Wind power plant electric energy quality optimization method and system based on improved particle swarm optimization

Abstract

The invention belongs to the technical field of new energy power generation and grid connection, and particularly relates to a wind power plant electric energy quality optimization method and system based on an improved particle swarm algorithm. Aiming at the problems of voltage fluctuation, frequency deviation and high harmonic content during grid connection of a wind power plant, a wind power plant simulation model is built through Matlab / Simulink, a multi-target fitness function is built according to the voltage deviation, the frequency deviation and the total harmonic distortion (THD), the self-adaptive inertia weight and learning factors of a particle swarm optimization algorithm are improved, and a wind power plant grid connection model is built. And dynamically adjusting voltage adjustment gain, reactive compensation amount and filter parameters. According to the embodiment, compared with a traditional method, the number of iterations is reduced by 40%, the voltage fluctuation suppression ratio is improved by 10%, the frequency adjusting time is shortened to be within 0.3 second, the THD is reduced to 2.1% from 5.2%, and the power grid stability and the electric energy quality are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of renewable energy power generation and grid connection, and particularly relates to a method and system for optimizing wind farm power quality based on an improved particle swarm algorithm. Background Art

[0002] Power quality is a key indicator of power system reliability and economic efficiency. It primarily encompasses voltage quality, frequency stability, and total harmonic distortion (THD). Voltage quality directly impacts user device performance and safety, frequency stability directly influences power system efficiency, and THD reflects the purity of the power supply, directly impacting the grid's power factor and equipment lifespan.

[0003] With the increasing complexity of modern power systems and the widespread use of renewable energy, power quality issues are becoming increasingly prominent. This is especially true in situations where nonlinear loads are concentrated, such as wind farms. Harmonic pollution, voltage fluctuations, and frequency deviations are serious issues, severely impacting the stable operation of the power grid and the proper functioning of user equipment.

[0004] Therefore, optimizing power quality has become an important issue in power system design and operation. By optimizing power quality, we can improve power supply reliability, reduce energy waste, and minimize equipment wear, while also contributing to the goal of achieving smart grids.

[0005] In existing technologies, the following technical solutions are mainly used to optimize the power quality of wind farms connected to the grid: Parameter optimization of traditional particle swarm optimization (PSO): The PSO algorithm is used for static optimization of wind farm control parameters, such as voltage regulation gain and reactive power compensation capacity. Its basic process includes initializing the particle swarm, iteratively updating particle positions and velocities, and evaluating the optimal solution based on a fitness function. Fitness functions typically optimize for a single objective (such as voltage deviation or frequency fluctuation) and lack a multi-objective collaborative optimization mechanism. Fixed parameter control strategies utilize a proportional-integral (PI) controller, whose parameters (such as the proportional coefficient and integral time constant) are pre-set before system operation and lack dynamic adjustment capabilities. LCL filters are used for harmonic suppression, but the filter parameters (inductance and capacitance) are designed for fixed operating conditions and cannot adapt to wind speed fluctuations or grid load changes.

[0006] Therefore, the inherent defects of the traditional PSO algorithm are: (1) Slow convergence speed: The inertia weight and learning factor are fixed, resulting in low global search efficiency in the early stage of the iteration and easy to fall into local optimality in the later stage. (2) Insufficient multi-objective optimization capability: A single fitness function is difficult to optimize multiple indicators such as voltage, frequency and harmonics at the same time, resulting in one-sided optimization results. Poor adaptability of fixed parameter control: (1) The preset PI control parameters cannot be dynamically adjusted according to the real-time status of the power grid (such as sudden changes in wind speed and load fluctuations), resulting in voltage / frequency regulation lag and poor dynamic response performance; (2) The fixed parameter design of the filter is difficult to cope with the dynamic changes of the harmonic spectrum, and the total harmonic distortion (THD) suppression effect is limited. Therefore, there is an urgent need for an adaptive multi-objective optimization method to improve the dynamic response and harmonic suppression capabilities of wind farms. Summary of the Invention

