A distributed power generation site selection and sizing planning method and device based on improved Hippo optimization algorithm

Abstract

The present invention discloses a method and device for distributed power site selection and sizing planning based on an improved Hippopotamus optimization algorithm, which relates to the technical field of distribution network planning and is used to solve the technical problem that the existing distributed power site selection and sizing planning methods lead to low efficiency of distributed power site selection and sizing planning. The method includes setting site selection and sizing constraints with the goal of minimizing network active power loss and voltage deviation when distributed power site selection and sizing planning is required, and constructing a distributed power site selection and sizing multi-objective model; based on the K-means clustering algorithm and the multi-objective optimization mechanism, using the improved Hippopotamus optimization algorithm to iteratively solve the distributed power site selection and sizing multi-objective model to generate a distributed power site selection and sizing planning scheme. The present invention utilizes the K-means clustering algorithm to improve sorting efficiency, and combines the improved Hippopotamus optimization algorithm to guide the solution to different target values, thereby improving the efficiency of obtaining distributed power site selection and sizing planning schemes.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network planning, and in particular to a method and device for distributed power source site selection and sizing planning based on an improved Hippo optimization algorithm. Background Art

[0002] Distributed power sources generally refer to a decentralized, non-centralized power generation method, usually referring to small, environmentally compatible power generation devices with a power range of several kilowatts to hundreds of kilowatts, which are used to meet the specific requirements of the power system and users, such as peak load regulation, power supply to remote users or residential areas, saving power supply and distribution investment, and improving power supply reliability.

[0003] When a large number of distributed generation sources randomly connect to or exit the distribution network, it will affect system reliability, relay protection, power quality and system losses, and increase the difficulty of system load forecasting, making the distribution network planning process more difficult. Therefore, system operation planners must re-evaluate the impact of distributed generation sources to implement new load forecasting methods and appropriate optimization algorithms to provide the optimal location and capacity of DG (Distributed Generation).

[0004] Most existing distributed generation (DG) site selection and sizing planning methods employ intelligent optimization algorithms (such as genetic algorithms or particle swarm algorithms). Even for a network with only 50 nodes, completing a complete site selection and sizing optimization calculation often takes over 30 minutes. For a distribution network with 200 nodes, the calculation time can reach several hours. This slow planning process is primarily due to the algorithm's repeated flow calculations to evaluate each candidate solution, each of which takes several seconds. Furthermore, the intelligent algorithm requires thousands or even tens of thousands of iterations to find the optimal solution, resulting in low efficiency in DG site selection and sizing planning. Summary of the Invention

[0005] The present invention provides a distributed power site selection and sizing planning method and device based on an improved Hippo optimization algorithm, which is used to solve the technical problem that the existing distributed power site selection and sizing planning methods lead to low efficiency of distributed power site selection and sizing planning.

[0006] A first aspect of the present invention provides a distributed power generation site selection and sizing planning method based on an improved Hippo optimization algorithm, comprising:

[0007] In response to planning requests, with the goal of minimizing network active power loss and voltage deviation, the site selection and sizing constraints are set to build a multi-objective model for distributed generation site selection and sizing;

[0008] Based on the K-means clustering algorithm and multi-objective optimization mechanism, the improved Hippo optimization algorithm is used to iteratively solve the distributed power generation site selection and sizing multi-objective model to generate a distributed power generation site selection and sizing planning scheme;

[0009] The multi-objective optimization mechanism includes a non-dominated sorting mechanism and a sinusoidal power chaos mapping mechanism; the K-means clustering algorithm and the multi-objective optimization mechanism are used to iteratively solve the distributed power site selection and sizing multi-objective model using an improved Hippo optimization algorithm to generate a distributed power site selection and sizing planning scheme, including:

[0010] Based on the sine power chaotic mapping mechanism, a chaotic sequence is generated;

[0011] An initial hippo population is generated using the chaotic sequence, and based on the distributed power generation site selection and sizing multi-objective model, a fitness vector corresponding to each hippo individual in the initial hippo population is output according to the decision vector corresponding to each hippo individual in the initial hippo population; the fitness vector includes a target value for network active power loss and a target value for voltage deviation; and the decision vector includes the installation location and capacity of the distributed power generation;

[0012] Using the K-means clustering algorithm to cluster the hippopotamus individuals according to the fitness vectors, and outputting a plurality of clusters;

[0013] Analyzing each of the clusters using the non-dominated sorting mechanism to generate an initial non-dominated sorting set at the current moment;

[0014] Screening the initial non-dominated sorted set at the current moment and the target non-dominated sorted set at the historical moment to determine the target non-dominated sorted set at the current moment;

[0015] Selecting, from the network active power loss target values ​​of the fitness vectors in the target non-dominated sorted set at the current moment, a decision vector corresponding to the minimum network active power loss target value as the first vector element;

[0016] Selecting a decision vector corresponding to a minimum voltage deviation target value among the voltage deviation target values ​​of the fitness vectors in the target non-dominated sorted set at the current moment as a second vector element;

[0017] Constructing a multi-target dominant hippopotamus at a current moment according to the first vector element and the second vector element;

[0018] The initial hippo population is updated using the multi-target dominant hippopotamus at the current moment and the multi-target dominant hippopotamus at multiple historical moments to generate an intermediate hippo population, and the number of updates is counted in real time;

[0019] Taking the target non-dominated sorted set at the current moment as the target non-dominated sorted set at the new historical moment;

[0020] Based on the K-means clustering algorithm and the non-dominated sorting mechanism, the distributed power generation site selection and sizing multi-objective model is adopted to generate a new target non-dominated sorting set at the current moment according to the target non-dominated sorting set at the new historical moment and the intermediate hippo population;

[0021] Determine whether the update number reaches a preset number threshold;

[0022] If so, the new target non-dominated sorted set at the current moment is taken as the Pareto optimal solution set.

