Improved Pareto method and system for optimizing comprehensive index of distribution network region division

Abstract

The invention discloses an improved Pareto method and system for distribution network region division comprehensive index optimization. The method comprises the steps that distribution network region division comprehensive indexes are determined, and a genetic algorithm and related parameters of Pareto optimization are set; carrying out chromosome coding of the distribution network based on the adjacent matrix, and carrying out crossover variation genetic operation of chromosomes of the distribution network; establishing a Pareto optimization objective function according to the distribution network region division comprehensive index, obtaining a Pareto optimization solution set through a Pareto dominating relation and a non-inferior solution, and obtaining a Pareto leading edge through a genetic algorithm crossover variation optimization objective function; and carrying out normalization processing on the Pareto leading edge, selecting a Pareto optimal solution with the closest index values according to a normalization result, and outputting the result. According to the method, the optimal or approximately optimal comprehensive indexes can be objectively obtained, and the comprehensive indexes are not gathered near a single index, so that the partitioning result is more beneficial to optimal operation of the power distribution network.

Description

Technical Field

[0001] This invention relates to the field of power distribution network area division technology, and in particular to an improved Pareto method and system for optimizing comprehensive indicators of power distribution network area division. Background Technology

[0002] With the large-scale integration of distributed renewable energy into the distribution network, the difficulty of centralized control and operation of the distribution network and the dimensionality curse have increased. In order to solve the above problems, regional autonomy based on regional division has been widely studied and applied. Regional autonomy can reduce the difficulty of control and operation and is an effective means to achieve efficient and accurate operation of complex distribution networks.

[0003] Regional division involves active and reactive power balance, voltage regulation capability, and net load fluctuation indicators within the region. It is the basis for achieving optimized operation of distribution networks based on regional autonomy. However, the current comprehensive indicators for regional division use artificially assigned weight coefficients for each objective function, which lacks objectivity and makes it difficult to achieve optimal or near-optimal operating results in actual operation. Summary of the Invention

[0004] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0005] In view of the aforementioned existing problems, this invention is proposed. Therefore, this invention provides an improved Pareto method for optimizing comprehensive indicators in distribution network area division, addressing the problem of how to optimize comprehensive indicators in distribution network area division.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides an improved Pareto method for optimizing the comprehensive index of distribution network area division, comprising:

[0008] Determine the comprehensive index for distribution network area division, and set the relevant parameters for genetic algorithm and Pareto optimization;

[0009] Chromosome encoding for the distribution network is performed based on the adjacency matrix, and crossover and mutation inheritance operations are performed on the chromosomes of the distribution network.

[0010] Based on the comprehensive index of distribution network area division, a Pareto optimization objective function is established. The Pareto optimization solution set is obtained through Pareto dominance relationship and non-dominated solution. The Pareto front is obtained by optimizing the objective function through crossover and mutation of genetic algorithm.

[0011] The Pareto front is normalized, and the Pareto optimal solution with the closest values ​​for each index is selected based on the normalization result and the result is output.

[0012] As a preferred embodiment of the improved Pareto method for optimizing the comprehensive index of distribution network area division according to the present invention, the following steps are taken: determining the comprehensive index of distribution network area division, and setting relevant parameters for the genetic algorithm and Pareto optimization, including...

[0013] The comprehensive indicators for the division of distribution network areas include net load fluctuation indicators, modularity indicators, reactive power balance indicators, and active power balance indicators.

[0014] Set the Pareto iteration number it = 1, and the maximum iteration number is it. max The critical threshold ε is initialized to a very large value.

[0015] As a preferred embodiment of the improved Pareto method for optimizing the comprehensive index of distribution network area division as described in this invention, it further includes:

[0016] The net load volatility index is the sum of the absolute values ​​of net load volatility for each distribution network area. The calculation formula is as follows:

[0017]

[0018]

[0019] ΔP adj,z =ΔP G,z +ΔP BESS,z +ΔP MG,z

[0020] Where, n z For the number of regions, This is a net load fluctuation indicator. Let ΔP be the net load fluctuation index for distribution network area z. L,z Net load is the sum of fluctuations in renewable energy output and load, ΔP. adj,z ΔP represents the regulated power of dispatchable distributed voltage within the distribution network area z. BESS,z ΔP represents the regulation power of the energy storage batteries within the distribution network area z. MG,z ΔP represents the adjustable power of the microgrid within the distribution network area z. G,z The adjustable power of dispatchable distributed power sources within the distribution network area z;

[0021] Modularity index, which measures the degree of electrical connection between nodes (busbars) within a distribution network area, is calculated using the following formula:

