Reactive power optimization method of wind power grid-connected power system and related equipment

Abstract

The invention relates to the technical field of power optimization, and provides a reactive power optimization method and related equipment for a wind power grid-connected power system, and the method comprises the steps: obtaining a plurality of historical wind power plant wind speeds of an environment where the wind power grid-connected power system is located; calculating a wind speed probability density function based on all historical wind power plant wind speeds, and calculating the scene probability of the wind power grid-connected power system in each operation scene according to the wind speed probability density function; constructing a reactive power optimization objective function of the wind power grid-connected power system according to all scene probabilities; solving the reactive power optimization objective function based on a genetic algorithm and a particle swarm algorithm to obtain the capacitor bank input group number of each power generation node in the wind power grid-connected power system and reactive power output of SVG compensation; and performing reactive power optimization on each power generation node in the wind power grid-connected power system according to the capacitor bank input group number of each power generation node and the reactive power output compensated by the SVG. According to the method provided by the invention, the reactive power optimization efficiency of the wind power grid-connected power system can be improved.

Description

Technical Field

[0001] This application relates to the field of power optimization technology, and in particular to a reactive power optimization method and related equipment for a wind power grid-connected power system. Background Technology

[0002] In recent years, wind power has played an increasingly important role in the global power system. While the widespread integration of wind farms promotes the decarbonization of the energy structure, it also presents new challenges to the operation and dispatch of the power system. Due to the randomness and intermittency of wind energy, the output of wind farms fluctuates significantly, easily causing voltage deviations and reactive power fluctuations, thus threatening the safe and stable operation of the power system. Therefore, reasonable reactive power optimization for new power systems incorporating wind farms is of great significance for ensuring stable grid operation, reducing grid losses, and improving power quality.

[0003] To address the voltage and reactive power fluctuations caused by wind farm grid connection, scholars have studied various reactive power optimization strategies to improve voltage stability and reduce grid losses. Traditional reactive power optimization methods, such as linear programming (LP), nonlinear programming (NLP), Newton's method, and Lagrange relaxation, have been widely applied in reactive power optimization scenarios of new power systems with wind farms. Among them, the optimization method based on quadratic programming can approximate the nonlinear optimization objective as a quadratic function and linearize the constraints. It solves continuous variables through convex optimization, which improves the optimization accuracy to some extent, but the computational cost is large and it is not suitable for dynamic reactive power optimization with wind farms. In addition, the Lagrange relaxation method relaxes complex constraints to the objective function and constructs a dual problem decomposition optimization to coordinate continuous and discrete variables. However, when solving large-scale power system optimization problems with new energy sources, the dimensionality of variables increases dramatically, and the coordination efficiency of the decomposed subproblems decreases significantly.

[0004] With increasing uncertainty in power systems, the limitations of traditional optimization methods in global search, multi-objective collaboration, and high-dimensional computation are becoming increasingly apparent. Intelligent algorithms, through mechanisms such as parallel search and gradient-free optimization, offer new avenues to overcome these bottlenecks. Particle swarm optimization (PSO) has become a crucial solution for optimization problems due to its high search efficiency and simple parameter settings. Classical PSO achieves rapid optimization through group collaboration, but its fixed inertial weights lead to an imbalance between global exploration and local development, easily getting trapped in local optima in the early stages and facing insufficient convergence speed in later stages. To address the shortcomings of classical PSO, existing methods optimize fixed compensation schemes based on improved genetic algorithms (GA). While this solves the problems of traditional algorithms (such as PSO) being prone to local optima and relying on precise models, it struggles to adapt to real-time fluctuations in reactive power output from wind farms. Some researchers have used deep reinforcement learning to achieve online control of wind power clusters and active distribution networks, but this requires a large amount of historical data to support offline training and has not been combined with traditional mathematical programming methods or heuristic algorithms. Furthermore, its generalization ability to sudden extreme scenarios in real-world systems has not been fully verified. While some researchers have improved convergence through sensitivity analysis using transmission and distribution co-optimization based on the generalized master-slave splitting method, it still suffers from problems such as numerous iterations and low computational efficiency under large-scale distributed power generation integration. Traditional Anderson acceleration algorithms used by some researchers rely solely on unidirectional historical information to correct boundary variables, making it difficult to handle the complex power flow interactions of bidirectional coupling in transmission and distribution networks, resulting in a slow reduction in the convergence radius. Therefore, the current reactive power optimization of wind-connected power systems suffers from low efficiency. Summary of the Invention

[0005] This application provides a reactive power optimization method for wind power grid-connected power systems, which can solve the problem of low reactive power optimization efficiency in wind power grid-connected power systems.

[0006] In a first aspect, embodiments of this application provide a reactive power optimization method for a wind power grid-connected power system, the reactive power optimization method comprising:

[0007] Obtain the historical wind speeds of multiple wind farms in the environment where the wind power grid-connected power system is located;

[0008] The wind speed probability density function is calculated based on the wind speed of all historical wind farms, and the probability of the wind power grid-connected power system being in each operating scenario is calculated based on the wind speed probability density function.

[0009] A reactive power optimization objective function for the wind power grid-connected power system is constructed based on the probabilities of all scenarios. The reactive power optimization objective function is used to describe the sum of network losses and voltage deviations of the wind power grid-connected power system under each operating scenario.

[0010] Solve the reactive power optimization objective function based on the genetic algorithm and the particle swarm algorithm to obtain the number of capacitor banks put into operation and the reactive power output compensated by SVG for each generating node in the wind power integrated power system; the reactive power output refers to the reactive power provided to the wind power integrated power system.

[0011] Perform reactive power optimization for each generating node in the wind power integrated power system according to the number of capacitor banks put into operation and the reactive power output compensated by SVG for each generating node.

[0012] Optionally, calculate the wind speed probability density function based on all historical wind farm wind speeds, including:

[0013] Through the formula:

[0014]

[0015] Calculate the wind speed probability density function f(V);

[0016] Among them, V represents the given wind farm wind speed, c represents the scale parameter calculated based on all historical wind farm wind speeds, k represents the shape parameter calculated based on all historical wind farm wind speeds, and e represents the natural constant.