[0023] Optionally, the multi-objective optimization function corresponding to the distributed generation site selection and sizing multi-objective model is specifically:

[0024] ;

[0025] ;

[0026] ;

[0027] Where X represents the vector of control variables; f loss and f volt They represent network active power loss and voltage deviation respectively; T represents the time of a cycle; t represents the time; b represents the total number of branches; j represents the branch; represents the current of branch j at time t; r j represents the resistance of branch j; v represents the number of nodes; i represents the node; U N Indicates rated voltage; represents the voltage of node i at time t.

[0028] Optionally, the site selection and sizing constraints include power balance constraints, node voltage constraints, branch current constraints, and power flow backflow constraints.

[0029] Optionally, it also includes:

[0030] If the number of updates does not reach the preset number threshold, jumping to the step of selecting the decision vector corresponding to the minimum network active power loss target value from the network active power loss target values ​​of the fitness vectors in the target non-dominated sorted set at the current moment as the first vector element, until the number of updates reaches the preset number threshold;

[0031] The new target non-dominated sorting set at the current moment determined when the number of updates reaches the preset number threshold is used as the Pareto optimal solution set.

[0032] Optionally, the sinusoidal power chaotic mapping mechanism is specifically:

[0033] ;

[0034] Among them, z h represents the hth chaotic sequence; z h+1 represents the h+1th chaotic sequence; and is a random number between 0 and 1; is the Gaussian perturbation of the standard normal distribution; the initial chaotic sequence z1 takes a random number from 0 to 1, and mod is the modulo operation.

[0035] A second aspect of the present invention provides a distributed power site selection and sizing planning device based on an improved Hippo optimization algorithm, which is applied to the above-mentioned distributed power site selection and sizing planning method based on the improved Hippo optimization algorithm, comprising:

[0036] The construction module is used to respond to planning requests, set site selection and sizing constraints with the goal of minimizing network active power loss and voltage deviation, and build a multi-objective model for distributed generation site selection and sizing;

[0037] The solution module is used to iteratively solve the distributed power site selection and sizing multi-objective model based on the K-means clustering algorithm and the multi-objective optimization mechanism using the improved Hippo optimization algorithm to generate a distributed power site selection and sizing planning scheme.

[0038] The third aspect of the present invention provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of the distributed power site selection and sizing planning method based on the improved Hippo optimization algorithm as described in any one of the above items.

[0039] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the steps of the distributed power site selection and sizing planning method based on the improved Hippo optimization algorithm as described in any one of the above items.

[0040] A fifth aspect of the present invention provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions, wherein when the program instructions are executed by a computer, the computer executes the steps of the distributed power site selection and sizing planning method based on the improved Hippo optimization algorithm as described in any one of the above items.

[0041] It can be seen from the above technical solutions that the present invention has the following advantages:

[0042] The above scheme of the present invention provides a distributed power site selection and sizing planning method based on the improved Hippopotamus optimization algorithm. When distributed power site selection and sizing planning is required, the network active power loss and voltage deviation are minimized, site selection and sizing constraints are set, and a distributed power site selection and sizing multi-objective model is constructed; based on the K-means clustering algorithm and the multi-objective optimization mechanism, the improved Hippopotamus optimization algorithm is used to iteratively solve the distributed power site selection and sizing multi-objective model to generate a distributed power site selection and sizing planning scheme. Based on the above scheme, the present invention uses the K-means clustering algorithm to improve the sorting efficiency, and combines the improved Hippopotamus optimization algorithm to guide the solution to different target values, thereby improving the efficiency of obtaining the distributed power site selection and sizing planning scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 A flowchart of the steps of a distributed power generation site selection and sizing planning method based on an improved Hippo optimization algorithm provided in Example 1 of the present invention;

[0045] Figure 2 A schematic diagram of a flow chart of a distributed power generation site selection and sizing planning method based on an improved Hippo optimization algorithm provided in the first embodiment of the present invention;

[0046] Figure 3 This is a structural block diagram of a distributed power generation site selection and sizing planning device based on an improved Hippo optimization algorithm provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0047] The embodiments of the present invention provide a distributed power site selection and sizing planning method and device based on an improved Hippo optimization algorithm, which are used to solve the technical problem that the existing distributed power site selection and sizing planning methods lead to low efficiency of distributed power site selection and sizing planning.

[0048] In order to make the purpose, features, and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0049] See also Figure 1 , Figure 1 A flowchart of the steps of a distributed power generation site selection and sizing planning method based on an improved Hippo optimization algorithm is provided in Example 1 of the present invention.

