An active power filter and a parameter determination method thereof, and a computer readable storage medium
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HENAN XJ INSTR
- Filing Date
- 2026-05-06
- Publication Date
- 2026-08-07
AI Technical Summary
[0007]本发明的目的在于提供一种有源电力滤波器及其参数确定方法、计算机可读存储介质，用以解决现有技术中有源电力滤波器的稳定性、性能、成本和响应速度无法兼顾的技术问题
[0025]本发明的有益效果为：本发明通过仿真模拟和寻优算法相结合，并且设置寻优算法的适应度函数从滤波器成本、响应速度、性能和稳定性四个方面考虑，设置寻优算法的适应度函数，对有源电力滤波器的滤波参数和控制参数进行寻优，将最终的寻优结果作为有源电力滤波器的参数，使得该有源电力滤波器，能够兼顾滤波器成本、响应速度、性能和稳定性。
Smart Images
Figure CN122532972A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of power system harmonic compensation technology, specifically relating to an active power filter and its parameter determination method, as well as a computer-readable storage medium. Background Technology
[0002] With the development of modern industry, the widespread use of nonlinear loads (such as rectifiers and frequency converters) in power systems has led to serious power quality problems, among which harmonic pollution is particularly prominent. Harmonics reduce the efficiency of power transmission and utilization, cause overheating and malfunction of electrical equipment, and even damage equipment, posing a threat to the safe and stable operation of the power grid. Active power filters (APFs), as a highly efficient dynamic harmonic compensation device, can detect and compensate for harmonics and reactive currents in real time, and have become key equipment for controlling power grid harmonics and improving power quality.
[0003] The performance of an APF largely depends on the rationality of its main circuit parameters (such as AC-side inductance) and control parameters (such as current loop PI controller parameters). These parameters are interdependent and jointly determine the APF's compensation accuracy, dynamic response speed, stability, and cost.
[0004] Currently, the methods for determining the parameters of APF mainly fall into the following categories: Traditional analytical design method: This method derives the parameter range by establishing a mathematical model based on circuit theory and operating principles. Although the physical meaning is clear, the design results tend to be conservative in order to ensure system stability, resulting in excessively large parameter margins. This prevents the APF from fully realizing its performance potential, potentially leading to increased costs (such as increased inductor size) and insufficient dynamic response.
[0005] Trial and error approach: In practical engineering, engineers often adjust parameters based on past experience and trial and error. This method relies heavily on the designer's personal experience, lacks systematic theoretical guidance, and results in a cumbersome and time-consuming design process. Furthermore, it is difficult to obtain a globally optimal solution, and the consistency and reliability of the design results cannot be guaranteed.
[0006] In summary, existing active power filter parameter setting methods cannot simultaneously address the stability, performance, cost, and response speed of active power filters. Summary of the Invention
[0007] The purpose of this invention is to provide an active power filter and a method for determining its parameters, as well as a computer-readable storage medium, to solve the technical problem that the stability, performance, cost, and response speed of active power filters in the prior art cannot be balanced.
[0008] To address the aforementioned technical problems, this invention provides a method for determining the parameters of an active power filter, comprising the following steps: 1) Construct a simulation model of the compensation system, which includes an active power filter, a power grid compensated by the active power filter, and a load supplied by the power grid; 2) With the optimization objectives of achieving the best performance, fastest response speed, strongest stability, and lowest cost for the active power filter, an iterative optimization algorithm is used to iteratively optimize the filtering parameters and control parameters of the active power filter. The optimal solution obtained is the parameter determination result of the active power filter. During the iterative optimization process, after each iteration, the filtering parameters and control parameters of the active power filter after the iteration are simulated using a simulation model to obtain the grid current waveform, thereby evaluating the performance, response speed, and stability of the active power filter.
[0009] Furthermore, the performance of the active power filter is evaluated using the distortion rate of the grid current waveform. Correspondingly, the fitness function for evaluating the performance of the active power filter is: F THD =1, THD ig > D m F THD =0, THD ig ≤ D m In the formula, F THD The fitness function used to evaluate the performance of active power filters is THD. ig The distortion rate of the grid current waveform. D m To set the distortion rate.