[0017] Optionally, multiple operating scenarios include full-load operation, shutdown state, and partial output;

[0018] Calculate the scenario probabilities of the wind power integrated power system in each operating scenario according to the wind speed probability density function, including:

[0019] Through the formula:

[0020]

[0021]

[0022]

[0023] Calculate the scenario probability P1 of full-load operation, the scenario probability P2 of shutdown state, and the scenario probability P3 of partial output;

[0024] Among them, v represents the wind farm wind speed, v0 represents the wind turbine cut-in wind speed, v r represents the rated wind speed, v m represents the cut-out wind speed, P(v r ≤v<v m ) represents the scenario probability when v r ≤v<v m is satisfied, P(v<v0) represents the scenario probability when v<v0 is satisfied, P(v≥v m ) represents the scenario probability when v≥v m is satisfied, P(v0≤v<vr ) indicates that v0 ≤ v <v r The probability of the scenario at that time.

[0025] Optionally, the reactive power optimization objective function is:

[0026]

[0027]

[0028]

[0029]

[0030] U i,min ≤U i ≤U i,max

[0031] C i,min ≤C i ≤C i,max

[0032] Q svgi,min ≤Q svgi ≤Q svgi,max

[0033] Where λ represents the reactive power optimization comprehensive index, α1 represents the weighting coefficient of network loss, α2 represents the weighting coefficient of voltage deviation, α1=1-α2, α2=sin(yπ / 2), and y represents the voltage deviation normalization. ΔU i,max P represents the maximum allowable voltage deviation for the i-th generating node during safe operation. l Let represent the probability of the l-th running scenario, and n represent the number of running scenarios, where n = 3. δ represents the grid loss of the wind-powered grid-connected power system under the l-th operating scenario. l P represents the voltage deviation of the wind power grid-connected power system under the l-th operating scenario. loss This indicates the network loss of a wind power grid-connected power system under operating conditions. Let δ represent the set of grid losses of the wind-powered grid-connected power system under all operating scenarios, and let δ represent the voltage deviation of the wind-powered grid-connected power system under the operating scenarios, where δ∈{δ l}, {δ l} represents the set of voltage deviations of the wind power grid-connected power system under all operating scenarios, N represents the number of generating nodes in the wind power grid-connected power system, and G represents the voltage deviation set of the wind power grid-connected power system under all operating scenarios. ij U represents the electrical conductance between the i-th and j-th power generation nodes. i U represents the voltage amplitude of the i-th generating node. jLet θ represent the voltage amplitude at the j-th generating node. ij This represents the phase difference between the i-th and j-th power generation nodes. P represents the reference voltage value of the i-th generating node. Gi P represents the active power of the power source at the i-th generating node. Ii B represents the active power of the load at the i-th generation node. ij Q represents the susceptance between the i-th generating node and the j-th generating node. Gi Q represents the reactive power of the power source at the i-th generation node. Ci Q represents the reactive power output of the capacitor bank at the i-th generation node. svgi Q represents the reactive power output compensated by the SVG at the i-th generation node. Ii U represents the reactive power of the load at the i-th generation node. i,min U represents the minimum voltage amplitude at the i-th generating node. i,max C represents the maximum voltage amplitude at the i-th generating node. i C represents the number of capacitor banks connected at the i-th power generation node. i,min C represents the minimum number of capacitor banks connected at the i-th power generation node. i,max Q represents the maximum number of capacitor banks connected at the i-th power generation node. svgi,min Q represents the minimum reactive power output compensated by the SVG at the i-th generation node. svgi,max This represents the maximum reactive power output compensated by the SVG at the i-th power generation node.

[0034] Optionally, the reactive power optimization objective function is solved using genetic algorithms and particle swarm optimization to obtain the number of capacitor banks connected to each generation node in the wind power grid-connected power system and the reactive power output compensated by SVG, including:

[0035] Initialize multiple initial populations; each initial population includes multiple individuals, including the number of capacitor banks connected to each generation node in the wind power grid-connected power system and the reactive power output value of SVG compensation.

[0036] The fitness of each individual in the initial population is calculated based on the reactive power optimization objective function.

[0037] Determine if the number of iterations is greater than the preset number of iterations;

[0038] If so, the individual with the highest fitness is taken as the target individual, and the number of capacitor banks connected and the reactive power output compensated by SVG for each generation node in the target individual are taken as the number of capacitor banks connected and the reactive power output compensated by SVG for each generation node in the wind power grid-connected power system.

[0039] Otherwise, increment the iteration count by 1, use the genetic operations in the genetic algorithm to generate multiple new populations based on all initial populations, and use the particle swarm optimization algorithm to calculate all new populations to obtain multiple updated populations. Use these multiple updated populations as multiple initial populations, return with the iteration count incremented by 1, and calculate the fitness of each individual in all initial populations according to the reactive power optimization objective function.

[0040] Optionally, the particle swarm optimization algorithm can be used to calculate all new populations, resulting in multiple updated populations, including:

[0041] Determine if the number of particle iterations is greater than the preset number of particle iterations;

[0042] If so, then treat all new populations as multiple updated populations;

[0043] Otherwise, calculate the adaptive inertia weight based on the number of particle iterations, update each individual in each new population based on the adaptive inertia weight, and take all the updated new populations as the whole new population. Increment the number of particle iterations by 1, and return to the step of determining whether the number of particle iterations is greater than the preset number of particle iterations.

[0044] Optionally, adaptive inertia weights are calculated based on the number of particle iterations, including:

[0045] By formula:

[0046]

[0047] Calculate the adaptive inertia weight ω;

[0048] Where, ω min ω represents the minimum value of the adaptive inertia weight. max represents the maximum value of the adaptive inertia weight, t represents the number of particle iterations, and T represents the maximum number of particle iterations.

[0049] Optionally, each individual in each new population is updated according to adaptive inertia weights, including:

[0050] The optimal individual is determined from all individuals in all new populations based on the reactive power optimization objective function;

[0051] Each individual in each new population is updated based on the optimal individual and adaptive inertia weights.