[0022]

[0023] Where ρ is the modularity index, m is the coefficient, and eij k represents the weight of the edge between nodes (buses) i and j. i =∑ i e ij The sum of the weights of the edges connected to the bus node (bus) i;

[0024] The formula for calculating the reactive power balance index is:

[0025]

[0026]

[0027] in, Q is an indicator of reactive power balance. z Q represents the reactive power balance of distribution network area z. sup Q represents the maximum reactive power supply within the distribution network area. ned For reactive power demand within the distribution network area;

[0028] The formula for calculating the active power balance index is:

[0029]

[0030]

[0031] in, As an indicator of active power balance, The desired active power of a dispatchable distributed power source. For the desired power of the energy storage battery, For the expected active power of the microgrid, Expected active power for renewable energy sources This represents the expected active power of the load.

[0032] As a preferred embodiment of the improved Pareto method for optimizing the comprehensive index of distribution network area division as described in this invention, the method includes: encoding the chromosomes of the distribution network based on the adjacency matrix, and performing crossover and mutation genetic operations on the chromosomes of the distribution network, including...

[0033] A chromosome represents an adjacency matrix of a distribution network. If the i-th and j-th columns of the adjacency matrix are 0, it means that nodes (buses) i and j are not connected. If they are 1, it means that nodes (buses) i and j are connected.

[0034] Genetic algorithms are used to perform chromosome crossover mutation on the distribution network, and the maximum value of the comprehensive index for each distribution network area is calculated.

[0035] As a preferred embodiment of the improved Pareto method for optimizing the comprehensive index of distribution network area division according to the present invention, the method includes: establishing a Pareto optimization objective function based on the comprehensive index of distribution network area division; obtaining a Pareto optimization solution set through Pareto dominance relations and non-dominated solutions; and obtaining the Pareto front through crossover and mutation optimization of the objective function using a genetic algorithm.

[0036] The Pareto objective function is expressed as:

[0037] maxF = (f1, f2, f3, f4)

[0038] in, The net load fluctuation index is f2 = ρ, where ρ is the modularity index. As an indicator of reactive power balance, The active power balance index;

[0039] The objective function is calculated using a genetic algorithm with crossover and mutation to obtain the Pareto optimization in the i-th iteration, expressed as:

[0040]

[0041] Where i = 1, 2, ..., n, and n is the number of solutions. Let ρ be the net load fluctuation index value of the i-th solution in the it-th iteration. it,i Let be the modularity index value of the i-th solution in the it-th iteration. Let be the reactive power balance index value of the i-th solution in the it-th iteration. Let be the active power balance index value of the i-th solution in the it-th iteration.

[0042] As a preferred embodiment of the improved Pareto method for optimizing the comprehensive index of distribution network area division according to the present invention, the Pareto front is normalized, including:

[0043] The Pareto front normalized representation is:

[0044]

[0045] Where i = 1, 2, ..., n, and n is the number of solutions. Let ρ be the net load fluctuation index value of the i-th solution in the it-th iteration. it,i Let be the modularity index value of the i-th solution in the it-th iteration. Let be the reactive power balance index value of the i-th solution in the it-th iteration. Let be the active power balance index value of the i-th solution in the it-th iteration;

[0046] The absolute value of the differences between the Pareto optimization metrics after the it-th iteration is used to obtain the Pareto front closest to each metric after Pareto normalization. The calculation formula is as follows:

[0047]

[0048] Where i = 1, 2, ..., n, and n is the number of solutions. Let ρ be the net load fluctuation index value of the i-th solution in the it-th iteration. it,i Let be the modularity index value of the i-th solution in the it-th iteration. Let be the reactive power balance index value of the i-th solution in the it-th iteration. Let be the active power balance index value of the i-th solution in the it-th iteration.

[0049] As a preferred embodiment of the improved Pareto method for optimizing the comprehensive index of distribution network area division according to the present invention, the method includes: obtaining the Pareto optimal solution based on the normalized result and outputting the result, including...

[0050] If the normalized Pareto front index is lower than the critical threshold ε, then update the critical threshold and retain the Pareto optimization result of the current iteration. If it has not reached the maximum number of iterations, continue to calculate the normalized Pareto front of each index. Otherwise, output the Pareto front corresponding to the critical threshold ε.

[0051] If the normalized Pareto front index is higher than the critical threshold ε and it has not reached the maximum number of iterations, continue to calculate the normalized Pareto front for each index; otherwise, output the Pareto front corresponding to the critical threshold ε.