[0010] Furthermore, the response speed of the active power filter is evaluated using the settling time of the grid current. Correspondingly, the fitness function for evaluating the response speed of the active power filter is: F t =1, t s > t m F t =0, t s ≤ t m In the formula, F t To evaluate the fitness function of the settling time of an active power filter, t s The regulation time of the grid current.t m To set the adjustment time.
[0011] Furthermore, the phase margin is used to evaluate the stability of the active power filter. Correspondingly, the fitness function for evaluating the stability of the active power filter is: F s =0, f min ≤ f ( oh c +180°≤ f max F s =1, f ( oh c +180° < f min or f ( oh c +180°> f max In the formula, F s The fitness function is used to evaluate the stability of active power filters. f ( oh c ) represents the grid current at the crossover frequency oh c Phase at that point, f min To compensate the system at the crossover frequency oh c The minimum set phase at that location, f max To compensate the system at the crossover frequency oh c The maximum set phase at that location.
[0012] Furthermore, the filtering parameters include the filtering inductance.
[0013] Furthermore, the fitness function for evaluating the cost of an active power filter is positively correlated with the inductance of the active power filter.
[0014] Furthermore, the control parameters include the proportional coefficient and integral coefficient of the current inner loop PI controller of the active power filter converter.
[0015] Furthermore, the fitness functions for evaluating the performance, response speed, stability, and cost of the active power filter are weighted and summed to obtain the overall fitness function of the optimization algorithm.
[0016] The beneficial effects of this invention are as follows: This invention combines simulation and optimization algorithm, and sets the fitness function of the optimization algorithm to consider four aspects: filter cost, response speed, performance and stability. The fitness function of the optimization algorithm optimizes the filtering parameters and control parameters of the active power filter, and uses the final optimization result as the parameters of the active power filter, so that the active power filter can take into account the filter cost, response speed, performance and stability.
[0017] To address the aforementioned technical problems, this invention also provides an active power filter. The control parameters and filtering parameters of the active power filter are designed using the parameter determination method for active power filters described below: 1) Construct a simulation model of the compensation system, which includes an active power filter, a power grid compensated by the active power filter, and a load supplied by the power grid; 2) With the optimization objectives of achieving the best performance, fastest response speed, strongest stability, and lowest cost for the active power filter, an iterative optimization algorithm is used to iteratively optimize the filtering parameters and control parameters of the active power filter. The optimal solution obtained is the parameter determination result of the active power filter. During the iterative optimization process, after each iteration, the filtering parameters and control parameters of the active power filter after the iteration are simulated using a simulation model to obtain the grid current waveform, thereby evaluating the performance, response speed, and stability of the active power filter.
[0018] Furthermore, the performance of the active power filter is evaluated using the distortion rate of the grid current waveform. Correspondingly, the fitness function for evaluating the performance of the active power filter is: F THD =1, THD ig > D m F THD =0, THD ig ≤ D m In the formula, F THD The fitness function used to evaluate the performance of active power filters is THD. ig The distortion rate of the grid current waveform. D m To set the distortion rate.
[0019] Furthermore, the response speed of the active power filter is evaluated using the settling time of the grid current. Correspondingly, the fitness function for evaluating the response speed of the active power filter is: F t =1, t s > t m F t =0, t s ≤ t m In the formula, F t To evaluate the fitness function of the settling time of an active power filter, t s The regulation time of the grid current. t m To set the adjustment time.
[0020] Furthermore, the phase margin is used to evaluate the stability of the active power filter. Correspondingly, the fitness function for evaluating the stability of the active power filter is: F s =0, f min ≤ f ( oh c +180°≤ f max F s =1, f ( oh c +180° < f min or f ( oh c +180°> f max In the formula, F s The fitness function is used to evaluate the stability of active power filters. f ( oh c ) represents the grid current at the crossover frequency oh c Phase at that point, f min To compensate the system at the crossover frequency oh c The minimum set phase at that location, fmax To compensate the system at the crossover frequency oh c The maximum set phase at that location.