[0007] The present invention provides a wind farm power quality optimization method and system based on an improved particle swarm algorithm to solve the technical problems existing in the prior art, such as slow convergence of the traditional particle swarm algorithm, insufficient multi-objective optimization capability, poor dynamic response of the fixed parameter control strategy, and limited harmonic suppression effect.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions: A power quality optimization control method based on an improved particle swarm optimization algorithm comprises the following steps: S1: Build a wind farm simulation model and obtain the initial voltage deviation, frequency deviation, and total harmonic distortion of the power grid based on the simulation model; S2: Based on the initial voltage deviation, frequency deviation and total harmonic distortion, the initial fitness is calculated using a multi-objective fitness function; S3: Based on the particle swarm optimization algorithm, the adaptive inertia weight and learning factor are improved to randomly generate the initial control parameters of the power grid; S4: According to the initial control parameters of the power grid, the simulation model is controlled to obtain the current voltage deviation, frequency deviation and total harmonic distortion. According to the current voltage deviation, frequency deviation and total harmonic distortion, the current fitness and current control parameters are calculated; S5: Preset the maximum number of iterations of the fitness. If the current number of iterations of the fitness is equal to the maximum number of iterations, output the grid control parameters corresponding to the current fitness; if the current number of iterations of the fitness does not reach the maximum number of iterations, re-enter step S3.

[0009] Based on the initial voltage deviation, frequency deviation and total harmonic distortion, the initial fitness is calculated using a multi-objective fitness function, and the multi-objective fitness function is as follows:

[0010]

[0011] Where, : voltage deviation; : frequency deviation; : Total harmonic distortion; : voltage deviation dynamic weight coefficient; : frequency deviation dynamic weight coefficient; : Total harmonic distortion dynamic weight coefficient; : voltage harmonic amplitude; : Fundamental wave amplitude.

[0012] In the particle swarm algorithm, in the early stage of particle evolution, a too small inertia weight will reduce the search space; in the later stage, a too large inertia weight is not conducive to algorithm convergence. Therefore, it is necessary to reasonably improve the adaptive inertia weight. The improvement method is as follows:

[0013] Where, : initial value of adaptive weight; : The end value of the adaptive weight; : Maximum number of iterations, : Current iteration number.

[0014] In the particle swarm algorithm, by dynamically adjusting the learning factor, the algorithm can have different search characteristics at different stages. The improvement method is as follows:

[0015]

[0016] Where, : Individual learning factor; : Social learning factor; : initial value of the learning factor, : Final value of the learning factor.

[0017] According to the initial control parameters of the power grid, the simulation model is controlled to obtain the current voltage deviation, frequency deviation and total harmonic distortion. Specifically, the initial control parameters of the power grid include the initial voltage regulation gain , initial reactive compensation , initial filter inductance and the initial filter capacitance , adjust the gain according to the initial voltage , initial reactive compensation , initial filter inductance and the initial filter capacitance , perform preliminary regulation of the simulation model on voltage, frequency and filter inductance and capacitance, and then use the SCADA system to collect the current voltage deviation, current frequency deviation and current total harmonic distortion of the simulation model after preliminary regulation in real time.

[0018] The current control parameters are adjusted according to the current voltage deviation and the initial voltage gain. , calculate the voltage control parameters as follows:

[0019] In the formula ， : voltage deviation threshold; : Current voltage deviation; : Voltage regulation gain parameter; : Current voltage regulation gain.

[0020] Among the current control parameters, according to the current frequency deviation and the initial reactive compensation amount , calculate the frequency control parameters, specifically:

[0021] Where, : Current frequency deviation; : Frequency control gain parameter; : Current reactive compensation amount.

[0022] Among the current control parameters, according to the current total harmonic distortion and the initial filter inductance , calculate the filter inductance control parameters, specifically:

[0023] Where, : Current total harmonic distortion; : Filter inductance control gain parameter; : Current filter inductance.

[0024] In the current control parameters, according to the current total harmonic distortion and the initial filter capacitance , calculate the filter capacitance control parameters, specifically:

[0025] Where, : Current total harmonic distortion; : filter electrolytic control gain parameter; : Current filter capacitance.