[0052] Secondly, this invention provides an improved Pareto system for optimizing the comprehensive index of distribution network area division, comprising,

[0053] The parameter determination module is used to determine the comprehensive index for distribution network area division and set the relevant parameters for genetic algorithm and Pareto optimization.

[0054] The distribution network chromosome encoding module is used to encode the chromosomes of the distribution network based on the adjacency matrix and to perform crossover and mutation inheritance operations on the chromosomes of the distribution network.

[0055] The Pareto and genetic algorithm optimization module is used to establish a Pareto optimization objective function based on the comprehensive index of the distribution network area division, obtain the Pareto optimization solution set through Pareto dominance relationship and non-dominated solution, and obtain the Pareto frontier by optimizing the objective function through crossover and mutation of genetic algorithm.

[0056] The results output module is used to normalize the Pareto front, select the Pareto optimal solution with the closest values ​​for each index based on the normalization result, and output the result.

[0057] Thirdly, the present invention provides a computing device, comprising:

[0058] Memory and processor;

[0059] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the improved Pareto method for optimizing the comprehensive index of the distribution network area division.

[0060] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the improved Pareto method for optimizing the comprehensive index of distribution network area division.

[0061] Compared with the prior art, the beneficial effects of the present invention are as follows: By improving the Pareto method, the present invention overcomes the shortcomings of the existing comprehensive index optimization process for distribution network area division, and can objectively obtain the optimal or near-optimal comprehensive index, avoiding the problem of index clustering around a single index, thereby improving the balance and scientific nature of the zoning results, which is conducive to the optimized operation of the distribution network, improves decision-making efficiency and accuracy, and has important practical application value and promotion prospects. Attached Figure Description

[0062] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0063] Figure 1 This is a schematic diagram of the overall process of the improved Pareto method for optimizing the comprehensive index of distribution network area division according to an embodiment of the present invention.

[0064] Figure 2 This is a schematic diagram of a 151-node (busbar) example of the improved Pareto method for optimizing the comprehensive index of distribution network area division according to an embodiment of the present invention. Detailed Implementation

[0065] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0066] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0067] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0068] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0069] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the system or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0070] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0071] Example 1

[0072] Reference Figure 1 As an embodiment of the present invention, an improved Pareto method for optimizing the comprehensive index of distribution network area division is provided, comprising:

[0073] S100: Determine the comprehensive index for distribution network area division, and set the relevant parameters for genetic algorithm and Pareto optimization;

[0074] Furthermore, the comprehensive indicators for the division of distribution network areas include net load fluctuation indicators, modularity indicators, reactive power balance indicators, and active power balance indicators.

[0075] Set the Pareto iteration number it = 1, and the maximum iteration number is it. max The critical threshold ε is initialized to a very large value;

[0076] Furthermore, the net load volatility index is the sum of the absolute values ​​of net load volatility for each distribution network area, calculated using the following formula:

[0077]

[0078] ΔP adj,z =ΔP G,z +ΔP BESS,z +ΔP MG,z

[0079] Where, n z For the number of regions, This is a net load fluctuation indicator. Let ΔP be the net load fluctuation index for distribution network area z. L,z Net load is the sum of fluctuations in renewable energy output and load, ΔP. adj,z ΔP represents the regulated power of dispatchable distributed voltage within the distribution network area z. BESS,z ΔP represents the regulation power of the energy storage batteries within the distribution network area z. MG,z ΔP represents the adjustable power of the microgrid within the distribution network area z. G,z The adjustable power of dispatchable distributed power sources within the distribution network area z;

[0080] Modularity index, which measures the degree of electrical connection between nodes (busbars) within a distribution network area, is calculated using the following formula:

[0081]

[0082] Where ρ is the modularity index, m is the coefficient, and e ij k represents the weight of the edge between nodes (buses) i and j. i =∑ i e ij The sum of the weights of the edges connected to the bus node (bus) i;

[0083] Preferred weight e ij Expressed as a formula:

[0084] e ij =1-L ij / max(L)

[0085]

[0086]

[0087]

[0088] Among them, L ij The electrical distance between nodes (busbars) i and j, taking into account the influence of other nodes (busbars), where n is the number of nodes (busbars) and d is the distance between them. ik and d jk S represents the ratio of the voltage change at node (bus) k to the voltage changes at nodes (buses) i and j, respectively, when the active and reactive power changes. VP and S VQ These are the active and reactive voltage sensitivity matrices, respectively.