[0021] Furthermore, the filtering parameters include the filtering inductance.
[0022] Furthermore, the fitness function for evaluating the cost of an active power filter is positively correlated with the inductance of the active power filter.
[0023] Furthermore, the control parameters include the proportional coefficient and integral coefficient of the current inner loop PI controller of the active power filter converter.
[0024] Furthermore, the fitness functions for evaluating the performance, response speed, stability, and cost of the active power filter are weighted and summed to obtain the overall fitness function of the optimization algorithm.
[0025] The beneficial effects of this invention are as follows: This invention combines simulation and optimization algorithm, and sets the fitness function of the optimization algorithm to consider four aspects: filter cost, response speed, performance and stability. The fitness function of the optimization algorithm optimizes the filtering parameters and control parameters of the active power filter, and uses the final optimization result as the parameters of the active power filter, so that the active power filter can take into account the filter cost, response speed, performance and stability.
[0026] To address the aforementioned technical problems, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed, implements a method for determining the parameters of an active power filter, comprising the following steps: 1) Construct a simulation model of the compensation system, which includes an active power filter, a power grid compensated by the active power filter, and a load supplied by the power grid; 2) With the optimization objectives of achieving the best performance, fastest response speed, strongest stability, and lowest cost for the active power filter, an iterative optimization algorithm is used to iteratively optimize the filtering parameters and control parameters of the active power filter. The optimal solution obtained is the parameter determination result of the active power filter. During the iterative optimization process, after each iteration, the filtering parameters and control parameters of the active power filter after the iteration are simulated using a simulation model to obtain the grid current waveform, thereby evaluating the performance, response speed, and stability of the active power filter.
[0027] Furthermore, the performance of the active power filter is evaluated using the distortion rate of the grid current waveform. Correspondingly, the fitness function for evaluating the performance of the active power filter is: F THD =1, THD ig > D m F THD =0, THD ig ≤ D m In the formula, F THD The fitness function used to evaluate the performance of active power filters is THD. ig The distortion rate of the grid current waveform. D m To set the distortion rate.
[0028] Furthermore, the response speed of the active power filter is evaluated using the settling time of the grid current. Correspondingly, the fitness function for evaluating the response speed of the active power filter is: F t =1, t s > t m F t =0, t s ≤ t m In the formula, F t To evaluate the fitness function of the settling time of an active power filter, t s The regulation time of the grid current. t m To set the adjustment time.
[0029] Furthermore, the phase margin is used to evaluate the stability of the active power filter. Correspondingly, the fitness function for evaluating the stability of the active power filter is: F s =0, f min ≤ f ( oh c +180°≤ f max F s =1, f ( oh c +180° < f min or f ( ohc +180°> f max In the formula, F s The fitness function is used to evaluate the stability of active power filters. f ( oh c ) represents the grid current at the crossover frequency oh c Phase at that point, f min To compensate the system at the crossover frequency oh c The minimum set phase at that location, f max To compensate the system at the crossover frequency oh c The maximum set phase at that location.
[0030] Furthermore, the filtering parameters include the filtering inductance.
[0031] Furthermore, the fitness function for evaluating the cost of an active power filter is positively correlated with the inductance of the active power filter.
[0032] Furthermore, the control parameters include the proportional coefficient and integral coefficient of the current inner loop PI controller of the active power filter converter.
[0033] Furthermore, the fitness functions for evaluating the performance, response speed, stability, and cost of the active power filter are weighted and summed to obtain the overall fitness function of the optimization algorithm.