[0050] The present invention provides a distributed power generation site selection and sizing planning method based on an improved Hippo optimization algorithm, comprising:

[0051] Step 101: In response to a planning request, with the goal of minimizing network active power loss and voltage deviation, setting site selection and sizing constraints, and constructing a distributed generation site selection and sizing multi-objective model.

[0052] The site selection and sizing constraints include power balance constraints, node voltage constraints, branch current constraints, and power flow backflow constraints.

[0053] It should be noted that for the multi-objective optimization function corresponding to the multi-objective model for distributed generation site selection and sizing, the location and capacity of distributed generation connected to the distribution network will have many impacts on the distribution network. The site selection and capacity of distributed generation will affect the voltage value and the magnitude and distribution of current. Among them, the two objectives of system operation network active power loss and voltage deviation can reflect the operating status of distributed generation when it is connected to the distribution network. Therefore, they are jointly constructed into a multi-objective function for analysis. The multi-objective optimization function corresponding to the multi-objective model for distributed generation site selection and sizing is specifically:

[0054] ;

[0055] ;

[0056] ;

[0057] Where X represents the vector of control variables, i.e., the decision vector, including the installation location and capacity of distributed power sources; f loss and f volt They represent network active power loss and voltage deviation respectively; T represents the time of a cycle; t represents the time; b represents the total number of branches; j represents the branch; represents the current of branch j at time t; r j represents the resistance of branch j; v represents the number of nodes; i represents the node; U N Indicates rated voltage; represents the voltage of node i at time t.

[0058] Furthermore, after the distributed power source is connected to the system, it needs to meet the power balance constraint (the current value can be calculated by combining the installation location and capacity of the distributed power source with this constraint). and voltage values ), the specific formula is as follows:

[0059] ;

[0060] in, and They represent the active output and reactive output of the distributed generation at node i respectively; and They represent the active load and reactive load of node i respectively; l represents the neighboring nodes of node i; 、 and They represent the conductance, susceptance and phase angle difference between node i and its neighbor node l respectively; is the voltage at node i; is the voltage of neighbor node l.

[0061] Furthermore, for node voltage constraints, the voltage of each node in the distribution network should satisfy:

[0062] ;

[0063] Among them, U min and U max Represent the lower voltage limit and upper voltage limit respectively.

[0064] Furthermore, for branch current constraints, the current of each branch in the distribution network should satisfy:

[0065] ;

[0066] Among them, I j,max Indicates the upper limit of the current in branch j.

[0067] Furthermore, for the power flow backfeed constraint, after the distributed generation is connected to the distribution network, the power it generates should be consumed locally and should not be backfeeded to the upper power grid. The specific formula is as follows:

[0068] ;

[0069] Among them, b f Represents the set of branches connecting the substation; represents the current in branch j at time t.

[0070] Step 102: Based on the K-means clustering algorithm and the multi-objective optimization mechanism, the improved Hippo optimization algorithm is used to iteratively solve the distributed generation site selection and sizing multi-objective model to generate a distributed generation site selection and sizing planning scheme.

[0071] The multi-objective optimization mechanism includes the non-dominated sorting mechanism and the sine-powered chaotic mapping mechanism, namely the SPM (Sine-Powered Map) chaotic mapping.

[0072] K-means clustering algorithm is a K-means clustering algorithm.

[0073] It should be noted that the traditional Hippo optimization algorithm is a single-objective optimization form. To achieve multi-objective optimization, the fitness function is expanded from a scalar to a vector form. That is, the fitness value output by the present invention is in vector form, which is composed of the network active loss and voltage deviation as the corresponding target values. The fitness function is specifically as follows:

[0074] ;

[0075] Among them, F is the fitness function, f m Represents the mth target value, which includes the network active power loss target value and the voltage deviation target value.

[0076] Specifically, step 102 may include the following sub-steps:

[0077] Step S21: Generate a chaotic sequence based on a sine power chaotic mapping mechanism;

[0078] It should be noted that in the traditional Hippo optimization algorithm, the generation of the initial population relies on the generation of random numbers. Although this method can ensure the randomness of the Hippo population to a certain extent, its randomness has limitations and may lead to insufficient population diversity. To solve this problem, the present invention introduces SPM chaotic mapping to improve the quality of the initial population. Its mathematical formula is as follows:

[0079] The sine power chaotic mapping mechanism is as follows:

[0080] ;

[0081] ;

[0082] Among them, z h represents the hth chaotic sequence; z h+1 represents the h+1th chaotic sequence; and is a random number between 0 and 1; is the Gaussian perturbation of the standard normal distribution; the initial chaotic sequence z1 takes a random number from 0 to 1, and mod is the modulus operation; is the i-th hippopotamus individual (i-th solution) in the initial hippopotamus population, representing the decision vector; X min and X max represent the lower and upper limits of the control variables in the vector, respectively.

[0083] Step S22: Generate an initial hippo population using a chaotic sequence, and output a fitness vector corresponding to each hippo individual in the initial hippo population based on the decision vector corresponding to each hippo individual in the distributed power generation site selection and sizing multi-objective model; the fitness vector includes a target value for network active power loss and a target value for voltage deviation; and the decision vector includes the installation location and capacity of the distributed power generation;

[0084] It should be noted that each hippopotamus individual will carry a decision vector, which includes two decision variables, namely the installation location of the distributed power source and the capacity of the distributed power source. Based on the installation location of the distributed power source and the capacity of the distributed power source, the current value and the voltage value are determined. The distributed power source site selection and capacity determination multi-objective model is used to calculate the network active power loss target value and voltage deviation target value corresponding to each hippopotamus individual according to the current value and voltage value corresponding to each hippopotamus individual, thereby forming a fitness vector.

