A distribution network framework optimization method and system

Abstract

The present invention discloses a distribution network grid optimization method and system, which belongs to the field of distribution network planning, wherein the method includes: obtaining initial grid characteristic parameters; calling a target distribution database; randomly obtaining a first distribution record; analyzing the first distribution network voltage record and the first distribution network power supply record based on a predetermined evaluation rule to obtain an initial grid index; iteratively optimizing the initial grid characteristic parameters based on a predetermined optimization rule to obtain the optimal grid characteristic parameters; and transforming and optimizing the target grid according to the optimal grid characteristic parameters. The present application solves the technical problem that the prior art lacks grid optimization means, making it difficult to systematically obtain the optimal grid solution, resulting in low stability and power supply reliability of the distribution network grid, and achieves the technical effect of realizing automated intelligent design and iterative optimization of the distribution network grid by establishing evaluation rules and optimization rules, thereby improving the stability and power supply reliability of the distribution network.

Description

Technical Field

[0001] The present invention relates to the field of distribution network planning, and in particular to a distribution network framework optimization method and system. Background Art

[0002] With the rapid development of social economy, the power load is growing continuously, which puts forward higher requirements for the stable power supply of power system. As the terminal part of power system, the operation quality of distribution network is directly related to the power use of users. The existing distribution network design mainly relies on the experience of designers. It is difficult to obtain different grid design schemes comprehensively and make systematic judgments. It is impossible to effectively obtain the optimal design parameters, resulting in low stability and power supply reliability of distribution network. Summary of the invention

[0003] The present application provides a distribution network framework optimization method and system, aiming to solve the technical problems in the prior art that there is a lack of grid optimization means, it is difficult to systematically obtain the optimal grid solution, and the stability of the distribution network framework and the power supply reliability are low.

[0004] In view of the above problems, the present application provides a distribution network framework optimization method and system.

[0005] In a first aspect disclosed in the present application, a distribution network grid optimization method is provided, the method comprising: obtaining initial grid characteristic parameters, the initial grid characteristic parameters are parameters obtained by collecting characteristics of a target grid of a target distribution network based on predetermined characteristic indicators; retrieving a target distribution database, the target distribution database refers to a historical record database of the target distribution network, and the target distribution database comprises multiple groups of distribution records; randomly obtaining a first distribution record from the multiple groups of distribution records, the first distribution record comprising a first distribution network voltage record and a first distribution network power supply record; analyzing the first distribution network voltage record and the first distribution network power supply record based on predetermined evaluation rules to obtain an initial grid index; iteratively optimizing the initial grid characteristic parameters based on predetermined optimization rules to obtain optimal grid characteristic parameters, the optimal grid characteristic parameters having an optimal grid index identifier, and the optimal grid index being greater than the initial grid index; and transforming and optimizing the target grid according to the optimal grid characteristic parameters.

[0006] Another aspect disclosed in the present application provides a distribution network grid optimization system, the system comprising: a characteristic parameter acquisition module, used to acquire initial grid characteristic parameters, the initial grid characteristic parameters are parameters obtained by collecting characteristics of a target grid of a target distribution network based on predetermined characteristic indicators; a database retrieval module, used to retrieve a target distribution database, the target distribution database refers to a historical record database of the target distribution network, and the target distribution database includes multiple groups of distribution records; a distribution record random module, used to randomly acquire a first distribution record from the multiple groups of distribution records, the first distribution record including a first distribution network voltage record and a first distribution network power supply record; a power supply record analysis module, used to analyze the first distribution network voltage record and the first distribution network power supply record based on a predetermined evaluation rule to obtain an initial grid index; an optimal parameter acquisition module, used to iteratively optimize the initial grid characteristic parameters based on a predetermined optimization rule to obtain an optimal grid characteristic parameter, the optimal grid characteristic parameter having an optimal grid index identifier, and the optimal grid index is greater than the initial grid index; a grid transformation optimization module, used to transform and optimize the target grid according to the optimal grid characteristic parameters.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The method adopts the method of obtaining initial grid characteristic parameters to lay a foundation for subsequent optimization; calling the target distribution database to provide a basis for extracting the actual distribution situation and ensure the comprehensive accuracy of the data based on which the subsequent optimization is based; randomly obtaining the first distribution record to provide a data source for subsequent analysis; analyzing the first record based on a predetermined evaluation rule to obtain an initial grid index to systematically and comprehensively judge different grid parameters to obtain an evaluation index result; iteratively searching for the optimal parameters based on a predetermined optimization rule to obtain the optimal grid characteristic parameters; using the optimal grid characteristic parameters to guide the transformation and optimization of the actual distribution network grid, solves the technical problems in the prior art that there is a lack of grid optimization means, it is difficult to systematically obtain the optimal grid solution, and the stability of the distribution network grid and the low power supply reliability are achieved, and the technical effect of realizing the automated intelligent design and iterative optimization of the distribution network grid and improving the stability and power supply reliability of the distribution network is achieved by establishing evaluation rules and optimization rules.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1A schematic diagram of a flow chart of a method for optimizing a power distribution network is provided for an embodiment of the present application;

[0011] Figure 2 A schematic diagram of a flow chart for obtaining optimal grid characteristic parameters in a distribution network grid optimization method is provided for an embodiment of the present application;

[0012] Figure 3 A structural schematic diagram of a distribution network optimization system is provided for an embodiment of the present application.