[0034] The beneficial effects of this invention are as follows: This invention combines simulation and optimization algorithm, and sets the fitness function of the optimization algorithm to consider four aspects: filter cost, response speed, performance and stability. The fitness function of the optimization algorithm optimizes the filtering parameters and control parameters of the active power filter, and uses the final optimization result as the parameters of the active power filter, so that the active power filter can take into account the filter cost, response speed, performance and stability. Attached Figure Description
[0035] Figure 1 This is a schematic diagram of the structure and control of the nonlinear load-parallel active power filter compensation system of the present invention; Figure 2 This is a flowchart of the active power filter parameter determination method of the present invention; Figure 3This is a diagram showing the iterative optimization results of the active power filter parameters using the present invention; Figure 4 The diagram shows the simulation results of the compensation system using the optimal active power filter parameters determined by this invention. Detailed Implementation
[0036] The technical concept of this invention is as follows: This invention combines simulation and optimization algorithm, and sets the fitness function of the optimization algorithm to obtain the total fitness function from aspects such as filter cost, response speed, and total distortion rate of grid-side current (performance), so that the APF parameters obtained by the final optimization can take into account filter cost, response speed, and total distortion rate of grid-side current (performance).
[0037] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.
[0038] Implementation methods for active power filters: The active power filter of this invention is a highly efficient harmonic dynamic compensation device used to detect and compensate for grid harmonics and reactive current in real time. In this embodiment, a nonlinear load with a parallel active power filter compensation system is used as an example to illustrate the active power filter of the invention.
[0039] Nonlinear load with parallel active power filter compensation system, such as Figure 1 As shown, the system includes a three-phase AC power grid, an AC / DC rectifier, a DC load, and the active power filter of this invention. The three-phase AC power grid converts AC power into DC power through the AC / DC rectifier to supply power to the DC load. The active power filter is connected in parallel with the three-phase AC power grid and compensates for harmonics and reactive current. The structure and control method of the active power filter are the same as those in existing technologies.
[0040] Figure 1 middle, L s For the inductance within the power grid, R s The resistance within the power grid, R d For DC load resistors, C This is a capacitor supporting the DC side of the inverter. L e Here, n is the inverter filter inductor, and n is the midpoint of the power grid. V dc DC side voltage i sx For three-phase grid side current, i dx For AC load current, i cxTo compensate for current in an active power filter (APF), i cd and i cq These represent the compensation currents along the d-axis and q-axis, respectively. u gx This is the three-phase grid voltage. u px This is the voltage at the common coupling point. i cx * This is the reference value for the three-phase compensation current. i cd * and i cq * These represent the compensation current setpoints for the d-axis and q-axis, respectively. V dcref This represents the DC-side voltage reference value for the inverter. PWM stands for Pulse Width Modulation. 3s-2r represents the coordinate system transformation from three-phase stationary to two-phase rotating, and 2r-3s represents the coordinate system transformation from two-phase rotating to three-phase stationary. x=a,b,c represent phases a, b, and c, respectively.
[0041] The system parameters of the nonlinear load-parallel active power filter compensation system are shown in Table 1.
[0042] Table 1. System Parameter Table parameters numerical value parameters numerical value <![CDATA[The effective value of the grid line voltage ( u gx_l )]]> 269V <![CDATA[DC load resistance( R d )]]> 10Ω <![CDATA[DC side voltage ( V dc )]]> 900 V <![CDATA[Internal impedance of the power grid ( R s , L s )]]> 0.1Ω,0.1mH An active power filter consists of a filter inductor and a converter. The converter in an active power filter employs dual closed-loop control with an outer voltage loop and an inner current loop, where the active current reference value... i cd * and reactive current reference value i cq * Harmonic current before the AC / DC rectifier i cx * The actual value of active current obtained after a coordinate system transformation from three-phase stationary to two-phase rotating (3s-2r). i cd and actual value of reactive current i cq To compensate the output current of the active power filter to the AC power grid i cx The coordinate system is obtained by transforming from a three-phase stationary system to a two-phase rotating system using a 3s-2r transformation.
[0043] The parameters of the active power filter of the present invention are obtained by the following method.
[0044] Step 1: Construct a simulation model of the active power filter (APF) closed-loop grid-connected system.
[0045] In this embodiment, the aforementioned nonlinear load-parallel active power filter compensation system is used as an example for explanation, and its simulation model is similarly as described above. Figure 1 As shown.