[0085] Step S23: clustering the hippopotamus individuals according to their fitness vectors using the K-means clustering algorithm, and outputting multiple clusters;

[0086] It should be noted that when updating the archive, the K-means clustering algorithm is used to group candidate solutions (each hippopotamus individual). By calculating similarity metrics such as Euclidean distance, the solutions are divided into K clusters, and the solutions within each cluster have similarity in the target value. The goal of the K-means algorithm is to minimize the sum of squared errors within the cluster, and its mathematical expression is:

[0087] ;

[0088] Among them, C k represents the kth cluster after clustering; is the i-th solution, i.e. the i-th decision vector; Represents the center of the k-th cluster, and the calculation formula is as follows:

[0089] ;

[0090] The above method is used to divide the solutions into multiple clusters according to the similarity of the target values. That is, the K-means clustering algorithm is used to cluster the hippopotamus individuals according to their fitness vectors, and multiple clusters are output, so as to quickly identify solution groups with similar characteristics and improve the screening efficiency.

[0091] Step S24: Analyze each cluster using a non-dominated sorting mechanism to generate an initial non-dominated sorted set at the current moment;

[0092] It should be noted that after K-means clustering, in order to ensure that the final selected solution is both diverse and close to the Pareto front, a non-dominated sorting method (non-dominated sorting mechanism) is used. The specific steps are as follows:

[0093] 1) Put the objects in each cluster into an unsorted set.

[0094] 2) Traverse all solutions and find those that are not dominated by any other solution, forming the first level front1 (that is, the initial non-dominated sorted set at the current moment). For each subsequent level front L, find those that are only dominated by Continue the above steps until all solutions are assigned to a level.

[0095] Step S25: Screen the initial non-dominated sorted set at the current moment and the target non-dominated sorted set at the historical moment to determine the target non-dominated sorted set at the current moment;

[0096] It should be noted that the initial non-dominated sorted set at the current moment includes multiple solutions that are not dominated by any other solutions, namely the selected hippopotamus individuals. Finally, the initial non-dominated sorted set at the current moment and the target non-dominated sorted set at the historical moment are screened to determine the target non-dominated sorted set at the current moment. For example, assuming that the initial non-dominated sorted set at the current moment includes hippopotamus individuals A, B and C, and the target non-dominated sorted set at the historical moment includes hippopotamus individuals D, E and F, compare the target values ​​corresponding to any two hippopotamus individuals (including two objective function values, namely, the network active power loss target value and the voltage deviation target value). If the network values ​​corresponding to hippopotamus individuals A and D are the same, then the target values ​​are the same. If the target values ​​of network active loss are equal, but the voltage deviation target values ​​corresponding to hippopotamus individuals A and D are not equal, and the voltage deviation target value corresponding to hippopotamus individuals A is smaller than the voltage deviation target value corresponding to hippopotamus individuals D, then hippopotamus individuals D will be eliminated, and hippopotamus individuals A will be retained in the target non-dominated sorted set at the current moment; if the target values ​​corresponding to hippopotamus individuals A and D are not equal, then both hippopotamus individuals A and Hippopotamus individuals D will be retained in the target non-dominated sorted set at the current moment, and then the target value corresponding to the retained hippopotamus individuals will continue to be compared with the target values ​​corresponding to other hippopotamus individuals until the target values ​​corresponding to all solutions (hippopotamus individuals) are compared, thereby obtaining the target non-dominated sorted set at the current moment.

[0097] Step S26: Selecting the decision vector corresponding to the minimum network active power loss target value from the network active power loss target values ​​of the fitness vectors in the target non-dominated sorted set at the current moment as the first vector element;

[0098] Step S27: Selecting the decision vector corresponding to the minimum voltage deviation target value from the voltage deviation target values ​​of the fitness vectors in the target non-dominated sorted set at the current moment as the second vector element;

[0099] It should be noted that in the traditional Hippopotamus optimization algorithm, the search process in the first phase relies on the globally optimal Hippopotamus as the dominant Hippopotamus to guide other solutions to move closer to it. However, in multi-objective optimization problems, there is no single globally optimal solution, but rather a set of Pareto optimal solutions. Therefore, the dominant Hippopotamus is expanded into a vector consistent with the number of targets, with each vector element representing the optimal value of a different target. The present invention selects the decision vector corresponding to the minimum network active power loss target value as the first vector element from the voltage deviation target value of the fitness vector corresponding to each Hippopotamus individual in the target non-dominated sorted set at the current moment, and selects the decision vector corresponding to the minimum voltage deviation target value as the second vector element from the voltage deviation target value of the fitness vector corresponding to each Hippopotamus individual in the target non-dominated sorted set at the current moment. For example, the target non-dominated sorting set at the current moment includes hippopotamus individual A and hippopotamus individual B. Hippopotamus individual A and hippopotamus individual B both have corresponding network active power loss target values ​​and voltage deviation target values. If the network active power loss target value corresponding to hippopotamus individual A is smaller than the network active power loss target value corresponding to hippopotamus individual B, then the decision vector corresponding to hippopotamus individual A is selected as the first vector element. Similarly, the second vector element can be obtained.

