the input of the harmonic generation PWM modulation strategy to generate a switching signal to act on the cascaded H-bridge to achieve multi-level output, and the other path passes through the transient observer Observer to form a closed-loop control with the input quantity.
[0118]
[0119] Take the DC bus voltage E = 20V, R = 50Ω; m = 0.7, m h3 =0.12, m h5 =0.1, m h7 =0.08. The desired controlled cascade output voltage v o The harmonic amplitude V h1 * =28V, V h3 * =4.8V, V h5 * =4V, V h7 * =3.2V; the first H-bridge output voltage v H1 The harmonic amplitude V h1.H1 * =14V, V h3.H1 * =2.4V, V h5.H1 * =2V; the second H-bridge output voltage v H2 The harmonic amplitude V h1.H2 * =14V, V h3.H2 * =2.4V, V h5.H2 * =2V. The cascade output voltage, each H-bridge output voltage and its FFT are shown in Figure 3 shown.
[0120] Depend on Figure 3 It can be seen that v H1 、v H2 、v o The fundamental amplitudes are 14V, 13.8V, and 28V, which are consistent with the theoretical fundamental amplitudes of 14V, 14V, and 28V. Similarly, the amplitudes of the third harmonics are 2.4V, 2.4V, and 4.8V, which are consistent with the theoretical third harmonic amplitudes of 2.4V, 2.4V, and 4.8V; the amplitudes of the fifth harmonics are 2V, 2V, and 4V, which are consistent with the theoretical fifth harmonic amplitudes of 2V, 2V, and 4V; v o The 7th harmonic amplitude Vh7 = 3.2V, which is consistent with the theoretical 7th harmonic amplitude of 3.2V.
Claims
1. A real-time selective harmonic generation method for a cascaded power grid simulator based on Newton-Ant Colony Algorithm, characterized in that: include: Construct the SHGPWM model of H-bridge inverter; The group ACA algorithm is used to solve the SHGPWM model and obtain the initial value of the switching angle; Substitute the initial value of the switch angle into the Newton iteration NR algorithm to obtain the final switch angle; The cascaded H-bridge is controlled in real time by SHGPWM according to the switching angle obtained by Newton algorithm.
2. The real-time selective harmonic generation method of a cascaded power grid simulator based on Newton-Ant Colony Algorithm according to claim 1 is characterized in that: The specific method of constructing the SHGPWM model is: Set the fundamental and harmonic modulation ratio; The output voltage waveform of N cascaded H-bridges is decomposed by Fourier transform to obtain the SHGPWM model of the H-bridge inverter.
3. The real-time selective harmonic generation method of a cascaded power grid simulator based on Newton-Ant Colony Algorithm according to claim 1 is characterized in that: The SHGPWM model of the H-bridge inverter is as follows: In the formula, θ1, θ2, …θ N represents the switching angle, m represents the fundamental modulation ratio, m h3 、m h5 、m h(2N-1) They represent 3rd, 5th, and 2N-1th harmonic modulation ratios respectively, and N is the number of H-bridge inverters.
4. The real-time selective harmonic generation method of a cascaded power grid simulator based on Newton-Ant Colony Algorithm according to claim 1 is characterized in that: The specific steps of solving the SHGPWM model using the ant colony ACA algorithm and obtaining the initial value of the switching angle are as follows: According to the SHGPWM model construct function y1,y2,y3…y N , specifically: In the formula, θ1, θ2, …θ N represents the switching angle, m represents the fundamental modulation ratio, m h3 、m h5 、m h(2N-1) They represent the 3rd, 5th, and 2N-1th harmonic modulation ratios respectively, and N is the number of H-bridge inverters; The constructed function is transformed into fitness y, specifically: Use the ant colony ACA algorithm to find the minimum value of fitness y, so that y1, y2, y3…y N All approach 0, the minimum value of y is -1, and the exact initial switch angle group θ0 = [θ 10 ,θ 20 ,θ 30 ,…θ N0 ] T .