[0052] Secondly, embodiments of this application provide a reactive power optimization device for a wind power grid-connected power system, comprising:

[0053] The acquisition module is used to acquire the historical wind speeds of multiple wind farms in the environment where the wind power grid-connected power system is located;

[0054] The calculation module is used to calculate the wind speed probability density function based on the wind speed of all historical wind farms, and to calculate the scenario probability of the wind power grid-connected power system under each operating scenario based on the wind speed probability density function.

[0055] The module is used to construct the reactive power optimization objective function of the wind power grid-connected power system based on the probabilities of all scenarios; the reactive power optimization objective function is used to describe the sum of network loss and voltage deviation of the wind power grid-connected power system under each operating scenario.

[0056] The solution module is used to solve the reactive power optimization objective function based on genetic algorithm and particle swarm algorithm to obtain the number of capacitor banks connected and the reactive power output of SVG compensation for each generation node in the wind power grid-connected power system; reactive power output refers to the reactive power provided to the wind power grid-connected power system.

[0057] The reactive power optimization module is used to optimize the reactive power of each generation node in the wind power grid-connected power system based on the number of capacitor banks put into operation at each generation node and the reactive power output compensated by SVG.

[0058] Thirdly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the reactive power optimization method of the wind power grid-connected power system described above.

[0059] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned reactive power optimization method for wind power grid-connected power systems.

[0060] The above-mentioned solution in this application has the following beneficial effects:

[0061] In the embodiments of this application, the wind speeds of multiple historical wind farms in the environment of the wind-powered grid-connected power system are obtained. Then, a wind speed probability density function is calculated based on all historical wind farm wind speeds. The probability of the wind-powered grid-connected power system operating under each scenario is calculated based on the wind speed probability density function. A reactive power optimization objective function for the wind-powered grid-connected power system is then constructed based on all scenario probabilities. The reactive power optimization objective function is then solved using a genetic algorithm and a particle swarm optimization algorithm to obtain the number of capacitor banks connected to each generation node in the wind-powered grid-connected power system and the reactive power output compensated by SVG. Finally, reactive power optimization is performed on each generation node in the wind-powered grid-connected power system based on the number of capacitor banks connected to each generation node and the reactive power output compensated by SVG. The construction of the reactive power optimization objective function quantifies the grid loss and voltage deviation of the wind-powered grid-connected power system. Solving the reactive power optimization objective function using a genetic algorithm and a particle swarm optimization algorithm combines the global search capability of the genetic algorithm and the fast convergence characteristic of the particle swarm optimization algorithm, improving the speed and accuracy of solving the reactive power optimization objective function, thereby improving the efficiency of reactive power optimization in the wind-powered grid-connected power system.

[0062] Other beneficial effects of this application will be described in detail in the following detailed description section. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 A flowchart of a reactive power optimization method for a wind power grid-connected power system provided in an embodiment of this application;

[0065] Figure 2 This is a schematic diagram of a wind speed probability curve provided in an embodiment of this application;

[0066] Figure 3 This is a schematic diagram of the power characteristic curve provided in an embodiment of this application;

[0067] Figure 4 A schematic diagram of the IEEE 33-node system architecture provided in an embodiment of this application;

[0068] Figure 5 A schematic diagram of the convergence curve provided in an embodiment of this application;

[0069] Figure 6 This is a schematic diagram of the capacitor bank operation and SVG output result curves provided in an embodiment of this application;

[0070] Figure 7 This is a schematic diagram of a voltage optimization curve provided in one embodiment of this application;

[0071] Figure 8 A schematic diagram of the reactive power optimization device for a wind power grid-connected power system provided in an embodiment of this application;

[0072] Figure 9 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. Detailed Implementation

[0073] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0074] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0075] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0076] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0077] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0078] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0079] To address the low reactive power optimization efficiency of existing wind power grid-connected power systems, this application provides an improved reactive power optimization method for wind power grid-connected power systems. This method obtains the wind speeds of multiple historical wind farms in the environment where the wind power grid-connected power system is located. Then, it calculates the wind speed probability density function based on all historical wind farm wind speeds, calculates the scenario probability of the wind power grid-connected power system under each operating scenario based on the wind speed probability density function, constructs the reactive power optimization objective function of the wind power grid-connected power system based on all scenario probabilities, and solves the reactive power optimization objective function based on a genetic algorithm and a particle swarm optimization algorithm to obtain the number of capacitor banks connected and the reactive power output compensated by SVG at each generation node in the wind power grid-connected power system. Finally, it optimizes the reactive power of each generation node in the wind power grid-connected power system based on the number of capacitor banks connected and the reactive power output compensated by SVG at each generation node. Among them, constructing a reactive power optimization objective function can quantify the grid loss and voltage deviation of the wind power grid-connected power system. The reactive power optimization objective function is solved based on genetic algorithm and particle swarm algorithm. By combining the global search capability of genetic algorithm and the fast convergence characteristic of particle swarm algorithm, the speed and accuracy of solving the reactive power optimization objective function are improved, thereby improving the efficiency of reactive power optimization of the wind power grid-connected power system.

[0080] The reactive power optimization method for wind power grid-connected power systems provided in this application will be illustrated by example below.

[0081] like Figure 1 As shown, the reactive power optimization method for wind power grid-connected power systems provided in this application includes the following steps:

[0082] Step 11: Obtain the historical wind speeds of multiple wind farms in the environment where the wind power grid-connected power system is located.

[0083] The aforementioned wind power grid-connected power system is a wind power grid-connected power system that requires reactive power optimization, and the historical wind farm wind speed refers to the wind farm wind speed at historical times.

[0084] In some embodiments of this application, historical wind speeds of wind farms in the environment can be obtained using devices such as sensors.

[0085] Step 12: Calculate the wind speed probability density function based on the wind speed of all historical wind farms, and calculate the scenario probability of the wind power grid-connected power system under each operating scenario based on the wind speed probability density function.