[0013] Explanation of the reference numerals: characteristic parameter acquisition module 11, database retrieval module 12, power distribution record random module 13, power supply record analysis module 14, optimal parameter acquisition module 15, grid transformation optimization module 16. DETAILED DESCRIPTION

[0014] The overall idea of ​​the technical solution provided by this application is as follows:

[0015] The embodiment of the present application provides a distribution network grid optimization method and system, which optimizes the grid characteristic parameters of the distribution network grid by acquiring initial grid characteristic parameters, calling a target distribution database, selecting distribution records, and applying predetermined evaluation and optimization rules, so as to obtain the optimal grid characteristic parameters and optimize the actually laid grid parameters, thereby improving the stability of the distribution network and the reliability of power supply.

[0016] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0017] Embodiment 1

[0018] like Figure 1 As shown, an embodiment of the present application provides a distribution network framework optimization method, the method comprising:

[0019] Acquire initial grid characteristic parameters, wherein the initial grid characteristic parameters are parameters obtained by collecting characteristics of a target grid of a target distribution network based on predetermined characteristic indicators;

[0020] Furthermore, the predetermined characteristic indicators include at least grid size, grid span, grid height, grid depth and cross-sectional shape.

[0021] In the embodiment of the present application, the predetermined characteristic index is a set of characteristic parameters used to describe the grid of the distribution network. In order to fully reflect the structure and performance of the grid of the distribution network, the predetermined characteristic index includes at least the grid size, grid span, grid height, grid depth and cross-sectional shape. Among them, the grid size is the distance between adjacent grids in the distribution network, reflecting the layout density of the grid; the grid span is the horizontal distance between two adjacent electric poles, reflecting the extension range of the grid; the grid height is the vertical height of the electric pole, which determines the safe passage height of the electric wire; the grid depth is the depth of the electric pole buried in the ground, which is related to the stability of the electric pole; the cross-sectional shape is the cross-sectional shape of the electric pole, such as round, square, etc., which affects its bearing strength.

[0022] First, determine the target grid of the target distribution network, that is, the distribution network grid that needs to be optimized and transformed, and select multiple measurement points on the target grid, such as points at the grid input end, output end and midpoint. Then, based on each measurement point, collect corresponding data according to the definition of the predetermined characteristic index, such as using a tape measure to measure the grid size and grid span; using a height gauge to measure the grid height; using an excavator to check the buried depth of the pole and measure the cross-sectional shape at the same time. Subsequently, for each predetermined characteristic index, calculate the data mean of all measurement points as the characteristic parameter corresponding to the characteristic index. Combine the characteristic parameters corresponding to all predetermined characteristic indicators to form the initial grid characteristic parameters.

[0023] Retrieving a target power distribution database, wherein the target power distribution database refers to a historical record database of the target power distribution network, and the target power distribution database includes multiple groups of power distribution records;

[0024] In the embodiment of the present application, the target power distribution database is used to store the historical operation and distribution data of the target distribution network, providing a reference for the grid optimization process. The target power distribution database includes multiple groups of distribution records, which are various types of distribution and operation logs generated by the target distribution network during operation, such as power supply records, accident records, voltage data, etc. By calling the target power distribution database, the operation of the target distribution network is analyzed, the performance of the grid is evaluated, and a basis is provided for grid optimization.

[0025] Randomly acquiring a first power distribution record from the plurality of groups of power distribution records, wherein the first power distribution record includes a first distribution network voltage record and a first distribution network power supply record;

[0026] In an embodiment of the present application, after retrieving the target power distribution database, the total number of multiple groups of power distribution records is confirmed, that is, the number of power distribution record groups stored in the target power distribution database. Then, a random number seed is set to generate a random integer between 1 and the number of power distribution record groups. According to the obtained random integer, the power distribution record corresponding to the random integer is determined from the multiple groups of power distribution records as the first power distribution record. The first power distribution record includes a first distribution network voltage record and a first distribution network power supply record. Among them, the first distribution network voltage record is a record of voltage parameters in the first distribution record, including voltage timing data of multiple nodes of the target distribution network; the first distribution network power supply record is a record of power supply and line loss conditions in the first distribution record, including total power supply and power loss data.

[0027] By randomly extracting distribution records, we avoid the bias caused by human selection and ensure that the first distribution record obtained meets the objective random requirements. It serves as the optimization starting point for iterative optimization of initial grid characteristic parameters, providing data support for subsequent grid optimization.

[0028] Analyze the first distribution network voltage record and the first distribution network power supply record based on a predetermined evaluation rule to obtain an initial grid index;

[0029] Furthermore, the embodiment of the present application also includes:

[0030] The predetermined evaluation rules include predetermined voltage evaluation rules and predetermined power supply evaluation rules.