[0026] A power quality optimization control system based on an improved particle swarm algorithm includes a simulation initialization module, an iterative initialization module, an optimization module, an iterative module, and an optimal output module; The simulation initialization module is used to build a wind farm simulation model and obtain the initial voltage deviation, frequency deviation and total harmonic distortion of the power grid based on the simulation model; The iterative initial module is used to calculate the initial fitness using a multi-objective fitness function based on the initial voltage deviation, frequency deviation and total harmonic distortion; The optimization module is used to improve the adaptive inertia weight and learning factor based on the particle swarm algorithm and randomly generate the initial control parameters of the power grid; The iterative module is used to control the simulation model according to the initial control parameters of the power grid, obtain the current voltage deviation, frequency deviation and total harmonic distortion, and calculate the current fitness and current control parameters based on the current voltage deviation, frequency deviation and total harmonic distortion; The optimal output module is used to preset the maximum number of iterations of fitness. If the current number of fitness iterations is equal to the maximum number of iterations, the grid control parameter corresponding to the current fitness is output; if the current number of fitness iterations does not reach the maximum number of iterations, the iteration initial module is re-entered.

[0027] Compared with the prior art, the present invention has the following beneficial effects: This paper discloses a power quality optimization and control method based on an improved particle swarm algorithm. This method uses voltage deviation, frequency deviation, and THD as comprehensive optimization targets. It achieves multi-target collaborative optimization through dynamic weighting coefficients. This method addresses the drawback of traditional single-target optimization, which focuses solely on a single power quality indicator. By adjusting the weighting coefficients, optimization targets can be flexibly configured based on grid requirements, improving system adaptability.

[0028] By using a particle swarm optimization algorithm, the adaptive inertia weight is improved. Compared to traditional fixed inertia weights, this method expands the search range in the early stages of iteration to avoid being trapped in local optima; in the later stages, it focuses on refined search to improve convergence speed. Furthermore, the learning factor is dynamically adjusted. Initially, a larger learning factor encourages particles to independently explore new areas, enhancing algorithm diversity. Later, the social factor dominates, prompting particles to converge toward the global optimal solution and accelerating convergence. This addresses the problem of slow convergence in the early stages and volatile fluctuations in the later stages of traditional PSO, which is caused by a fixed learning factor, and improves optimization efficiency.

[0029] The parameters improved by the particle swarm optimization algorithm are used to calculate the current control parameters to dynamically adjust the adjustment strategy. The frequency static error is eliminated through integral control, and the filter parameters adapt to the harmonic changes in real time, solving the problem that traditional fixed parameter filters are unable to suppress time-varying harmonics.

[0030] Furthermore, the present invention implements a closed-loop process of "parameter adjustment → simulation → data acquisition → fitness evaluation → parameter update" and repeatedly iterates to the maximum number of times to achieve adaptive optimization of parameters, so that the control strategy continuously matches the real-time status of the power grid. Compared with traditional open-loop control, this method dynamically corrects parameters through a feedback mechanism to improve system robustness. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is an optimization flowchart of the improved particle swarm algorithm in an embodiment of the present invention; Figure 2 A comparison diagram of fitness function values ​​in an embodiment of the present invention; Figure 3 A comparison diagram of voltage deviation adjustment time in an embodiment of the present invention; Figure 4 A frequency adjustment time comparison diagram in an embodiment of the present invention; Figure 5 Schematic diagram of a power quality optimization control method according to an embodiment of the present invention; Figure 6 This is a module diagram of the power quality optimization control system in an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to further understand the content of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the embodiments are only for explaining the present invention and are not intended to limit it.