[0086] In some embodiments of this application, the steps of calculating the wind speed probability density function based on the wind speed of all historical wind farms, and calculating the scenario probability of the wind power grid-connected power system under each operating scenario based on the wind speed probability density function include:

[0087] The first step is to calculate the wind speed probability density function based on the wind speeds of all historical wind farms.

[0088] Specifically, through the formula:

[0089]

[0090] Calculate the wind speed probability density function f(V);

[0091] Where V represents the given wind speed of the wind farm (a preset value), c represents the scale parameter calculated based on the wind speeds of all historical wind farms, k represents the shape parameter calculated based on the wind speeds of all historical wind farms, and e represents the natural constant.

[0092] For example, a curve relating wind speed to wind speed probability can be plotted based on all historical wind farm wind speeds (the expression for this curve is the formula mentioned above), and the curve can be fitted using the least squares method to obtain shape and scale parameters. The fitted wind speed probability curve is shown below. Figure 2 As shown, Figure 2 The horizontal axis represents wind speed in meters per second (m / s), and the vertical axis represents probability density. The different curves are the wind speed probability curves for k=1, c=7, k=3, c=7, k=2, c=7, and k=1.5, c=7, respectively.

[0093] The second step is to calculate the scenario probability of the wind power grid-connected power system under each operating scenario based on the wind speed probability density function.

[0094] Multiple operating scenarios include full-load operation (wind farm wind speed v, wind turbine cut-in wind speed v0, rated wind speed v). r The relationship between v r ≤v<v m (Stopped state (satisfying v < v0 or v ≥ v)) m ) and partial output (satisfying v0≤v<v r ).

[0095] Specifically, through the formula:

[0096]

[0097]

[0098]

[0099] Calculate the scenario probability P1 of full-load operation, the scenario probability P2 of shutdown state, and the scenario probability P3 of partial output.

[0100] Among them, v represents the wind farm wind speed, v0 represents the wind turbine cut-in wind speed, v r represents the rated wind speed, v m represents the cut-out wind speed, P(v r ≤v<v m ) represents the scenario probability when v r ≤v<v m , P(v<v0) represents the scenario probability when v<v0, P(v≥v m ) represents the scenario probability when v≥v m , P(v0≤v<v r ) represents the scenario probability when v0≤v<v r .

[0101] It should be noted that partial output can be discretized into multiple sub-scenarios (such as v0≤v1≤v2≤...≤v m-1 ≤v r ), and the calculation principle of the scenario probability of each sub-scenario is the same as the above formula.

[0102] Exemplarily, the power characteristic curve between the wind speed and the power of the wind power grid-connected power system is as Figure 3 shown, Figure 3 where the horizontal axis represents the wind speed and the vertical axis represents the power P v of the wind power grid-connected power system, p r represents the rated capacity of the wind turbine, v0 represents the wind turbine cut-in wind speed, v r represents the rated wind speed, v m represents the cut-out wind speed, and the expression is:

[0103]

[0104] Among them, k1 and k2 represent intermediate parameters, k1 = p r / (v r -v0), k2 = -v0.

[0105] Step 13, construct the reactive power optimization objective function of the wind power grid-connected power system according to all scenario probabilities.

[0106] The reactive power optimization objective function is used to describe the sum of network losses and voltage deviations of the wind power grid-connected power system under each operating scenario.

[0107] Specifically, the reactive power optimization objective function is:

[0108]

[0109]

[0110]

[0111]

[0112] U i,min ≤U i ≤U i,max

[0113] C i,min ≤C i ≤C i,max

[0114] Q svgi,min ≤Q svgi ≤Q svgi,max

[0115] Where λ represents the reactive power optimization comprehensive index, α1 represents the weighting coefficient of network loss, α2 represents the weighting coefficient of voltage deviation, α1=1-α2, α2=sin(yπ / 2), and y represents the voltage deviation normalization. ΔU i,max P represents the maximum allowable voltage deviation for the i-th generating node during safe operation. l Let represent the probability of the l-th running scenario, and n represent the number of running scenarios, where n = 3. δ represents the grid loss of the wind-powered grid-connected power system under the l-th operating scenario. l P represents the voltage deviation of the wind power grid-connected power system under the l-th operating scenario. loss This indicates the network loss of a wind power grid-connected power system under operating conditions. Let δ represent the set of grid losses of the wind-powered grid-connected power system under all operating scenarios, and let δ represent the voltage deviation of the wind-powered grid-connected power system under the operating scenarios, where δ∈{δ l}, {δ l} represents the set of voltage deviations of the wind power grid-connected power system under all operating scenarios, N represents the number of generating nodes in the wind power grid-connected power system, and G represents the voltage deviation set of the wind power grid-connected power system under all operating scenarios. ij U represents the electrical conductance between the i-th and j-th power generation nodes. i U represents the voltage amplitude of the i-th generating node.j Let θ represent the voltage amplitude at the j-th generating node. ij This represents the phase difference between the i-th and j-th power generation nodes. P represents the reference voltage value of the i-th generating node. Gi P represents the active power of the power source at the i-th generating node. Ii B represents the active power of the load at the i-th generation node. ij Q represents the susceptance between the i-th generating node and the j-th generating node. Gi Q represents the reactive power of the power source at the i-th generation node. Ci Q represents the reactive power output of the capacitor bank at the i-th generation node. svgi Q represents the reactive power output compensated by the SVG at the i-th generation node. Ii U represents the reactive power of the load at the i-th generation node. i,min U represents the minimum voltage amplitude at the i-th generating node. i,max C represents the maximum voltage amplitude at the i-th generating node. i C represents the number of capacitor banks connected at the i-th power generation node. i,min C represents the minimum number of capacitor banks connected at the i-th power generation node. i,max Q represents the maximum number of capacitor banks connected at the i-th power generation node. svgi,min Q represents the minimum reactive power output compensated by the SVG at the i-th generation node. svgi,max This represents the maximum reactive power output compensated by the SVG at the i-th power generation node.