[0100] Step S28: construct a multi-target dominant hippopotamus at the current moment according to the first vector element and the second vector element;

[0101] It should be noted that the process of constructing the multi-target dominant hippopotamus at the current moment can be expressed as:

[0102] ;

[0103] ;

[0104] Among them, X D Indicates a dominant hippopotamus; X D,m Represents the mth dominant hippopotamus, that is, a vector element, including: when m is 1, it is the first vector element, and when m is 2, it is the second vector element.

[0105] Step S29, using the multi-target dominant hippopotamus at the current moment and the multi-target dominant hippopotamus at multiple historical moments to update the initial hippopotamus population, generate an intermediate hippopotamus population, and count the number of updates in real time;

[0106] It should be noted that the process of updating the positions of female hippos and juvenile hippos in the initial hippo population is consistent with the existing update process, while the male hippos in the initial hippo population will randomly select a dominant hippo (that is, randomly select a dominant hippo from the multiple dominant hippos at the current moment and the multiple dominant hippos at multiple historical moments) for position update. The specific formula is as follows:

[0107] ;

[0108] Among them, X M,i represents the i-th male hippopotamus; and is a learning factor that controls the speed at which the hippopotamus approaches the dominant hippopotamus and the intensity of random disturbances; the choice function represents randomly selecting a dominant hippopotamus from the multi-target dominant hippopotamus at the current moment and the multi-target dominant hippopotamus at multiple historical moments.

[0109] Step S210: Use the target non-dominated sorted set at the current moment as the target non-dominated sorted set at the new historical moment;

[0110] Step S211: Based on the K-means clustering algorithm and the non-dominated sorting mechanism, a distributed generation site selection and sizing multi-objective model is adopted to generate a new target non-dominated sorted set at the current moment according to the target non-dominated sorted set at the new historical moment and the intermediate hippo population;

[0111] Step S212: determine whether the update times reaches a preset times threshold;

[0112] Step S213: If yes, then the new target non-dominated sorted set at the current moment is used as the Pareto optimal solution set.

[0113] Step S214: Generate a distributed generation site selection and sizing planning scheme based on the Pareto optimal solution set.

[0114] It's important to note that the traditional Hippo optimization algorithm outputs a single global optimal solution. However, in the context of multi-objective optimization, there's no single global optimal solution. Instead, there's a set of solutions known as Pareto optimal solutions. Therefore, the algorithm needs to be adjusted to output a Pareto optimal solution set.

[0115] After the improved Hippo optimization algorithm is solved, the Pareto optimal solutions it outputs include:

[0116] 1) Decision variable matrix, an n×s matrix, where n is the number of Pareto optimal solutions and s is the dimension of the decision variables. Specifically:

[0117] ;

[0118] in, is the matrix element in the nth row and sth column of the decision variable matrix, representing the decision variable.

[0119] 2) Objective function value matrix , an n×m matrix:

[0120] ;

[0121] in, is the matrix element in the nth row and mth column of the objective function value matrix, representing the objective function value, that is, the target value of network active power loss and the target value of voltage deviation; m is the dimension of the objective function value matrix.

[0122] Through these improvements, the multi-objective Hippo optimization algorithm can effectively output a Pareto-optimal solution set, providing comprehensive reference information for decision makers. Finally, planners select the appropriate solution (i.e., the installation location and capacity of the distributed generation) from the Pareto-optimal solution set based on the actual situation, resulting in a distributed generation site selection and sizing plan.

[0123] Optionally, it also includes:

[0124] If the number of updates does not reach the preset threshold, the process jumps to the step of selecting the decision vector corresponding to the minimum target network active power loss value from the target network active power loss values ​​of each fitness vector in the target non-dominated sorted set at the current moment as the first vector element, until the number of updates reaches the preset threshold;

[0125] The new target non-dominated sorting set at the current moment determined when the number of updates reaches a preset number threshold is taken as the Pareto optimal solution set.

[0126] It should be noted that if the number of updates does not reach the preset threshold, the process jumps to step S26 until the number of updates reaches the preset threshold, and the new target non-dominated sorted set at the current moment determined when the number of updates reaches the preset threshold is used as the Pareto optimal solution set. Finally, a suitable solution is selected from the Pareto optimal solution set according to demand as the distributed power site selection and sizing planning scheme.

[0127] To compare technical performance, existing technologies can be used as a reference. With the increasing penetration of renewable energy in distribution networks, the siting and sizing of distributed generation (DGs) have become critical for optimizing power system operation. Although various approaches have been proposed, existing technologies still have significant shortcomings in multi-objective collaborative optimization and computational efficiency. While existing research has proposed multi-objective models targeting network losses and voltage quality, the dimensions of the objectives vary significantly, and the weight assignment relies on manual experience, making it difficult to achieve global optimality in the optimization results. Current mainstream optimization algorithms, such as genetic algorithms and particle swarm optimization, are prone to local optima or premature convergence when solving complex nonlinear models. The Hippo optimization algorithm, a novel metaheuristic optimization algorithm, utilizes a unique balancing mechanism between exploration and exploitation, enabling it to quickly and efficiently find effective solutions within the search space. However, this algorithm is a single-objective optimization algorithm and cannot be directly applied to solving multi-objective models. This patent proposes a multi-objective site selection and sizing method based on the improved Hippo optimization algorithm. By introducing SPM chaotic mapping, K-means clustering algorithm, non-dominated sorting mechanism, and multi-objective dominant Hippo, it significantly improves the algorithm convergence, solution set diversity and model practicality, effectively overcoming the limitations of existing technologies.