[0046] Based on the system parameters of the APF closed-loop grid-connected system, the parameters of the active power filter are determined, including filtering parameters and control parameters. In this embodiment, the filtering parameters include the filter inductance. L e The control parameters include the proportional coefficient of the current inner loop PI regulator. K p Integral coefficient K i .
[0047] In this embodiment, the particle swarm optimization algorithm is used. Other iterative optimization algorithms may also be used in other embodiments.
[0048] Step 2: An optimization algorithm is used to iteratively optimize the parameters of the active power filter to obtain the optimal active filter parameters. Specific steps are as follows: Figure 2 As shown, it includes: S1-a: Initialize system parameters, particle swarm parameters, and community.
[0049] S1-b: Establish the performance evaluation index and fitness function of the active power filter. In this embodiment, the specific performance evaluation index is the grid current distortion rate (THD) obtained using the simulation model in step 1. ig If it is less than the set distortion rate D m Then its fitness function F THD The value is 0 if it is not 1 otherwise. In this embodiment, the distortion rate is set to 1.5%.
[0050] S1-c: Establish the evaluation index and fitness function corresponding to the response time of the active power filter. In this embodiment, the evaluation index corresponding to the response time is the settling time obtained using the simulation model in step 1. t s If it is less than the set adjustment time t m Then its corresponding fitness function F t The value is 0 if it is not 1 otherwise. In this embodiment, an adjustment time is set. t m It takes 35ms.
[0051] S1-d: Construct the evaluation index and fitness function corresponding to the stability of the active power filter. In this embodiment, the evaluation index corresponding to stability is the stability margin. The fitness function is determined when the following conditions are met. F s The value is 0 otherwise the value is 1. [ f ( oh c [+180°] greater than f min and less than f max ,in f ( oh c To compensate the system at the crossover frequency oh c Phase at that point, f min To compensate the system at the crossover frequency oh c The minimum set phase at that location, f max To compensate the system at the crossover frequency oh c The maximum set phase at that location. In this embodiment, f min =30°, f max =60°.
[0052] S1-e: Constructing the fitness function for the cost of the active power filter. In this embodiment, the fitness function for the cost of the active power filter is: F cost = L e In the formula, F cost Let the fitness function be the cost of the active power filter. L e This is a filter inductor.
[0053] S1-f: Based on the fitness functions of the active power filter's performance, response time, stability, and cost, the overall fitness function of the active power filter is constructed. In this embodiment, the overall fitness function is as follows: F total = F THD + F t + F s + F cost In the formula Ftotal The overall fitness function is the sum of the individual fitness functions in this embodiment. In other embodiments, the weights of the individual fitness functions may be unequal.
[0054] S1-g: Update the individual's optimal solution and the value of the individual's optimal fitness function.
[0055] S1-h: Update the global optimal solution and the global optimal fitness function value.
[0056] S1-i: Determine if the set number of iterations has been reached. If it has, execute S2; otherwise, execute s1-b.
[0057] S2: Output filter inductor L e PI controller parameters K p and K i The optimal active power filter parameters were obtained.
[0058] After each iteration, a simulation model is used to simulate the filtering and control parameters of the iterated active power filter, obtaining the grid current waveform, thereby evaluating the performance, response speed, and stability of the active power filter. In this embodiment, 80 iterations are used as an example. The change in filter inductance with the number of iterations during the iteration process is as follows: Figure 3 As shown in (a), the PI controller parameters K p The changes with the number of iterations are as follows: Figure 3 As shown in (b), the PI controller parameters K i The variation with the number of iterations is as follows: Figure 3 As shown in (c), the value of the total fitness function changes with the number of iterations as follows: Figure 3 As shown in (d).
[0059] Depend on Figure 3 Therefore, in this embodiment, the filter inductance of the active power filter is determined to be Le = 0.12mH, and the PI regulator parameters are as follows: K p =11.51、 K i =332.