[0089] The formula for calculating the reactive power balance index is:

[0090]

[0091]

[0092] in, Q is an indicator of reactive power balance. z Q represents the reactive power balance of distribution network area z. sup Q represents the maximum reactive power supply within the distribution network area. ned For reactive power demand within the distribution network area;

[0093] Preferred, Q ned This represents the reactive power demand within the region, including not only normal reactive power demand but also the minimum reactive power Q required to regulate overvoltage at node (bus) i. V The calculation formula is:

[0094]

[0095] Among them, C z It is the region z, ΔV i S is the voltage change at node (bus) i. VQ,ii It is the reactive voltage sensitivity of node (bus) i itself;

[0096] The formula for calculating the active power balance index is:

[0097]

[0098]

[0099] in, As an indicator of active power balance, The desired active power of a dispatchable distributed power source. For the desired power of the energy storage battery, For the expected active power of the microgrid, Expected active power for renewable energy sources The expected active power of the load;

[0100] It should be noted that this scheme ensures the accuracy and flexibility of the optimization process by setting the number of Pareto iterations and the maximum number of iterations, as well as initializing the critical threshold. At the same time, by combining genetic algorithms and Pareto optimization, it can find the optimal or near-optimal comprehensive index among multiple indicators, avoiding the index clustering problem that may occur in traditional methods, and making the partitioning results more scientific and reasonable.

[0101] It should also be noted that this scheme provides a comprehensive evaluation system for the division of distribution network areas by comprehensively considering net load fluctuation index, modularity index, reactive power balance index, and active power balance index. This comprehensive evaluation method can ensure that the zoning results not only meet the requirements of electrical connection tightness, but also guarantee the active and reactive power balance of the system, while reducing net load fluctuation, thereby improving the stability and efficiency of the entire distribution network.

[0102] S102: Chromosome encoding for the distribution network is performed based on the adjacency matrix, and crossover and mutation inheritance operations are performed on the chromosomes of the distribution network.

[0103] Furthermore, a chromosome represents an adjacency matrix of the distribution network. If the i-th and j-th columns of the adjacency matrix are 0, it means that nodes (buses) i and j are not connected. If they are 1, it means that nodes (buses) i and j are connected.

[0104] Genetic algorithms are used to perform chromosome crossover mutation on the distribution network, and the maximum value of the comprehensive index for each distribution network area is calculated.

[0105] Preferably, the maximum value of each indicator is represented as:

[0106] Preferably, the number of individuals in the genetic algorithm population is set to 50, the maximum number of iterations is 500, the crossover probability is [0.3-0.6], and the mutation probability is [0.1-0.4].

[0107] It should be noted that this technical solution combines adjacency matrix encoding and genetic algorithm, and achieves efficient optimization of distribution network partitioning by intuitively reflecting the distribution network topology, efficiently searching for the optimal partitioning scheme, optimizing multiple indicators, flexibly setting parameters, and providing strong support for decision-makers. This improves the performance and stability of the distribution network and provides decision-makers with a scientific and reasonable partitioning scheme.

[0108] S104: According to

[0109] A Pareto optimization objective function is established based on the comprehensive index of distribution network area division. The Pareto optimization solution set is obtained through Pareto dominance relationship and non-dominated solution. The Pareto front is obtained by optimizing the objective function through crossover and mutation of genetic algorithm.

[0110] The Pareto objective function is expressed as:

[0111] maxF = (f1, f2, f3, f4)

[0112] in, The net load fluctuation index is f2 = ρ, where ρ is the modularity index. As an indicator of reactive power balance, The active power balance index;

[0113] Preferably, for any two feasible solutions x i and x j If x i Any sub-target is greater than x j The sub-goals, namely:

[0114]

[0115] Where, x i For a feasible solution, x j This is a feasible solution;

[0116] Preferably, there exists a value m0, m0∈(1,2,3,4) that satisfies the dominance formula:

[0117] f m0 (x j ) < f m0 (x i )

[0118] Where, x i For a feasible solution, x j This is a feasible solution;

[0119] The preferred solution is x. i dominatex j If there is no dominant x in the solution vector i A feasible solution is called x. i It is a nondominated solution of the dominant form, and all nondominated solutions constitute a Pareto optimization of the dominant form.

[0120] The objective function is calculated using a genetic algorithm with crossover and mutation to obtain the Pareto optimization in the i-th iteration, expressed as:

[0121]

[0122] Where i = 1, 2, ..., n, and n is the number of solutions. Let ρ be the net load fluctuation index value of the i-th solution in the it-th iteration. it,i Let be the modularity index value of the i-th solution in the it-th iteration. Let be the reactive power balance index value of the i-th solution in the it-th iteration. Let be the active power balance index value of the i-th solution in the it-th iteration;

[0123] Preferably, the number of individuals in the genetic algorithm population is set to 50, the maximum number of iterations is 500, the crossover probability is [0.3-0.6], and the mutation probability is [0.1-0.4].