5. The real-time selective harmonic generation method of a cascaded power grid simulator based on Newton-Ant Colony Algorithm according to claim 4 is characterized in that: Substitute the initial value of the switch angle into the Newton iteration NR algorithm, and the specific method to obtain the final switch angle is: Step 3.1: Transpose the SHGPWM model to obtain: In the formula, θ1, θ2, …θ N represents the switching angle, m represents the fundamental modulation ratio, m h3 、m h5 、m h(2N-1) They represent the 3rd, 5th, and 2N-1th harmonic modulation ratios respectively, and N is the number of H-bridge inverters; The above formula can be simplified as F(θ)=0 Step 3.2: Use the switch angle θ0 obtained by the ant colony algorithm as the initial value of the Newton iteration method, substitute it into the above formula, and determine the initial value F0 of the switch angle function F(θ0)=F0 Step 3.3: Linearize the equation in step 3.2 to obtain: F(θ)≈F0+F′(θ0)dθ=0 in: dθ is the switching angle error; Step 3.4: Solve the linearized equations in step 3.2 to obtain the error dθ of the switching angle, which is: dθ=-(F′(θ0)) -1 ×F0 Step 3.5: Update the initial value of the Newton iteration method in step 3.2 to θ n =θ n-1 +dθ, n is the number of iterations, and steps 3.2 to 3.4 are repeated until the error of the switching angle meets the set conditions.
6. The real-time selective harmonic generation method of a cascaded power grid simulator based on Newton-Ant Colony Algorithm according to claim 1, characterized in that: The specific method of real-time SHGPWM control of the cascaded H-bridge according to the switching angle obtained by the Newton algorithm is: Step 4.1: Put the final switching angle obtained by the NR-ACA algorithm into the lookup table LUT as the initial switching angle θ for real-time SHGPWM control 10 ,θ 20 ,θ 30 …θ N0 , calculated from the initial switching angle θ 10 ,θ 20 ,θ 30 …θ N0 The matrix T l and stored in the lookup table at the same time; Step 4.2: Set the given fundamental, 3rd, 5th, and 2N-1th harmonic amplitudes V h1 * 、V h3 * 、V h5 * …V h(2N-1) * Multiply it by the unit sinusoidal signal sin(ωt) in phase with the AC power grid and the actual cascade H-bridge output voltage U o The difference is controlled by quasi-proportional QPR to obtain the fundamental wave, 3rd harmonic, 5th harmonic, ..., 2N-1th harmonic amplitude V of the output voltage after QPR control. h1QPR 、V h3QPR 、V h5QPR …V h(2N-1)QPR ; Extract the fundamental wave and harmonic amplitudes of the output voltage after QPR control submitV h1QPR 、submit V h3QPR 、submit V h5QPR …submit V h(2N-1)QPR ; Step 4.3: Multiply the fundamental wave and harmonic amplitude of the output voltage after QPR control by the DC side voltage NE after N H bridges are cascaded and then add them to the output value V controlled by SHGPWM. h1 、V h3 、V h5 …V h(2N-1) Subtract, and get through the proportional controller K and the integral controller 1 / s Then we can get the switching angle disturbance value Step 4.4: Set the switch angle disturbance value The actual switching angles θ1, θ2, …θ are added to the switching angles calculated by the NR-ACA algorithm and stored in the lookup table to obtain the actual switching angles θ1, θ2, …θ N One path is used as the switching signal of the cascaded H-bridge CMI to achieve multi-level output, and the other path passes through the transient observer to obtain the output of the overall closed-loop control V h1 、V h3 、V h5 …V h(2N-1) .
7. The real-time selective harmonic generation method of a cascaded power grid simulator based on Newton-Ant Colony Algorithm according to claim 6 is characterized in that: Switching angle disturbance value Specifically: In the formula, It means that after the proportional controller K and the integral controller 1 / s, A collection of .
8. The real-time selective harmonic generation method of a cascaded power grid simulator based on Newton-Ant Colony Algorithm according to claim 6 is characterized in that: Actual switching angles θ1, θ2…θ N After the transient observer is removed, the output V of the overall closed-loop control is obtained. h1 、V h3 、V h5 …V h(2N-1) , the specific formula is: Where E is the DC side voltage.