[0128] Furthermore, when selecting the site and sizing the distributed power generation, it is necessary to improve the voltage quality and reduce network losses as much as possible. However, the current solutions are often converted into single-objective optimization through weighted form when solving using algorithms. When multiple constraints conflict with each other, the single-objective optimization cannot effectively balance them and may fall into no solution or suboptimal solution.

[0129] To address the above issues, this paper proposes a distributed power generation site selection and sizing planning method based on an improved Hippo optimization algorithm. The objective function is defined as a multi-objective function, the traditional Hippo optimization algorithm is expanded into a multi-objective optimization algorithm, the SPM chaotic mapping is introduced to improve the quality of the initialization population, the archive and external population are combined to ensure that excellent solutions are not lost, the non-dominated sorting is combined, and the algorithm's search strategy is adjusted to adapt to the characteristics of the multi-objective problem. Ultimately, the multi-objective optimization configuration of distributed power generation can be achieved. Specifically, please refer to Figure 2 First, a mathematical model for distributed generation site selection and sizing is constructed. Next, the upper and lower bounds of the variables are determined. The lower bound of the site selection variable is 1, the upper bound is the number of nodes, and the variable is an integer. The lower bound of the sizing variable is 0, and the upper bound depends on the actual situation. The variable is continuous. Next, the parameters of the Hippo optimization algorithm are set, specifically the population size, number of iterations, and number of archives. Finally, the improved Hippo optimization algorithm is used to solve the problem and generate a Pareto optimal solution set. Planners can then select the appropriate solution based on the actual situation.

[0130] In summary, in the distributed power site selection and sizing planning method based on the improved Hippo optimization algorithm proposed in the present invention, a multi-objective optimization model for distributed power site selection and sizing is constructed, and a flow backflow constraint is introduced into it to ensure that the distributed power can be locally consumed after access. In view of the fact that the traditional Hippo optimization algorithm cannot solve multi-objective optimization problems, the single objective is expanded to a multi-objective form, an external archiving and non-dominated sorting mechanism is introduced, and the K-means clustering algorithm is used to improve the sorting efficiency. The solutions are divided into multiple clusters according to the similarity of the target values, thereby improving the efficiency of non-dominated sorting. Combined with the improved first-stage search strategy of the Hippo optimization algorithm, it is ensured that the solutions can move closer to different objectives, thereby improving the efficiency of obtaining the Pareto optimal set. In addition, the SPM chaotic mapping is introduced to improve the quality of the initialized population, accelerate the convergence efficiency of the algorithm, and improve the randomness and traversal of the population, thereby strengthening the global search capability. Finally, the improved Hippo optimization algorithm is used to solve the distributed power site selection and sizing model, and the Pareto optimal set of distributed power site selection and sizing is generated. Compared with the single-objective optimization algorithm, the present invention can find a set of Pareto optimal solutions to fully reflect the trade-offs between different objectives. Planners can select the most appropriate distributed site selection and sizing scheme from the Pareto optimal solution set according to specific needs. At the same time, the multi-objective optimization configuration of distributed power sources is realized, which can effectively reduce the network loss of the system and improve the voltage quality.

[0131] In an embodiment of the present invention, the present invention provides a distributed power site selection and sizing planning method based on an improved Hippopotamus optimization algorithm. When distributed power site selection and sizing planning is required, first, with the goal of minimizing network active power loss and voltage deviation, site selection and sizing constraints are set, and a distributed power site selection and sizing multi-objective model is constructed; based on the K-means clustering algorithm and the multi-objective optimization mechanism, the improved Hippopotamus optimization algorithm is used to iteratively solve the distributed power site selection and sizing multi-objective model to generate a distributed power site selection and sizing planning scheme. Based on the above scheme, the present invention uses the K-means clustering algorithm to improve the sorting efficiency, and combines the improved Hippopotamus optimization algorithm to guide the solution to different target values, thereby improving the efficiency of obtaining the distributed power site selection and sizing planning scheme.

[0132] See also Figure 3 , Figure 3 This is a structural block diagram of a distributed power generation site selection and sizing planning device based on an improved Hippo optimization algorithm provided in Example 2 of the present invention.