[0033] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0034] Example 1 according to Figure 5 The schematic diagram of the power quality optimization control method based on the improved particle swarm algorithm is shown. This embodiment proposes a power quality optimization control method based on the improved particle swarm algorithm, which specifically includes the following steps: S1: Build a wind farm simulation model and obtain the initial voltage deviation, frequency deviation, and total harmonic distortion of the power grid based on the simulation model; S2: Based on the initial voltage deviation, frequency deviation and total harmonic distortion, the initial fitness is calculated using a multi-objective fitness function; S3: Based on the particle swarm algorithm, the adaptive inertia weight and learning factor are improved to randomly generate the initial control parameters of the power grid; S4: According to the initial control parameters of the power grid, the simulation model is controlled to obtain the current voltage deviation, frequency deviation and total harmonic distortion. According to the current voltage deviation, frequency deviation and total harmonic distortion, the current fitness and current control parameters are calculated; S5: Preset the maximum number of iterations of the fitness. If the current number of iterations of the fitness is equal to the maximum number of iterations, output the grid control parameters corresponding to the current fitness; if the current number of iterations of the fitness does not reach the maximum number of iterations, re-enter step S3.

[0035] Example 2 Based on the power quality optimization control method based on the improved particle swarm algorithm proposed in Example 1, this embodiment proposes a power quality optimization control system based on the improved particle swarm algorithm, such as Figure 6 As shown, the steps of the power quality optimization control method based on the improved particle swarm algorithm described in Example 1 are implemented through the simulation initialization module, the iterative initial module, the optimization module, the iterative module and the optimal output module, specifically as follows: The simulation initialization module is used to build a wind farm simulation model and obtain the initial voltage deviation, frequency deviation and total harmonic distortion of the power grid based on the simulation model; The iterative initial module is used to calculate the initial fitness using a multi-objective fitness function based on the initial voltage deviation, frequency deviation and total harmonic distortion; The optimization module is used to improve the adaptive inertia weight and learning factor based on the particle swarm optimization algorithm and randomly generate the initial control parameters of the power grid; The iterative module is used to control the simulation model according to the initial control parameters of the power grid, obtain the current voltage deviation, frequency deviation and total harmonic distortion, and calculate the current fitness and current control parameters based on the current voltage deviation, frequency deviation and total harmonic distortion; The optimal output module is used to preset the maximum number of iterations of fitness. If the current number of fitness iterations is equal to the maximum number of iterations, the grid control parameter corresponding to the current fitness is output; if the current number of fitness iterations does not reach the maximum number of iterations, the iteration initial module is re-entered.

[0036] Example 3 This embodiment builds a wind farm simulation model based on the Matlab / Simulink platform to complete the power quality optimization control method based on the improved particle swarm algorithm proposed in Example 1, as shown in FIG. Figure 1 As shown, the specific implementation method is as follows: Based on the actual electric field, a wind farm simulation model was constructed in the simulation platform integrated with Matlab and Simulink. The main modules include the following: simulating the power output characteristics of the wind turbine by simulating the doubly fed wind turbine; simulating the LCL filter to filter out the high-frequency harmonics generated by the converter in the simulation model; and simulating the SCADA data acquisition system to monitor the grid voltage, frequency, and harmonic data of the simulation model in real time. After the simulation model is constructed, the wind farm simulation is carried out. After the preliminary simulation, the SCADA data acquisition system is used to collect the initial grid voltage deviation in the wind farm simulation model. , initial frequency and initial filter inductance and initial filter capacitors , SCADA data acquisition system for initial voltage , initial frequency and initial filter inductance and initial filter capacitors And the rated voltage, frequency and filter capacitor inductance are calculated to obtain the initial voltage deviation , initial frequency deviation and initial total harmonic distortion SCADA data acquisition system based on voltage deviation , frequency deviation and total harmonic distortion To optimize the target, a multi-objective fitness function is constructed, as shown in the following formula (1) and formula (2): (1) (2) Where, : voltage deviation; : frequency deviation; : Total harmonic distortion; : voltage deviation dynamic weight coefficient; : is the dynamic weight coefficient of frequency deviation; : Total harmonic distortion dynamic weight coefficient; : voltage harmonic amplitude; : Fundamental wave amplitude. It is used to evaluate the comprehensive optimization effect of power quality. It combines three key indicators (voltage deviation ΔU, frequency deviation Δf, and total harmonic distortion THD) into a scalar value through weighted summation, which serves as the optimization target of the particle swarm algorithm. The smaller FF is, the closer the voltage and frequency are to the rated values, the lower the harmonic distortion is, and the better the power quality is. In this embodiment, the initial voltage deviation is used to calculate the power quality of the system. , initial frequency deviation and initial total harmonic distortion , using the above multi-objective fitness function, calculate the initial fitness , at this time, the voltage deviation dynamic weight coefficient , frequency deviation dynamic weight coefficient , dynamic weight coefficient of total harmonic distortion , the number of fitness iterations after the initial calculation is 1, and the maximum number of fitness iterations preset in this embodiment is 100.