[0060] Simulation verification of the determined active power filter parameters yields the following results: Figure 4The simulation results shown (the left image is the simulated waveform of the grid-side current output of the compensation system, and the right image is a schematic diagram of the distortion rate calculation) show a total harmonic distortion rate of 1.41%, which is less than the set distortion rate. The adjustment time... t s =30ms, less than the set adjustment time t m =40ms. Therefore, the parameter determination method for the active power filter of the present invention enables the parameters of the active power filter to simultaneously meet the requirements of performance, settling time, stability, and cost.
[0061] Implementation method for determining parameters of active power filters: The active power filter parameter determination method of the present invention includes the following steps: Step 1: Construct a simulation model of the active power filter (APF) closed-loop grid-connected system.
[0062] In this embodiment, the aforementioned nonlinear load-parallel active power filter compensation system is used as an example for explanation, and its simulation model is similarly as described above. Figure 1 As shown.
[0063] Based on the system parameters of the APF closed-loop grid-connected system, the parameters of the active power filter are determined, including filtering parameters and control parameters. In this embodiment, the filtering parameters include the filter inductance. L e The control parameters include the proportional coefficient of the current inner loop PI regulator. K p Integral coefficient K i .
[0064] In this implementation, the optimization algorithm is the particle swarm optimization algorithm.
[0065] Step 2: An optimization algorithm is used to iteratively optimize the parameters of the active power filter to obtain the optimal active filter parameters. Specific steps are as follows: Figure 2 As shown, it includes: S1-a: Initialize system parameters, particle swarm parameters, and community.
[0066] S1-b: Establish the performance evaluation index and fitness function of the active power filter. In this embodiment, the specific performance evaluation index is the grid current distortion rate (THD) obtained using the simulation model in step 1. ig If it is less than the set distortion rate D m Then its fitness function F THD The value is 0 if it is not 1 otherwise. In this embodiment, the distortion rate is set to 1.5%.
[0067] S1-c: Establish the evaluation index and fitness function corresponding to the response time of the active power filter. In this embodiment, the evaluation index corresponding to the response time is the settling time obtained using the simulation model in step 1. t s If it is less than the set adjustment time t m Then its corresponding fitness function F t The value is 0 if it is not 1 otherwise. In this embodiment, an adjustment time is set. t m It takes 35ms.
[0068] S1-d: Construct the evaluation index and fitness function corresponding to the stability of the active power filter. In this embodiment, the evaluation index corresponding to stability is the stability margin. The fitness function is determined when the following conditions are met. F s The value is 0 otherwise the value is 1. [ f ( oh c [+180°] greater than f min and less than f max ,in f ( oh c To compensate the system at the crossover frequency oh c Phase at that point, f min To compensate the system at the crossover frequency oh c The minimum set phase at that location, f max To compensate the system at the crossover frequency oh c The maximum set phase at that location. In this embodiment, f min =30°, f max =60°.
[0069] S1-e: Constructing the fitness function for the cost of the active power filter. In this embodiment, the fitness function for the cost of the active power filter is: F cost = L e In the formula, F cost Let the fitness function be the cost of the active power filter. L e This is a filter inductor.
[0070] S1-f: Based on the fitness functions of the active power filter's performance, response time, stability, and cost, the overall fitness function of the active power filter is constructed. In this embodiment, the overall fitness function is as follows: F total = F THD + F t + F s + F cost In the formula F total The overall fitness function is the sum of the individual fitness functions in this embodiment. In other embodiments, the weights of the individual fitness functions may be unequal.
[0071] S1-g: Update the individual's optimal solution and the value of the individual's optimal fitness function.
[0072] S1-h: Update the global optimal solution and the global optimal fitness function value.
[0073] S1-i: Determine if the set number of iterations has been reached. If it has, execute S2; otherwise, execute s1-b.
[0074] S2: Output filter inductor L e PI controller parameters K p and K i The optimal active power filter parameters were obtained.
[0075] In this embodiment, the number of iterations is set to 80 as an example. During the iteration process, the change of the filter inductance with the number of iterations is as follows: Figure 3 As shown in (a), the PI controller parameters K p The variation with the number of iterations is as follows: Figure 3 As shown in (b), the PI controller parameters K i The variation with the number of iterations is as follows: Figure 3 As shown in (c), the value of the total fitness function changes with the number of iterations as follows: Figure 3 As shown in (d).