[0133] The present invention provides a distributed power site selection and sizing planning device based on an improved Hippo optimization algorithm, which is applied to the above-mentioned distributed power site selection and sizing planning method based on an improved Hippo optimization algorithm, comprising:

[0134] A construction module 301 is used to respond to planning requests, set site selection and sizing constraints with the goal of minimizing network active power loss and voltage deviation, and construct a multi-objective model for distributed generation site selection and sizing;

[0135] A solution module 302 is configured to iteratively solve the distributed generation site selection and sizing multi-objective model using an improved Hippo optimization algorithm based on a K-means clustering algorithm and a multi-objective optimization mechanism, and generate a distributed generation site selection and sizing planning scheme;

[0136] The solution module 302 is specifically configured to generate a chaotic sequence based on a sine power chaotic mapping mechanism;

[0137] The initial hippo population is generated using a chaotic sequence. Based on the distributed power generation site selection and sizing multi-objective model, the fitness vector corresponding to each hippo individual in the initial hippo population is output according to the decision vector corresponding to each hippo individual in the initial hippo population. The fitness vector includes the target value of network active power loss and the target value of voltage deviation. The decision vector includes the installation location and capacity of the distributed power generation.

[0138] The K-means clustering algorithm is used to cluster the hippopotamus individuals according to their fitness vectors and output multiple clusters;

[0139] The non-dominated sorting mechanism is used to analyze each cluster and generate the initial non-dominated sorting set at the current moment;

[0140] The initial non-dominated sorted set at the current moment and the target non-dominated sorted set at the historical moment are screened to determine the target non-dominated sorted set at the current moment;

[0141] Among the network active power loss target values ​​of each fitness vector in the target non-dominated sorted set at the current moment, the decision vector corresponding to the minimum network active power loss target value is selected as the first vector element;

[0142] Among the voltage deviation target values ​​of each fitness vector in the target non-dominated sorted set at the current moment, the decision vector corresponding to the minimum voltage deviation target value is selected as the second vector element;

[0143] Construct the multi-target dominant hippopotamus at the current moment according to the first vector element and the second vector element;

[0144] The initial hippo population is updated using the multi-target dominant hippopotamus at the current moment and the multi-target dominant hippopotamus at multiple historical moments to generate an intermediate hippo population, and the number of updates is counted in real time;

[0145] The target non-dominated sorting set at the current moment is used as the target non-dominated sorting set at the new historical moment;

[0146] Based on the K-means clustering algorithm and non-dominated sorting mechanism, a distributed generation site selection and sizing multi-objective model is adopted to generate a new target non-dominated sorting set at the current moment according to the target non-dominated sorting set at the new historical moment and the intermediate hippo population;

[0147] Determine whether the number of updates reaches a preset threshold;

[0148] If so, the new current-time target non-dominated sorting set is taken as the Pareto optimal solution set;

[0149] Based on the Pareto optimal solution set, a distributed generation site selection and sizing planning scheme is generated.

[0150] Furthermore, the multi-objective optimization function corresponding to the multi-objective model of distributed generation site selection and sizing is specifically:

[0151] ;

[0152] ;

[0153] ;

[0154] Where X represents the vector of control variables; f loss and f volt They represent network active power loss and voltage deviation respectively; T represents the time of a cycle; t represents the time; b represents the total number of branches; j represents the branch; represents the current of branch j at time t; r j represents the resistance of branch j; v represents the number of nodes; i represents the node; U N Indicates rated voltage; represents the voltage of node i at time t.

[0155] Furthermore, the site selection and sizing constraints include power balance constraints, node voltage constraints, branch current constraints, and power flow backflow constraints.

[0156] In an optional embodiment of the device, the device further comprises:

[0157] The first module is configured to, if the number of updates does not reach a preset number threshold, jump to the step of selecting the decision vector corresponding to the minimum network active power loss target value from the network active power loss target values ​​of the fitness vectors in the target non-dominated sorted set at the current moment as the first vector element, until the number of updates reaches the preset number threshold;

[0158] The second module is used to use the new target non-dominated sorting set at the current moment determined when the number of updates reaches a preset number threshold as the Pareto optimal solution set.

[0159] Furthermore, the sine power chaotic mapping mechanism is specifically:

[0160] ;

[0161] Among them, z h represents the hth chaotic sequence; z h+1 represents the h+1th chaotic sequence; and is a random number between 0 and 1; is the Gaussian perturbation of the standard normal distribution; the initial chaotic sequence z1 takes a random number from 0 to 1, and mod is the modulo operation.

[0162] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0163] An embodiment of the present invention also provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory; when the computer program is executed by the processor, the processor executes the steps of the distributed power site selection and sizing planning method based on the improved Hippo optimization algorithm as in any of the above embodiments.

[0164] An embodiment of the present invention also provides a computer-readable storage medium having a computer program / instruction stored thereon. When the computer program / instruction is executed by a processor, the steps of the distributed power site selection and sizing planning method based on the improved Hippo optimization algorithm as in any of the above embodiments are implemented.

[0165] An embodiment of the present invention also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the distributed power site selection and sizing planning method based on the improved Hippo optimization algorithm as in any of the above embodiments.