[0037] In the further optimization scheme of the present invention, the particle swarm optimization algorithm is used to improve the adaptive inertia weight and learning factor, and randomly generate the initial control parameters of the power grid; at the same time, the particle swarm optimization algorithm is configured with parameters, which are as follows: the particle swarm size is 100, the initial value of the adaptive weight is , the end value of the adaptive weight , the initial value of the learning factor .

[0038] In the particle swarm algorithm, in the early stages of particle evolution, a too small inertia weight will reduce the search space; in the later stages, a too large inertia weight is not conducive to algorithm convergence. Therefore, it is necessary to reasonably improve the adaptive inertia weight to enhance the global search capability and avoid local optimality. The improvement method is as follows:

[0039] Where, : initial value of adaptive weight; : The end value of the adaptive weight; : Maximum number of iterations, : Current iteration number.

[0040] In the particle swarm algorithm, by dynamically adjusting the learning factor, the algorithm has global search capabilities in the early stage, searches the entire optimization space, and converges to the global minimum as quickly as possible in the later stage, accelerating the convergence speed to achieve different search characteristics at different stages. The improvement method is as follows:

[0041]

[0042] Where, : Individual learning factor; : Social learning factor; : initial value of the learning factor, : Final value of the learning factor.

[0043] Through the above dynamic adjustment method, the adaptive inertia weight and learning factor are improved to change the speed and displacement of the particles. The initial position and speed of 100 particles are randomly generated in the search space. According to the initial position of the particles, the initial control parameters of the power grid are generated. The control parameters of the power grid accident include: initial voltage regulation gain , initial reactive compensation , initial filter inductance and the initial filter capacitance , adjust the gain according to the initial voltage , initial reactive compensation , initial filter inductance and the initial filter capacitance , perform preliminary regulation of the voltage, frequency and filter inductance and capacitance of the simulation model, and then use the SCADA system to collect the current voltage after the preliminary regulation of the simulation model in real time , current frequency And the current filter inductor and the current filter capacitor , by the current voltage , current frequency And the current filter inductor and the current filter capacitor And the rated voltage, frequency and filter capacitor inductance are calculated to obtain the current voltage deviation , Current frequency deviation And the current total harmonic distortion . According to the current voltage deviation , Current frequency deviation And the current total harmonic distortion , calculate the current fitness, the number of fitness iterations is increased by one, that is, 2; at the same time, according to the current frequency deviation And the current total harmonic distortion , calculate the current control parameters, specifically: according to the current voltage deviation and the initial voltage regulation gain , calculate the voltage control parameters as follows:

[0044] In the formula ， : voltage deviation threshold; :The current voltage deviation is ; : Voltage regulation gain parameter, in this embodiment ; : Current voltage regulation gain.

[0045] According to the current frequency deviation And the initial reactive compensation , calculate the frequency control parameters, specifically:

[0046] Where, : Current frequency deviation, which is ; : Frequency control gain parameter, in this embodiment ; : Current reactive compensation amount.

[0047] According to the current total harmonic distortion and the initial filter inductance , calculate the filter inductance control parameters, specifically:

[0048] Where, : The current total harmonic distortion, that is ; : The filter inductance control gain parameter, in this embodiment ; : Current filter inductance.

[0049] According to the current total harmonic distortion and the initial filter capacitance , calculate the filter capacitance control parameters, specifically:

[0050] Where, : The current total harmonic distortion, that is ; : Filter electrolytic control gain parameter, in this embodiment ; : Current filter capacitance.