[0116] It should be noted that P loss The expressions for δ and δ are applicable to calculations in all operating scenarios. When substituting specific parameters, substitute the parameters corresponding to the operating scenario, such as for... and δ l When performing the calculation, substitute G ij It should be the electrical conductance between the i-th and j-th power generation nodes under the l-th operating scenario, U i It should be the voltage amplitude of the i-th generation node under the l-th operating scenario, U j It should be the voltage amplitude of the j-th generation node under the l-th operating scenario, θ ij It should be the phase difference between the i-th and j-th power generation nodes under the l-th operating scenario. It should be the reference voltage value of the i-th power generation node under the l-th operating scenario.

[0117] Step 14: Solve the reactive power optimization objective function based on the genetic algorithm and particle swarm optimization algorithm to obtain the number of capacitor banks put into operation and the reactive power output of SVG compensation for each generation node in the wind power grid-connected power system.

[0118] The aforementioned power generation nodes refer to the nodes in a wind power grid-connected power system that generate wind power, including power sources, capacitor banks, SVG devices, and wind power generation equipment. Reactive power output refers to the reactive power provided to the wind power grid-connected power system. The reactive power output compensated by the aforementioned SVG refers to the reactive power provided by the Static Var Generator (SVG) to the wind power grid-connected power system.

[0119] In some embodiments of this application, the steps of solving the reactive power optimization objective function based on genetic algorithm and particle swarm optimization algorithm to obtain the number of capacitor banks connected and the reactive power output of SVG compensation at each generation node in the wind power grid-connected power system include:

[0120] The first step is to initialize multiple initial populations.

[0121] The initial population includes multiple individuals, which include the number of capacitor banks connected to each generation node in the wind power grid system and the reactive power output value of SVG compensation.

[0122] The second step is to calculate the fitness of each individual in the initial population based on the reactive power optimization objective function.

[0123] Specifically, each individual is substituted into the reactive power optimization objective function to calculate the corresponding reactive power optimization comprehensive index, which is then used as the fitness of the individual.

[0124] For example, assuming the initial power factor of the wind turbine is 1, the Newton-Raphson method is used to perform power flow calculations on the new power system containing the wind farm based on the active power output of the wind turbine, so as to obtain the generation node voltage, active power loss and reactive power demand under various operating scenarios.

[0125] The third step is to determine whether the number of iterations is greater than the preset number of iterations.

[0126] If so, the individual with the highest fitness is taken as the target individual, and the number of capacitor banks connected and the reactive power output compensated by SVG for each generation node in the target individual are taken as the number of capacitor banks connected and the reactive power output compensated by SVG for each generation node in the wind power grid-connected power system.

[0127] Otherwise, increment the iteration count by 1, use the genetic operations in the genetic algorithm to generate multiple new populations based on all initial populations, and use the particle swarm optimization algorithm to calculate all new populations to obtain multiple updated populations. Use these multiple updated populations as multiple initial populations, return with the iteration count incremented by 1, and calculate the fitness of each individual in all initial populations according to the reactive power optimization objective function.

[0128] It should be noted that the genetic operations in genetic algorithms include selection, crossover, and mutation. The steps described above, which use particle swarm optimization to calculate all new populations and obtain multiple updated populations, are as follows:

[0129] Determine if the number of particle iterations is greater than the preset number of particle iterations.

[0130] If so, then all new populations will be treated as multiple updated populations.

[0131] Otherwise, calculate the adaptive inertia weight based on the number of particle iterations, update each individual in each new population based on the adaptive inertia weight, and take all the updated new populations as the whole new population. Increment the number of particle iterations by 1, and return to the step of determining whether the number of particle iterations is greater than the preset number of particle iterations.

[0132] The above calculation of adaptive inertia weights based on particle iteration count is as follows:

[0133] By formula:

[0134]

[0135] Calculate the adaptive inertia weight ω.

[0136] Where, ω min ω represents the minimum value of the adaptive inertia weight. max represents the maximum value of the adaptive inertia weight, t represents the number of particle iterations, and T represents the maximum number of particle iterations.

[0137] The above update is performed on each individual in each new population based on adaptive inertia weights, specifically as follows:

[0138] First, the optimal individual is determined from all individuals in all new populations based on the reactive power optimization objective function.

[0139] Specifically, each individual in the new population is substituted into the reactive power optimization objective function to calculate the corresponding reactive power optimization comprehensive index, which is then used as the individual's fitness. The individual with the highest fitness is selected as the optimal individual.

[0140] Then, each individual in each new population is updated based on the optimal individual and the adaptive inertia weight.

[0141] For example, the initial values ​​for both the number of iterations and the number of particle iterations are 0. Individual particles can be updated using the formula from the particle swarm optimization algorithm:

[0142] x i (t+1)=x i (t)+v i (t+1)

[0143] v i (t+1)=ωv i (t)+c1r1(p best -x i (t))+c2r2(g best -x i (t))

[0144] Where, x i (t+1) represents the updated individual, x i (t) represents an individual, v i (t+1) represents the updated velocity of this individual, v i (t) represents the velocity of the individual, c1 and c2 are acceleration constants, r1 and r2 are random numbers, and p best G represents the historical best individual of this individual. best Let ω represent the optimal individual among all individuals, and let ω represent the adaptive inertia weight.

[0145] Step 15: Perform reactive power optimization for each generation node in the wind power grid-connected power system based on the number of capacitor banks put into operation at each generation node and the reactive power output of SVG compensation.

[0146] In related technologies, reactive power compensation for the output power of a power generation node is achieved through an SVG (Static Var Generator) device installed in the node. An SVG device is a type of device used for reactive power compensation.

[0147] Specifically, for each power generation node, the capacitor banks are configured according to the number of capacitor banks put into operation at that power generation node, so that the number of capacitor banks put into operation at that power generation node is the corresponding number of capacitor banks put into operation. The SVG device at that power generation node is configured according to the reactive power compensation of the SVG device, so that the reactive power compensation of the SVG device at that power generation node is the corresponding reactive power compensation of the SVG, thereby achieving reactive power optimization for that power generation node.