[0124] It should be noted that the Pareto optimization objective function comprehensively considers multiple indicators such as net load volatility, modularity, reactive power balance and active power balance, ensuring that the performance and stability of the distribution network can be fully considered during the optimization process. Through the Pareto dominance relationship and the concept of non-dominated solutions, a set of solutions that cannot be further improved without sacrificing any indicator can be found, namely the Pareto front.

[0125] It should also be noted that genetic algorithms, by simulating the evolutionary process in nature, utilize operations such as selection, crossover, and mutation to efficiently search for Pareto fronts in the solution space. This method is not only fast in its search speed but also capable of finding the global optimum, avoiding the problem of traditional optimization methods getting trapped in local optima. The Pareto fronts obtained through genetic algorithms and Pareto optimization provide decision-makers with multiple feasible solutions that perform well under different indicators, thereby improving the scientific and rational nature of decision-making.

[0126] S106: Normalize the Pareto front, select the Pareto optimal solution with the closest values ​​of each index based on the normalization result, and output the result.

[0127] Furthermore, the Pareto front normalized representation is:

[0128]

[0129] Where i = 1, 2, ..., n, and n is the number of solutions. Let ρ be the net load fluctuation index value of the i-th solution in the it-th iteration. it,i Let be the modularity index value of the i-th solution in the it-th iteration. Let be the reactive power balance index value of the i-th solution in the it-th iteration. Let be the active power balance index value of the i-th solution in the it-th iteration;

[0130] The absolute value of the differences between the Pareto optimization metrics after the it-th iteration is used to obtain the Pareto front closest to each metric after Pareto normalization. The calculation formula is as follows:

[0131]

[0132] Where i = 1, 2, ..., n, and n is the number of solutions. Let ρ be the net load fluctuation index value of the i-th solution in the it-th iteration. it,i Let be the modularity index value of the i-th solution in the it-th iteration. Let be the reactive power balance index value of the i-th solution in the it-th iteration. Let be the active power balance index value of the i-th solution in the it-th iteration;

[0133] Furthermore, if the normalized Pareto front index is lower than the critical threshold ε, the critical threshold is updated and the Pareto optimization result of the current iteration is retained. If it has not reached the maximum number of iterations, the normalized Pareto front of each index is calculated again. Otherwise, the Pareto front corresponding to the critical threshold ε is output.

[0134] If the normalized Pareto front index is higher than the critical threshold ε and it has not reached the maximum number of iterations, continue to calculate the normalized Pareto front for each index; otherwise, output the Pareto front corresponding to the critical threshold ε.

[0135] It should be noted that by normalizing the Pareto front, the differences in dimensions and values ​​between different optimization indices are eliminated, allowing them to be compared and weighed within the same framework. Combined with the strategy of selecting the Pareto optimal solution with the closest values ​​for each index, this ensures that the selected solution performs well across multiple indices. The critical threshold update mechanism further enhances the stability and convergence of the algorithm, causing the solution to gradually approach the true Pareto front. This not only improves decision-making efficiency and accuracy but also provides decision-makers with clear and reliable decision support, ensuring that the solution gradually approaches the optimal solution as the number of iterations increases. Simultaneously, by combining the iterative process of genetic algorithms and Pareto optimization, the critical threshold is continuously updated and the optimal solution is selected, gradually approaching the true Pareto front, ensuring the convergence of the algorithm. That is, as the number of iterations increases, the obtained solution gradually approaches the optimal solution.

[0136] The above is an illustrative scheme of an improved Pareto method for optimizing the comprehensive index of distribution network area division according to this embodiment. It should be noted that the technical solution of this improved Pareto system for optimizing the comprehensive index of distribution network area division belongs to the same concept as the technical solution of the improved Pareto method for optimizing the comprehensive index of distribution network area division described above. Details not described in detail in the technical solution of the improved Pareto system for optimizing the comprehensive index of distribution network area division in this embodiment can be found in the description of the improved Pareto method for optimizing the comprehensive index of distribution network area division described above.

[0137] The improved Pareto system for optimizing the comprehensive index of distribution network area division in this embodiment includes:

[0138] The parameter determination module is used to determine the comprehensive index for distribution network area division and set the relevant parameters for genetic algorithm and Pareto optimization.

[0139] The distribution network chromosome encoding module is used to encode the chromosomes of the distribution network based on the adjacency matrix and to perform crossover and mutation inheritance operations on the chromosomes of the distribution network.