[0166] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0167] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0168] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A distributed power generation site selection and sizing planning method based on an improved Hippo optimization algorithm, characterized in that: include: In response to planning requests, with the goal of minimizing network active power loss and voltage deviation, the site selection and sizing constraints are set to build a multi-objective model for distributed generation site selection and sizing; Based on the K-means clustering algorithm and multi-objective optimization mechanism, the improved Hippo optimization algorithm is used to iteratively solve the distributed power generation site selection and sizing multi-objective model to generate a distributed power generation site selection and sizing planning scheme; The multi-objective optimization mechanism includes a non-dominated sorting mechanism and a sinusoidal power chaos mapping mechanism; the K-means clustering algorithm and the multi-objective optimization mechanism are used to iteratively solve the distributed power site selection and sizing multi-objective model using an improved Hippo optimization algorithm to generate a distributed power site selection and sizing planning scheme, including: Based on the sine power chaotic mapping mechanism, a chaotic sequence is generated; An initial hippo population is generated using the chaotic sequence, and based on the distributed power generation site selection and sizing multi-objective model, a fitness vector corresponding to each hippo individual in the initial hippo population is output according to the decision vector corresponding to each hippo individual in the initial hippo population; the fitness vector includes a target value for network active power loss and a target value for voltage deviation; and the decision vector includes the installation location and capacity of the distributed power generation; Using the K-means clustering algorithm to cluster the hippopotamus individuals according to the fitness vectors, and outputting a plurality of clusters; Analyzing the plurality of clusters using the non-dominated sorting mechanism to generate an initial non-dominated sorting set at a current moment; Screening the initial non-dominated sorted set at the current moment and the target non-dominated sorted set at the historical moment to determine the target non-dominated sorted set at the current moment; Selecting, from the network active power loss target values ​​of the fitness vectors in the target non-dominated sorted set at the current moment, a decision vector corresponding to the minimum network active power loss target value as the first vector element; Selecting a decision vector corresponding to a minimum voltage deviation target value among the voltage deviation target values ​​of the fitness vectors in the target non-dominated sorted set at the current moment as a second vector element; Constructing a multi-target dominant hippopotamus at a current moment according to the first vector element and the second vector element; The initial hippo population is updated using the multi-target dominant hippopotamus at the current moment and the multi-target dominant hippopotamus at multiple historical moments to generate an intermediate hippo population, and the number of updates is counted in real time; Taking the target non-dominated sorted set at the current moment as the target non-dominated sorted set at the new historical moment; Based on the K-means clustering algorithm and the non-dominated sorting mechanism, the distributed power generation site selection and sizing multi-objective model is adopted to generate a new target non-dominated sorting set at the current moment according to the target non-dominated sorting set at the new historical moment and the intermediate hippo population; Determine whether the update number reaches a preset number threshold; If so, the new target non-dominated sorting set at the current moment is taken as the Pareto optimal solution set; Based on the Pareto optimal solution set, a distributed power generation site selection and sizing planning scheme is generated.

2. The distributed power generation site selection and sizing planning method based on the improved Hippo optimization algorithm according to claim 1 is characterized in that: The multi-objective optimization function corresponding to the distributed generation site selection and sizing multi-objective model is specifically: ； ； ； Where X represents the vector of control variables; f loss and f volt They represent network active power loss and voltage deviation respectively; T represents the time of a cycle; t represents the time; b represents the total number of branches; j represents the branch; represents the current of branch j at time t; r j represents the resistance of branch j; v represents the number of nodes; i represents the node; U N Indicates rated voltage; represents the voltage of node i at time t.

3. The distributed power generation site selection and sizing planning method based on the improved Hippo optimization algorithm according to claim 1 is characterized in that: The site selection and sizing constraints include power balance constraints, node voltage constraints, branch current constraints, and power flow backflow constraints.

4. The distributed power generation site selection and sizing planning method based on the improved Hippo optimization algorithm according to claim 1 is characterized in that: Also includes: If the number of updates does not reach the preset number threshold, jumping to the step of selecting the decision vector corresponding to the minimum network active power loss target value from the network active power loss target values ​​of the fitness vectors in the target non-dominated sorted set at the current moment as the first vector element, until the number of updates reaches the preset number threshold; The new target non-dominated sorting set at the current moment determined when the number of updates reaches the preset number threshold is used as the Pareto optimal solution set.

5. The distributed power generation site selection and sizing planning method based on the improved Hippo optimization algorithm according to claim 1 is characterized in that: The sinusoidal power chaotic mapping mechanism is specifically: ； Among them, z h represents the hth chaotic sequence; z h+1 represents the h+1th chaotic sequence; and is a random number between 0 and 1; is the Gaussian perturbation of the standard normal distribution; the initial chaotic sequence z1 takes a random number from 0 to 1, and mod is the modulo operation.

6. A distributed power site selection and sizing planning device based on an improved Hippo optimization algorithm, applied to the distributed power site selection and sizing planning method based on an improved Hippo optimization algorithm according to claim 1, characterized in that: include: The construction module is used to respond to planning requests, set site selection and sizing constraints with the goal of minimizing network active power loss and voltage deviation, and build a multi-objective model for distributed generation site selection and sizing; The solution module is used to iteratively solve the distributed power site selection and sizing multi-objective model based on the K-means clustering algorithm and the multi-objective optimization mechanism using the improved Hippo optimization algorithm to generate a distributed power site selection and sizing planning scheme.

7. A computer device, characterized in that: It includes a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of the distributed power site selection and sizing planning method based on the improved Hippo optimization algorithm as described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the distributed power site selection and sizing planning method based on the improved Hippo optimization algorithm as described in any one of claims 1 to 5 is implemented.

9. A computer program product, characterized in that The computer program product includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions, wherein when the program instructions are executed by a computer, the computer is caused to execute the distributed power site selection and sizing planning method based on the improved Hippo optimization algorithm as described in any one of claims 1 to 5.

Images