[0148] It is worth mentioning that constructing a reactive power optimization objective function can quantify the grid loss and voltage deviation of the wind power grid-connected power system. The reactive power optimization objective function is solved based on genetic algorithm and particle swarm algorithm. By combining the global search capability of genetic algorithm and the fast convergence characteristic of particle swarm algorithm, the speed and accuracy of solving the reactive power optimization objective function are improved, thereby improving the efficiency of reactive power optimization of the wind power grid-connected power system.

[0149] The method of this application will be illustrated below with a specific example.

[0150] use Figure 4 The IEEE 33-bus standard distribution system, which includes wind power and is used for power system analysis and research, is shown in the figure. Multi-scenario reactive power optimization simulation test is conducted on the system. Figure 4 The system includes 33 nodes (i.e., generation nodes), generators, wind turbines, a capacitor bank installed at node 24, and an SVG device installed at node 25. The apparent power reference value of the system is 100MVA, and the voltage reference value is 12.66kV. The total active load of the system is 3715.0kW, and the total reactive load is 2300.0kvar. Detailed load bus data and branch parameters are provided in Appendix A. Seven switchable parallel capacitors are installed at node 24, and an SVG reactive power compensation device is installed at node 25. Each capacitor has a capacity of 0.2Mvar, and the SVG has a capacity of 3Mvar. A 1.5MW doubly-fed wind turbine is connected to node 30. The bus outlet voltage is 690V, the cut-in wind speed v0 is 3m / s, and the rated wind speed v r The cutoff wind speed is 12 m / s, v m The speed is 25 m / s, and the parameters k and c are 2.4 and 9.5, respectively.

[0151] The parameters of the 1.5MW doubly fed induction wind turbine connected to the system are as follows: stator resistance is 0.0071pu, stator leakage reactance is 0.171pu, rotor resistance is 0.005pu, rotor leakage reactance is 0.156pu, and excitation reactance is 2.9pu.

[0152] The parameters for the genetic algorithm are set as follows: population size of 50, maximum number of iterations of 100, crossover rate of 0.7, and mutation rate of 0.1. The parameters for the embedded particle swarm optimization algorithm are set as follows: maximum number of iterations of 100, 20 subpopulations randomly selected from the global population, learning factors c1 = c2 = 1.5, maximum inertia weight of 0.9, and minimum inertia weight of 0.4.

[0153] The calculated probability of occurrence and wind power output for the three operating scenarios are shown in Table 1.

[0154] Table 1

[0155] Shutdown status Partial output Full load operation Scenario Probability 0.1099 0.7035 0.1866 Power / MW 0 0.2839 1.5

[0156] To highlight the superiority of the proposed method in solving the adaptive optimization objective, the convergence curves of four different algorithms are compared. Figure 5 The convergence curves of the four algorithms during the optimization process are shown. Figure 5 The horizontal axis represents the number of iterations, the vertical axis represents the value of the objective function, PSO represents the convergence curve of the particle swarm optimization algorithm, GA represents the convergence curve of the genetic algorithm, GA_PSO represents the convergence curve of the particle swarm optimization algorithm based on the genetic algorithm, and SGA represents the convergence curve of the method in this application.

[0157] The SGA algorithm (i.e., the method of this application) stabilizes its objective function after 20 iterations, demonstrating a significantly faster convergence speed than the standard PSO and GA algorithms. Furthermore, in terms of convergence accuracy, through an adaptive inertia weighting mechanism and a unique embedded particle swarm optimization logic, the steady-state objective function value of SGA is reduced by 29.7%, 11.4%, and 4.1% compared to PSO, GA, and GA_PSO, respectively. The standard deviation of SGA's fluctuation in the later stages of iteration is only 0.0003, significantly lower than other algorithms, verifying its robustness. Compared to the other three algorithms, the SGA algorithm proposed in this application shows significant improvements in both convergence speed and optimization capability.

[0158] The final network loss optimization results of the four algorithms are shown in Table 2.

[0159] Table 2

[0160] Before optimization PSO GA GA-PSO SGA Active network loss / MW 6.4085 5.0533 4.9695 4.8531 4.5774

[0161] Before optimization, the total network loss of the system was 6.4085 MW. After algorithm optimization, the active power losses of the system were reduced to a certain extent. Among them, the SGA algorithm performed best, with a network loss reduction of 28.6%, which was 9.4%, 7.9%, and 5.7% higher than PSO, GA, and GA-PSO, respectively. This significant advantage stems from the hybrid optimization mechanism of SGA: it avoids local optima through the global search capability of genetic algorithms, combines the fast convergence characteristics of particle swarm optimization, and introduces adaptive weight dynamic balancing exploration and development capabilities.

[0162] After optimization, the capacitor bank action and SVG output results are as follows: Figure 6 As shown, Figure 6 The horizontal axis represents time, in hours (h). Figure 6 'a' represents the operating logic curve of the switchable capacitor bank, with the vertical axis representing the capacitor bank's position. Figure 6 b represents the SVG output curve, and the vertical axis represents the SVG power in Mvar.

[0163] Four optimization algorithms Figure 4 The 24-hour voltage optimization curve for node 25 in the system shown is as follows: Figure 7 As shown, the horizontal axis represents time in hours (h), the vertical axis represents the voltage per unit value, PSO represents the voltage optimization curve of the particle swarm optimization algorithm, GA represents the voltage optimization curve of the genetic algorithm, GA_PSO represents the voltage optimization curve of the particle swarm optimization algorithm based on the genetic algorithm, and SGA represents the voltage optimization curve of the method of this application.

[0164] As shown in the figure, the SGA algorithm exhibits a significant advantage in voltage stability. Compared with PSO, the average voltage deviation is reduced by 64.3% and the standard deviation is reduced by 64.4%. Its voltage value closely follows the reference value of 1.0 pu throughout the day, with the smallest fluctuation range (±0.015 pu). There is no voltage limit violation throughout the day, and the maximum deviation is optimized by 40.2% compared with GA-PSO. Other algorithms have varying degrees of voltage deviation or limit violation risk at different times.