[0051] After the above two iterations, the number of iterations of the fitness is 2, which does not meet the maximum number of iterations of the fitness preset in this embodiment. Therefore, the adaptive inertia weight and the learning factor are improved again, and the initial control parameters of the power grid in the next iteration are randomly generated again, and the current fitness and the current control parameters in the next iteration are recalculated; until the number of iterations of the fitness is equal to 100, the current number of iterations meets the maximum number of iterations, and the control parameters calculated in the current iteration are output, that is, the optimal solution for power quality control.

[0052] Furthermore, this implementation compares the above optimization control method with the traditional PSO and fixed parameter strategies in terms of convergence speed, dynamic response and harmonic suppression. Figures 2 to 4As shown in Figure 2, the fitness value (FF) of the traditional PSO decreases slowly in the early stages of iteration, and it takes about 80 iterations to reach stability, and the final FF value is about 5.58. This shows that the fixed inertia weight and learning factor of the traditional method lead to insufficient global search capabilities in the early stages, and it is easy to oscillate around the local optimal solution. However, the fitness value of the present invention decreases rapidly in the early stages of iteration, and it only takes about 50 iterations to converge to FF≈5.53, reducing the number of iterations by 40%. Figure 3 As shown in the figure, the voltage deviation of the traditional PSO takes about 0.8 seconds to stabilize within the range of ±1% of the rated value, indicating that its fixed-parameter controller cannot dynamically adjust the gain parameters according to the real-time voltage deviation, resulting in a response lag. In the present invention, the voltage deviation is stabilized within 0.4 seconds, the adjustment time is shortened by 50%, and the suppression rate is improved by 10%. Figure 4 As shown, the conventional PSO takes approximately 0.6 seconds to converge to within ±0.02 Hz of the frequency deviation. This indicates that the conventional method's fixed reactive compensation lacks integral control over the frequency deviation, failing to eliminate steady-state errors. In contrast, the present invention stabilizes the frequency within 0.3 seconds, shortening the adjustment time by 50%. While the conventional method's fixed filter has a total harmonic distortion of 5.2%, exceeding grid standards, the present invention reduces this to 2.1%, meeting grid standards.

[0053] In addition, it should be understood that although this specification describes the embodiments, not every embodiment contains only one independent technical solution. This description is for clarity only. Those skilled in the art should consider the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art. The above content is only for the purpose of illustrating the technical concept of the present invention and cannot be used to limit the scope of protection of the present invention. Any changes made based on the technical solution in accordance with the technical concept proposed by the present invention fall within the scope of protection of the claims of the present invention.

Claims

1. A power quality optimization control method based on improved particle swarm optimization algorithm, characterized in that: The following steps are involved: S1: Build a wind farm simulation model and obtain the initial voltage deviation, frequency deviation, and total harmonic distortion of the power grid based on the simulation model; S2: Based on the initial voltage deviation, frequency deviation and total harmonic distortion, the initial fitness is calculated using a multi-objective fitness function; S3: Based on the particle swarm algorithm, the adaptive inertia weight and learning factor are improved to randomly generate the initial control parameters of the power grid; S4: According to the initial control parameters of the power grid, the simulation model is controlled to obtain the current voltage deviation, frequency deviation and total harmonic distortion. According to the current voltage deviation, frequency deviation and total harmonic distortion, the current fitness and current control parameters are calculated; S5: Preset the maximum number of iterations of the fitness. If the current number of iterations of the fitness is equal to the maximum number of iterations, output the grid control parameters corresponding to the current fitness; if the current number of iterations of the fitness does not reach the maximum number of iterations, re-enter step S3.

2. The power quality optimization control method based on improved particle swarm optimization algorithm according to claim 1, characterized in that: Based on the initial voltage deviation, frequency deviation and total harmonic distortion, the initial fitness is calculated using a multi-objective fitness function, and the multi-objective fitness function is as follows: Where, : voltage deviation; : frequency deviation; : Total harmonic distortion; : voltage deviation dynamic weight coefficient; : is the dynamic weight coefficient of frequency deviation; : Total harmonic distortion dynamic weight coefficient; : voltage harmonic amplitude; : Fundamental wave amplitude.