[0076] Depend on Figure 3 Therefore, in this embodiment, the filter inductance of the active power filter is determined to be Le = 0.12mH, and the PI regulator parameters are as follows:K p =11.51、 K i =332.
[0077] The parameter determination method for the active power filter of the present invention enables the parameters of the active power filter to simultaneously meet the requirements of performance, settling time, stability and cost.
[0078] Implementation of computer-readable storage media: The present invention provides a computer-readable storage medium having a computer program stored therein, wherein the computer program, when executed, implements the active power filter parameter determination method of the present invention, the method comprising the following steps: Step 1: Construct a simulation model of the active power filter (APF) closed-loop grid-connected system.
[0079] In this embodiment, the aforementioned nonlinear load-parallel active power filter compensation system is used as an example for explanation, and its simulation model is similarly as described above. Figure 1 As shown.
[0080] Based on the system parameters of the APF closed-loop grid-connected system, the parameters of the active power filter are determined, including filtering parameters and control parameters. In this embodiment, the filtering parameters include the filter inductance. L e The control parameters include the proportional coefficient of the current inner loop PI regulator. K p Integral coefficient K i .
[0081] In this implementation, the optimization algorithm is the particle swarm optimization algorithm.
[0082] Step 2: An optimization algorithm is used to iteratively optimize the parameters of the active power filter to obtain the optimal active filter parameters. Specific steps are as follows: Figure 2 As shown, it includes: S1-a: Initialize system parameters, particle swarm parameters, and community.
[0083] S1-b: Establish the performance evaluation index and fitness function of the active power filter. In this embodiment, the specific performance evaluation index is the grid current distortion rate (THD) obtained using the simulation model in step 1. ig If it is less than the set distortion rate D m Then its fitness function F THD The value is 0 if it is not 1 otherwise. In this embodiment, the distortion rate is set to 1.5%.
[0084] S1-c: Establish the evaluation index and fitness function corresponding to the response time of the active power filter. In this embodiment, the evaluation index corresponding to the response time is the settling time obtained using the simulation model in step 1. t s If it is less than the set adjustment time t m Then its corresponding fitness function F t The value is 0 if it is not 1 otherwise. In this embodiment, an adjustment time is set. t m It takes 35ms.
[0085] S1-d: Construct the evaluation index and fitness function corresponding to the stability of the active power filter. In this embodiment, the evaluation index corresponding to stability is the stability margin. The fitness function is determined when the following conditions are met. F s The value is 0 otherwise the value is 1. [ f ( oh c [+180°] greater than f min and less than f max ,in f ( oh c To compensate the system at the crossover frequency oh c Phase at that point, f min To compensate the system at the crossover frequency oh c The minimum set phase at that location, f max To compensate the system at the crossover frequency oh c The maximum set phase at that location. In this embodiment, f min =30°, f max =60°.
[0086] S1-e: Constructing the fitness function for the cost of the active power filter. In this embodiment, the fitness function for the cost of the active power filter is: F cost = L e In the formula, F cost Let the fitness function be the cost of the active power filter. L e This is a filter inductor.
[0087] S1-f: Based on the fitness functions of the active power filter's performance, response time, stability, and cost, the overall fitness function of the active power filter is constructed. In this embodiment, the overall fitness function is as follows: F total = F THD + F t + F s + F cost In the formula F total The overall fitness function is the sum of the individual fitness functions in this embodiment. In other embodiments, the weights of the individual fitness functions may be unequal.
[0088] S1-g: Update the individual's optimal solution and the value of the individual's optimal fitness function.
[0089] S1-h: Update the global optimal solution and the global optimal fitness function value.
[0090] S1-i: Determine if the set number of iterations has been reached. If it has, execute S2; otherwise, execute s1-b.
[0091] S2: Output filter inductor L e PI controller parameters K p and K i The optimal active power filter parameters were obtained.