[0165] The results above demonstrate that the proposed method exhibits significant advantages in multi-scenario comprehensive optimization. Compared to traditional optimization algorithms, it reduces network losses more significantly, achieves greater voltage fluctuation and stability, and eliminates the risk of exceeding limits. This indicates that the hybrid optimization framework and dynamic weighting mechanism of SGA can effectively coordinate the actions of reactive power compensation equipment, balancing grid economy and safety, and verifying the effectiveness and engineering applicability of the algorithm improvement.

[0166] The method described in this application has the following main advantages:

[0167] Based on the scenario division of wind speed probability distribution, the impact of wind power output uncertainty on the power grid is quantified. Through multi-scenario collaborative optimization, a more robust reactive power regulation strategy is generated, which improves the adaptability of the optimization model to random fluctuations in wind power.

[0168] By integrating the global search capability of genetic algorithms with the fast convergence characteristics of particle swarm optimization, and introducing dynamic inertia weights, the algorithm effectively balances its exploration and development capabilities. Numerical examples demonstrate that this algorithm significantly reduces system network losses and suppresses voltage fluctuations, exhibiting faster convergence and stronger robustness.

[0169] The reactive power optimization device for wind power grid-connected power systems provided in this application is described below as an example.

[0170] like Figure 8 As shown, this application embodiment provides a reactive power optimization device for a wind power grid-connected power system. The reactive power optimization device 800 for the wind power grid-connected power system includes:

[0171] The acquisition module 801 is used to acquire the wind speeds of multiple historical wind farms in the environment of the wind power grid-connected power system.

[0172] The calculation module 802 is used to calculate the wind speed probability density function based on the wind speed of all historical wind farms, and to calculate the scenario probability of the wind power grid-connected power system under each operating scenario based on the wind speed probability density function.

[0173] Module 803 is used to construct the reactive power optimization objective function of the wind power grid-connected power system according to the probabilities of all scenarios; the reactive power optimization objective function is used to describe the sum of network loss and voltage deviation of the wind power grid-connected power system under each operating scenario.

[0174] The solver module 804 is used to solve the reactive power optimization objective function based on the genetic algorithm and the particle swarm algorithm to obtain the number of capacitor banks connected to each generation node in the wind power grid-connected power system and the reactive power output of SVG compensation; reactive power output refers to the reactive power provided to the wind power grid-connected power system.

[0175] The reactive power optimization module 805 is used to optimize the reactive power of each generation node in the wind power grid-connected power system based on the number of capacitor banks put into operation at each generation node and the reactive power output compensated by SVG.

[0176] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0177] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0178] like Figure 9 As shown, an embodiment of this application provides a terminal device, wherein the terminal device D10 of this embodiment includes: at least one processor D100 ( Figure 9 The diagram shows only one processor, a memory D101, and a computer program D102 stored in the memory D101 and executable on the at least one processor D100, wherein the processor D100 executes the computer program D102 to implement the steps in any of the above method embodiments.

[0179] Specifically, when the processor D100 executes the computer program D102, it acquires the wind speeds of multiple historical wind farms in the environment of the wind power grid-connected power system, calculates the wind speed probability density function based on all historical wind farm wind speeds, calculates the scenario probability of the wind power grid-connected power system under each operating scenario based on the wind speed probability density function, constructs the reactive power optimization objective function of the wind power grid-connected power system based on all scenario probabilities, and then solves the reactive power optimization objective function based on genetic algorithm and particle swarm optimization algorithm to obtain the number of capacitor banks connected and the reactive power output compensated by SVG for each generation node in the wind power grid-connected power system. Finally, it performs reactive power optimization for each generation node in the wind power grid-connected power system based on the number of capacitor banks connected and the reactive power output compensated by SVG for each generation node. Among them, constructing a reactive power optimization objective function can quantify the grid loss and voltage deviation of the wind power grid-connected power system. The reactive power optimization objective function is solved based on genetic algorithm and particle swarm algorithm. By combining the global search capability of genetic algorithm and the fast convergence characteristic of particle swarm algorithm, the speed and accuracy of solving the reactive power optimization objective function are improved, thereby improving the efficiency of reactive power optimization of the wind power grid-connected power system.

[0180] The processor D100 can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0181] In some embodiments, the memory D101 may be an internal storage unit of the terminal device D10, such as a hard disk or memory of the terminal device D10. In other embodiments, the memory D101 may be an external storage device of the terminal device D10, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the terminal device D10. Furthermore, the memory D101 may include both internal and external storage units of the terminal device D10. The memory D101 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory D101 can also be used to temporarily store data that has been output or will be output.

[0182] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0183] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0184] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to the reactive power optimization method device / terminal equipment of the wind power grid-connected power system, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0185] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0186] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0187] The above description is the preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principles described in this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A reactive power optimization method for a wind power grid-connected power system, characterized in that, include: Obtain the historical wind speeds of multiple wind farms in the environment where the wind power grid-connected power system is located; The wind speed probability density function is calculated based on the wind speed of all historical wind farms, and the probability of the wind power grid-connected power system being in each operating scenario is calculated based on the wind speed probability density function. The reactive power optimization objective function of the wind power grid-connected power system is constructed based on the probabilities of all scenarios; the reactive power optimization objective function is used to describe the sum of grid loss and voltage deviation of the wind power grid-connected power system under each operating scenario; The reactive power optimization objective function is solved using a genetic algorithm and a particle swarm optimization algorithm to obtain the number of capacitor banks connected to each generation node in the wind power grid-connected power system and the reactive power output compensated by SVG; the reactive power output refers to the reactive power provided to the wind power grid-connected power system. Based on the number of capacitor banks put into operation at each generation node and the reactive power output compensated by SVG, reactive power optimization is performed on each generation node in the wind power grid-connected power system.

2. The reactive power optimization method according to claim 1, characterized in that, The wind speed probability density function calculated based on the wind speeds of all historical wind farms includes: By formula: Calculate the wind speed probability density function f(V); Where V represents the given wind speed of the wind farm, c represents the scale parameter calculated based on the wind speeds of all historical wind farms, k represents the shape parameter calculated based on the wind speeds of all historical wind farms, and e represents the natural constant.