3. The power quality optimization control method based on improved particle swarm optimization algorithm according to claim 1, characterized in that: In the particle swarm algorithm, in the early stage of particle evolution, a too small inertia weight will reduce the search space; in the later stage, a too large inertia weight is not conducive to algorithm convergence. Therefore, it is necessary to reasonably improve the adaptive inertia weight. The improvement method is as follows: Where, : initial value of adaptive weight; : The end value of the adaptive weight; : Maximum number of iterations, : Current iteration number.

4. The power quality optimization control method based on improved particle swarm optimization algorithm according to claim 3 is characterized in that: In the particle swarm algorithm, by dynamically adjusting the learning factor, the algorithm can have different search characteristics at different stages. The improvement method is as follows: Where, : Individual learning factor; : Social learning factor; : initial value of the learning factor, : Final value of the learning factor.

5. The power quality optimization control method based on improved particle swarm optimization algorithm according to claim 1, characterized in that: According to the initial control parameters of the power grid, the simulation model is controlled to obtain the current voltage deviation, frequency deviation and total harmonic distortion. Specifically, the initial control parameters of the power grid include the initial voltage regulation gain , initial reactive compensation , initial filter inductance and the initial filter capacitance , adjust the gain according to the voltage , reactive compensation , filter inductor and filter capacitors , the simulation model is preliminarily regulated in voltage, frequency and filter inductance and capacitance, and then the SCADA system is used to collect the voltage deviation, frequency deviation and total harmonic distortion of the simulation model after the preliminary regulation in real time.

6. The power quality optimization control method based on improved particle swarm optimization algorithm according to claim 5 is characterized in that: The current control parameters are adjusted according to the current voltage deviation and the initial voltage gain. , calculate the voltage control parameters as follows: In the formula ， : voltage deviation threshold; : Current voltage deviation; : Voltage regulation gain parameter; : Current voltage regulation gain.

7. The power quality optimization control method based on improved particle swarm optimization algorithm according to claim 5, characterized in that: Among the current control parameters, according to the current frequency deviation and the initial reactive compensation amount , calculate the frequency control parameters, specifically: Where, : Current frequency deviation; : Frequency control gain parameter; : Current reactive compensation amount.

8. The power quality optimization control method based on improved particle swarm optimization algorithm according to claim 5, characterized in that: Among the current control parameters, according to the current total harmonic distortion and the initial filter inductance , calculate the filter inductance control parameters, specifically: Where, : Current total harmonic distortion; : Filter inductance control gain parameter; : Current filter inductance.

9. The power quality optimization control method based on improved particle swarm optimization algorithm according to claim 5, characterized in that: In the current control parameters, according to the current total harmonic distortion and the initial filter capacitance , calculate the filter capacitance control parameters, specifically: Where, : Current total harmonic distortion; : filter electrolytic control gain parameter; : Current filter capacitance.

10. A power quality optimization control system based on an improved particle swarm algorithm, a power quality optimization control method based on an improved particle swarm algorithm according to any one of claims 1 to 9, characterized in that: Including simulation initialization module, iteration initial module, optimization module, iteration module, and optimal output module; The simulation initialization module is used to build a wind farm simulation model and obtain the initial voltage deviation, frequency deviation and total harmonic distortion of the power grid based on the simulation model; The iterative initial module is used to calculate the initial fitness based on the initial voltage deviation, frequency deviation and total harmonic distortion using a multi-objective fitness function; The optimization module is used to improve the adaptive inertia weight and learning factor based on the particle swarm algorithm and randomly generate the initial control parameters of the power grid; The iterative module is used to control the simulation model according to the initial control parameters of the power grid, obtain the current voltage deviation, frequency deviation and total harmonic distortion, and calculate the current fitness and current control parameters based on the current voltage deviation, frequency deviation and total harmonic distortion; The optimal output module is used to preset the maximum number of iterations of fitness. If the current number of fitness iterations is equal to the maximum number of iterations, the grid control parameter corresponding to the current fitness is output; if the current number of fitness iterations does not reach the maximum number of iterations, the iteration initial module is re-entered.

Images