[0140] The Pareto and genetic algorithm optimization module is used to establish a Pareto optimization objective function based on the comprehensive index of distribution network area division, obtain the Pareto optimization solution set through Pareto dominance relationship and non-dominated solution, and obtain the Pareto frontier by optimizing the objective function through crossover and mutation of genetic algorithm.

[0141] The results output module is used to normalize the Pareto front, select the Pareto optimal solution with the closest values ​​for each index based on the normalization result, and output the result.

[0142] This embodiment also provides a computing device applicable to the improved Pareto case of optimizing the comprehensive index of distribution network area division, including:

[0143] The system includes a memory and a processor. The memory stores computer-executable instructions, and the processor executes these instructions to implement the improved Pareto method for optimizing the comprehensive index of distribution network area division, as proposed in the above embodiments.

[0144] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the improved Pareto method for optimizing the comprehensive index of distribution network area division as proposed in the above embodiments.

[0145] The storage medium proposed in this embodiment and the improved Pareto method for optimizing the comprehensive index of distribution network area division proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0146] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0147] Example 2

[0148] Referring to Tables 1-5, an improved Pareto method for optimizing the comprehensive index of distribution network area division is provided as an embodiment of the present invention. Real experimental results are provided to verify its beneficial effects.

[0149] This embodiment uses a 151-node (busbar) distribution network in a certain area of ​​Yunnan Province as an example to verify the effectiveness and accuracy of the present invention.

[0150] 151 nodes (busbars) Figure 2 As shown, 1 is the root node (bus), the voltage is 10kV, the voltage limit is [9.3, 10.7]kV, the active load is 5.738MW, the reactive load is 2.787Mvar, and it is assumed that the active load fluctuation range is ±2% of the current value.

[0151] To verify the superiority of the partitioning method of this invention, the power generation of a single photovoltaic power generation is 150kW, with a fluctuation range of ±10% of the expected value and a fluctuation power of [-15, 15]kW; the power generation of each distributed dispatchable power source is 100kW, and the adjustable power is [-30, 30]kW; assuming that energy storage is in the power generation state, the adjustable power is [-20, 20]kW; the power exchange limit between the microgrid and the distribution network depends on branch 32-33, with a maximum exchange power of 500kW.

[0152] The optimization results for testing a single metric are shown in Table 1.

[0153] Table 1 Optimization results for a single indicator as the objective

[0154] index numerical values Net load variability 0.845 Modularity 0.952 reactive power balance 0.941 Active power balance 0.955

[0155] The Pareto fronts obtained through genetic algorithms and Pareto analysis yielded two non-dominated solutions, as shown in Table 2.

[0156] Table 2 Non-dominated solutions obtained from improved Pareto optimization.

[0157] index Non-inferior solution 1 Non-inferior solution 2 Net load variability 0.631 0.607 Modularity 0.722 0.735 reactive power balance 0.920 0.915 Active power balance 0.835 0.824

[0158] The partitions calculated based on each indicator are shown in Table 3.

[0159] Table 3 lists the nodes (buses) included in the partitions of non-dominated solutions 1 and 2.

[0160] Partition Number Non-inferior solution 1 Non-inferior solution 2 1 1-17 1-17 2 18-31 18-31 3 45-56 45-60 4 32-44 32-44 5 57-74,92,93 61-74, 6 75-91 75-91,92,93 7 94-102,121-128 94-102,121-128,129,130 8 103-120 103-120 9 129-151 131-151

[0161] To verify the accuracy of this invention, as a comparison, the indicators obtained under conditions where weighting coefficients were artificially assigned are shown in Table 4.

[0162] Table 4 specifies the comprehensive index with weighted coefficients.

[0163]

[0164] Meanwhile, for comparison, the Pareto front obtained by directly applying Pareto optimization, i.e., the non-dominated solution, is shown in Table 5.

[0165] Table 5 Non-dominated solutions obtained from Pareto optimization

[0166] index Non-inferior solution 1 Non-inferior solution 2 Net load variability 0.629 0.618 Modularity 0.718 0.935 reactive power balance 0.912 0.914 Active power balance 0.823 0.826

[0167] By comparing the optimization results with those obtained through Pareto optimization, the effectiveness and accuracy of the multi-objective optimization method used in this invention can be clearly seen.

[0168] First, as shown in Table 1, the optimal values ​​achievable by each indicator when optimized individually are: net load volatility 0.845, modularity 0.952, reactive power balance 0.941, and active power balance 0.955. However, these single-objective optimization results cannot guarantee that they will be optimal simultaneously for all indicators.