[0092] In this embodiment, the number of iterations is set to 80 as an example. During the iteration process, the change of the filter inductance with the number of iterations is as follows: Figure 3 As shown in (a), the PI controller parameters K p The variation with the number of iterations is as follows: Figure 3 As shown in (b), the PI controller parameters K i The variation with the number of iterations is as follows: Figure 3 As shown in (c), the value of the total fitness function changes with the number of iterations as follows: Figure 3 As shown in (d).
[0093] Depend on Figure 3 Therefore, in this embodiment, the filter inductance of the active power filter is determined to be Le = 0.12mH, and the PI regulator parameters are as follows:K p =11.51、 K i =332.
[0094] The parameter determination method for the active power filter of the present invention enables the parameters of the active power filter to simultaneously meet the requirements of performance, settling time, stability and cost.
Claims
1. A method for determining the parameters of an active power filter, characterized in that, Includes the following steps: 1) Construct a simulation model of the compensation system, which includes an active power filter, a power grid compensated by the active power filter, and a load supplied by the power grid; 2) With the optimization objectives of achieving the best performance, fastest response speed, strongest stability, and lowest cost for the active power filter, an iterative optimization algorithm is used to iteratively optimize the filtering parameters and control parameters of the active power filter. The optimal solution obtained is the parameter determination result of the active power filter. During the iterative optimization process, after each iteration, the filtering parameters and control parameters of the active power filter after the iteration are simulated using a simulation model to obtain the grid current waveform, thereby evaluating the performance, response speed, and stability of the active power filter.
2. The method for determining the parameters of an active power filter according to claim 1, characterized in that, The performance of an active power filter is evaluated using the distortion rate of the grid current waveform. Correspondingly, the fitness function for evaluating the performance of an active power filter is: F THD =1,THD ig > D m F THD =0,THD ig ≤ D m In the formula, F THD The fitness function used to evaluate the performance of active power filters is THD. ig The distortion rate of the grid current waveform. D m To set the distortion rate.
3. The method for determining the parameters of an active power filter according to claim 1, characterized in that, The response speed of an active power filter is evaluated using the settling time of the grid current. Correspondingly, the fitness function for evaluating the response speed of an active power filter is: F t =1， t s ＞ t m F t =0， t s ≤ t m In the formula, F t To evaluate the fitness function of the settling time of an active power filter, t s The regulation time of the grid current. t m To set the adjustment time.
4. The method for determining the parameters of an active power filter according to claim 1, characterized in that, The stability of an active power filter is evaluated using phase margin. Correspondingly, the fitness function for evaluating the stability of an active power filter is: F s =0， φ min ≤ φ ( ω c )+180°≤ φ max F s =1, φ ( ω c +180° < φ min or φ ( ω c +180°> φ max In the formula, F s The fitness function is used to evaluate the stability of active power filters. φ ( ω c ) represents the grid current at the crossover frequency ω c Phase at that point, φ min To compensate the system at the crossover frequency ω c The minimum set phase at that location, φ max To compensate the system at the crossover frequency ω c The maximum set phase at that location.
5. The method for determining the parameters of an active power filter according to claim 1, characterized in that, The filtering parameters include the filter inductance.
6. The method for determining the parameters of an active power filter according to claim 5, characterized in that, The fitness function for evaluating the cost of an active power filter is positively correlated with the inductance of the active power filter.
7. The method for determining the parameters of an active power filter according to claim 1, characterized in that, The control parameters include the proportional and integral coefficients of the inner loop PI controller of the active power filter converter.
8. The method for determining the parameters of an active power filter according to any one of claims 1-7, characterized in that, The fitness functions for evaluating the performance, response speed, stability, and cost of the active power filter are weighted and summed to obtain the overall fitness function of the optimization algorithm.
9. An active power filter, characterized in that, The control parameters and filtering parameters of the active power filter are designed using the parameter determination method for the active power filter as described in any one of claims 1-8.
10. A computer-readable storage medium having a computer program stored internally, characterized in that, When the computer program is executed, it implements the parameter determination method for the active power filter as described in any one of claims 1-8.