3. The reactive power optimization method according to claim 1, characterized in that, Multiple operating scenarios include full-load operation, shutdown, and partial output; The calculation of the scenario probability of the wind power grid-connected power system under each operating scenario based on the wind speed probability density function includes: By formula: Calculate the probability of a scenario with full load operation (P1), a scenario with shutdown (P2), and a scenario with partial output (P3). Among them, v represents the wind farm wind speed, v0 represents the wind turbine cut-in wind speed, v r represents the rated wind speed, v m represents the cut-out wind speed, P(v r ≤v<v m ) represents the scenario probability when v r ≤v<v m , P(v<v0) represents the scenario probability when v<v0, P(v≥v m ) represents the scenario probability when v≥v m , P(v0≤v<v r ) represents the scenario probability when v0≤v<v r .

4. The reactive power optimization method according to claim 3, characterized in that, The reactive power optimization objective function is: IN i,min ≤U i ≤U i,max C i,min ≤C i ≤C i,max Q svgi,min ≤Q svgi ≤Q svgi,max Where λ represents the reactive power optimization comprehensive index, α1 represents the weighting coefficient of network loss, α2 represents the weighting coefficient of voltage deviation, α1=1-α2, α2=sin(yπ / 2), and y represents the voltage deviation normalization. ΔU i,max P represents the maximum allowable voltage deviation for the i-th generating node during safe operation. l Let represent the probability of the l-th running scenario, and n represent the number of running scenarios, where n = 3. δ represents the grid loss of the wind power grid-connected power system under the l-th operating scenario. l P represents the voltage deviation of the wind power grid-connected power system under the l-th operating scenario. loss This indicates the network loss of the wind power grid-connected power system under the operating scenario. Let δ represent the set of grid losses of the wind power grid-connected power system under all operating scenarios, and let δ represent the voltage deviation of the wind power grid-connected power system under the operating scenarios, where δ∈{δ l }, {δ l } represents the set of voltage deviations of the wind power grid-connected power system under all operating scenarios, N represents the number of generating nodes in the wind power grid-connected power system, and G represents the voltage deviation set of the wind power grid-connected power system under all operating scenarios. ij U represents the electrical conductance between the i-th and j-th power generation nodes. i U represents the voltage amplitude of the i-th generating node. j Let θ represent the voltage amplitude at the j-th generating node. ij This represents the phase difference between the i-th and j-th power generation nodes. P represents the reference voltage value of the i-th generating node. Gi P represents the active power of the power source at the i-th generating node. Ii B represents the active power of the load at the i-th generation node. ij Q represents the susceptance between the i-th generating node and the j-th generating node. Gi Q represents the reactive power of the power source at the i-th generation node. Ci Q represents the reactive power output of the capacitor bank at the i-th generation node. svgi Q represents the reactive power output compensated by the SVG at the i-th generation node. Ii U represents the reactive power of the load at the i-th generation node. i,min U represents the minimum voltage amplitude at the i-th generating node. i,max C represents the maximum voltage amplitude at the i-th generating node. i C represents the number of capacitor banks connected at the i-th power generation node. i,min C represents the minimum number of capacitor banks connected at the i-th power generation node. i,max Q represents the maximum number of capacitor banks connected at the i-th power generation node. svgi,min Q represents the minimum reactive power output compensated by the SVG at the i-th generation node. svgi,max This represents the maximum reactive power output compensated by the SVG at the i-th power generation node.

5. The reactive power optimization method according to claim 4, characterized in that, The reactive power optimization objective function is solved using a genetic algorithm and a particle swarm optimization algorithm to obtain the number of capacitor banks connected and the reactive power output compensated by SVG at each generation node in the wind power grid-connected power system, including: Initialize multiple initial populations; each initial population includes multiple individuals, and each individual includes the number of capacitor banks connected to each generation node in the wind power grid-connected power system and the reactive power output value of SVG compensation; The fitness of each individual in the initial population is calculated based on the reactive power optimization objective function. Determine if the number of iterations is greater than the preset number of iterations; If so, the individual with the highest fitness is taken as the target individual, and the number of capacitor banks connected and the reactive power compensated by SVG for each power generation node in the target individual are taken as the number of capacitor banks connected and the reactive power compensated by SVG for each power generation node in the wind power grid-connected power system. Otherwise, increment the iteration count by 1, use the genetic operations in the genetic algorithm to generate multiple new populations based on all initial populations, and use the particle swarm optimization algorithm to calculate all new populations to obtain multiple updated populations. Use the multiple updated populations as the multiple initial populations, return to the step of incrementing the iteration count by 1, and calculate the fitness of each individual in all initial populations according to the reactive power optimization objective function.

6. The reactive power optimization method according to claim 5, characterized in that, The particle swarm optimization algorithm is used to calculate all new populations, resulting in multiple updated populations, including: Determine if the number of particle iterations is greater than the preset number of particle iterations; If so, then all new populations will be treated as the multiple updated populations; Otherwise, the adaptive inertia weight is calculated based on the number of particle iterations, and each individual in each new population is updated according to the adaptive inertia weight. All updated new populations are then used as all new populations. The number of particle iterations is incremented by 1, and the process returns to the step of determining whether the number of particle iterations is greater than the preset number of particle iterations.

7. The reactive power optimization method according to claim 6, characterized in that, The calculation of adaptive inertia weights based on the number of particle iterations includes: By formula: Calculate the adaptive inertia weight ω; Where, ω min ω represents the minimum value of the adaptive inertia weight. max represents the maximum value of the adaptive inertia weight, t represents the number of particle iterations, and T represents the maximum number of particle iterations.

8. The reactive power optimization method according to claim 7, characterized in that, The step of updating each individual in each new population according to the adaptive inertia weight includes: The optimal individual is determined from all individuals in all new populations based on the reactive power optimization objective function; Each individual in each new population is updated based on the optimal individual and the adaptive inertia weight.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the reactive power optimization method for wind power grid-connected power systems as described in any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the reactive power optimization method for wind power grid-connected power systems as described in any one of claims 1 to 8.

Images