[0169] Table 5 shows the non-dominated solutions obtained through Pareto optimization, which are solutions that achieve relative optimality on multiple metrics simultaneously. Among them, non-dominated solution 2 is very close to the single objective extreme value of 0.952 in terms of modularity, and also performs well on other metrics. This shows that Pareto optimization can find a balance among multiple metrics, so that the solution achieves overall optimality.

[0170] The line losses calculated by the method of this invention after partitioning are 145.23kW and 147.32kW, which are significantly less than the line losses calculated by manually specifying weight coefficients (150.77kW, 164.32kW, and 149.45kW), and also less than the line losses calculated by Pareto optimization (148.69kW and 157.32kW). By comparing the line loss values ​​calculated by different methods, the effectiveness and accuracy of the method of this invention can be clearly seen. After using the method of this invention to partition the distribution network area, the calculated line loss values ​​are 145.23kW and 147.32kW, which is significantly lower than the line loss values ​​obtained by other methods.

[0171] Therefore, the Pareto optimization-based multi-objective optimization method adopted in this invention can not only achieve good optimization results on a single index, but also find a balance among multiple indexes to ensure that the solution is optimal overall. At the same time, the application of the method of this invention in the division and optimization of distribution network areas, by comprehensively considering multiple indexes and finding the optimal solution, significantly reduces the calculated line loss value. This result proves the effectiveness and accuracy of the method of this invention, and can provide a more scientific and reliable basis for the planning and operation of distribution networks.

[0172] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An improved Pareto method for distribution network area division comprehensive index optimization, characterized in that, The method comprises the following steps: determining a comprehensive index of distribution network area division, setting parameters of genetic algorithm and Pareto optimization; coding chromosomes of the distribution network based on an adjacency matrix, and performing crossover and mutation genetic operations on the chromosomes of the distribution network; establishing a Pareto optimization objective function according to the comprehensive index of distribution network area division, obtaining a Pareto optimization solution set through a Pareto dominance relationship and a non-inferior solution, and obtaining a Pareto frontier through crossover and mutation optimization of the objective function by genetic algorithm; performing normalization processing on the Pareto frontier, and selecting a Pareto optimal solution closest to each index value according to a normalization result and outputting the result.

2. The improved Pareto method for optimization of network deployment area division comprehensive index according to claim 1, wherein, The method for determining a comprehensive index of distribution network area division, setting parameters of genetic algorithm and Pareto optimization comprises the following steps: The comprehensive index of distribution network area division comprises a net load fluctuation index, a modularity index, a reactive power balance index and an active power balance index. Set the number of iterations of the Pareto it = 1, the maximum number of iterations it max , the critical threshold value of ε is a very large number of initial.

3. The improved Pareto method for network area division comprehensive index optimization according to claim 1 or 2, characterized in that, The method further comprises the following steps: The net load fluctuation index is the sum of absolute values of net load fluctuations of each distribution network area, and a calculation formula is as follows: ΔP adj,z = ΔP G,z + ΔP BESS,z + ΔP MG,z wherein n z is the number of areas, is the net load fluctuation index, is the net load fluctuation index of the distribution network area z, ΔP L,z is the net load, being the sum of the fluctuation values of the renewable energy output and the load, ΔP adj,z is the adjustable power of the dispatchable distributed voltage within the distribution network area z, ΔP BESS,z is the adjustable power of the energy storage battery within the distribution network area z, ΔP MG,z is the adjustable power of the microgrid within the distribution network area z, ΔP G,z is the adjustable power of the dispatchable distributed power source within the distribution network area z; The modularity index is the tightness of electrical connection of nodes (buses) in the distribution network area, and a calculation formula is as follows: where p is the modularity index, m is a coefficient, e ij is the weight of the edge between busbar nodes (busbars) i and j, k i =∑ i e ij is the sum of the weights of the edges connected to busbar node (busbar) i. A calculation formula of the reactive power balance index is as follows: wherein, is the reactive power balance index, Q z is the reactive power balance of the distribution network area z, Q sup is the maximum reactive power supply in the distribution network area, Q ned is the reactive power demand in the distribution network area; A calculation formula of the active power balance index is as follows: wherein, is an active balance index, is a dispatchable distributed generator expected active power, is an energy storage battery expected power, is a microgrid expected active power, is a renewable energy expected active power, is a load expected active power.

4. The improved Pareto method for network area division integrated index optimization of claim 3, wherein, The method for coding chromosomes of the distribution network based on an adjacency matrix, and performing crossover and mutation genetic operations on the chromosomes of the distribution network comprises the following steps: One chromosome represents one adjacency matrix of the distribution network, and if the i-th and j-th columns of the adjacency matrix are 0, it indicates that nodes (buses) i and j are not connected, and if the i-th and j-th columns of the adjacency matrix are 1, it indicates that nodes (buses) i and j are connected. The method for performing crossover and mutation genetic operations on the chromosomes of the distribution network by genetic algorithm comprises the following steps:

5. The improved Pareto method for optimization of network deployment area division comprehensive index according to claim 4, characterized in that, The method for establishing a Pareto optimization objective function according to the comprehensive index of distribution network area division, obtaining a Pareto optimization solution set through a Pareto dominance relationship and a non-inferior solution, and obtaining a Pareto frontier through crossover and mutation optimization of the objective function by genetic algorithm comprises the following steps: The Pareto objective function is expressed as follows: max F = (f1, f2, f3, f4) wherein, is a net load fluctuation index, f2= p, p is a modularity index, is a reactive power balance index, is an active power balance index; The method for performing crossover and mutation optimization of the objective function by genetic algorithm to obtain the ith iteration of Pareto optimization is expressed as follows: wherein i = 1, 2,..., n, n is the number of solutions, is the net load fluctuation index value of the ith solution in the itth iteration, ρ it,i is the modularity index value of the ith solution in the itth iteration, is the reactive power balance degree index value of the ith solution in the itth iteration, is the active power balance degree index value of the ith solution in the itth iteration.

6. The improved Pareto method for optimization of network deployment area division comprehensive index according to claim 5, characterized in that, The method for performing normalization processing on the Pareto frontier comprises the following steps: The normalization of the Pareto frontier is expressed as follows: Wherein, i = 1, 2,..., n, n is the number of solutions, is the net load fluctuation index value of the ith solution in the ith iteration, ρ it,i is the modularity index value of the ith solution in the ith iteration, is the reactive power balance degree index value of the ith solution in the ith iteration, is the active power balance degree index value of the ith solution in the ith iteration; The method for obtaining the Pareto frontier closest to each index after normalization of the Pareto optimization after the ith iteration comprises the following steps: wherein i = 1, 2,..., n, n is the number of solutions, is the net load fluctuation index value of the ith solution in the itth iteration, ρ it,i is the modularity index value of the ith solution in the itth iteration, is the reactive power balance degree index value of the ith solution in the itth iteration, is the active power balance degree index value of the ith solution in the itth iteration.

7. The improved Pareto method for network area division integrated index optimization of claim 6, wherein, The method for selecting a Pareto optimal solution closest to each index value according to a normalization result and outputting the result comprises the following steps: If the index of the normalized Pareto frontier is lower than a critical threshold ε, the critical threshold is updated and the Pareto optimization result of the current iteration is retained, if the ith iteration does not reach a maximum iteration number, the normalized Pareto frontier of each index is continuously calculated, otherwise, the Pareto frontier corresponding to the critical threshold ε is outputted; If the index of the normalized Pareto frontier is higher than the critical threshold ε, and the ith iteration does not reach the maximum iteration number, the normalized Pareto frontier of each index is continuously calculated, otherwise, the Pareto frontier corresponding to the critical threshold ε is outputted.

8. An improved Pareto system for distribution network area division integrated index optimization, characterized in that, The method comprises the following steps: The parameter determination module is configured to determine a comprehensive index of distribution network area division, and set parameters of genetic algorithm and Pareto optimization. The network distribution chromosome coding module is configured to code chromosomes of network distribution based on an adjacency matrix and to perform crossover and mutation genetic operations on chromosomes of network distribution. The Pareto and genetic algorithm optimization module is configured to establish a Pareto optimization objective function according to the network distribution area division comprehensive index, to obtain a Pareto optimization solution set through a Pareto dominance relationship and a non-inferior solution, and to obtain a Pareto frontier through crossover and mutation of the genetic algorithm optimization objective function. The result output module is configured to normalize the Pareto frontier, to select a Pareto optimal solution with the closest index value according to a normalized result, and to output the result. 9.An electronic device, comprising: a memory and a processor; the memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions, and the computer executable instructions, when executed by the processor, implement the steps of the improved Pareto method for optimizing the network distribution area division comprehensive index according to any one of claims 1 to 7. 10.A computer readable storage medium storing computer executable instructions, and the computer executable instructions, when executed by a processor, implement the steps of the improved Pareto method for optimizing the network distribution area division comprehensive index according to any one of claims 1 to 7.

Images