[0031] The first distribution network voltage record includes a plurality of voltage time series at multiple points in the target distribution network;

[0032] Matching a first voltage timing of a first point in the plurality of voltage timings, wherein the first point is any one of the plurality of points;

[0033] Performing multi-feature collection on a first voltage curve generated based on the first voltage time series to obtain a first voltage curve feature set;

[0034] Analyzing the first voltage curve feature set based on the predetermined voltage evaluation rule to obtain a first voltage index;

[0035] The first power distribution network power supply record includes a first total power supply and a first power loss;

[0036] Based on the predetermined power supply evaluation rule, a normalized result of the ratio of the first power loss to the first total power supply is obtained and recorded as a first power supply index;

[0037] An average value of a first grid index obtained by weighting the first voltage index and the first power supply index is taken as the initial grid index.

[0038] Furthermore, the first voltage curve feature set includes a first voltage qualification rate, a first voltage flicker frequency and a first voltage fluctuation amplitude.

[0039] In a preferred embodiment, the predetermined evaluation rules are rules for analyzing power distribution records and evaluating grid performance, including predetermined voltage evaluation rules and predetermined power supply evaluation rules. Among them, the predetermined voltage evaluation rules are specific rules for judging the voltage quality of the distribution network formulated in combination with the operating characteristics of the target distribution network. The main indicators include voltage qualification rate, voltage flicker frequency, voltage fluctuation amplitude, etc., for example, including: qualification rate not less than 95%, flicker frequency not more than 20 times / day, fluctuation amplitude not exceeding ±5%, etc.; the predetermined power supply evaluation rules are actually based on the target distribution network operation history data, determine the line loss evaluation rules, and judge the corresponding relationship between power loss and power supply to evaluate the grid power supply efficiency.

[0040] The first distribution network voltage record includes multiple voltage time series of multiple points in the target distribution network. Randomly select any point from the multiple points of the target distribution network as the first point, match the corresponding voltage time series in the voltage record of the first distribution network, and find the voltage time series corresponding to the first point, which is the first voltage time series. Then, according to the obtained first voltage time series, set the time as the X-axis and the voltage value as the Y-axis to draw the first voltage curve. Subsequently, calculate the number of qualified sections of the first voltage curve within the specified voltage qualified range, and calculate the ratio of the number of qualified sections to the total number of sections of the first voltage curve to obtain the first voltage qualified rate; at the same time, count the number of jumps in a certain time interval of the first voltage curve, calculate the ratio of the number of jumps to the total duration, and obtain the first voltage flicker frequency; at the same time, search for the maximum voltage value and the minimum voltage value of the first voltage curve, and calculate the difference between the maximum voltage value and the minimum voltage value to obtain the first voltage fluctuation amplitude. After that, the first voltage qualified rate, the first voltage flicker frequency, and the first voltage fluctuation amplitude are summarized to obtain the first voltage curve feature set. Then, the first voltage curve feature set is substituted into the predetermined voltage evaluation rules for item-by-item inspection, and a comprehensive first voltage index result is obtained by using a multi-index weighted average method, providing a basis for judging voltage quality for grid optimization.

[0041] The first distribution network power supply record includes the first total power supply and the first power loss, which reflects the power supply and line loss of the target distribution network when the first distribution record is generated. Among them, the first total power supply is the total power supply of the target distribution network to the user end during the first distribution record cycle; the first power loss is the power loss incurred by the target distribution network during the power transmission and distribution process during the first distribution record cycle. Then, the ratio of the first power loss to the first total power supply is calculated to obtain the first line loss rate, and then normalized so that the result value is mapped to the range of 0-100, which is the first power supply index, reflecting the power supply efficiency and line loss level of the current grid.

[0042] At this point, the first voltage index obtained reflects the grid power supply voltage quality, and the first power supply index reflects the grid power supply efficiency. 1 Set weight ratios for the first voltage index and the first power supply index, such as 0.6 and 0.4, respectively. Then, by weighted averaging the first voltage index and the first power supply index, the initial grid index is obtained, which fully reflects the comprehensive performance of the grid and serves as the performance benchmark of the current grid for subsequent grid optimization.

[0043] Iteratively optimizing the initial grid characteristic parameters based on a predetermined optimization rule to obtain optimal grid characteristic parameters, wherein the optimal grid characteristic parameters have an identifier of an optimal grid index, and the optimal grid index is greater than the initial grid index;

[0044] Furthermore, this step includes:

[0045] Randomly extracting a first characteristic index from the predetermined characteristic indexes based on the predetermined optimization rule;

[0046] Obtaining a predetermined grid characteristic parameter according to a first characteristic parameter randomly obtained in combination with a first value range of the first characteristic index;

[0047] Calculate the similarity between a first grid characteristic parameter and a first characteristic of the predetermined grid characteristic parameter in a plurality of groups of grid characteristic parameters of a plurality of distribution networks in a distribution database, wherein the distribution database is a distribution network historical distribution database established based on big data crawling;

[0048] If the first feature similarity reaches a predetermined similarity threshold, a predetermined grid group is formed based on the first grid feature parameter and the predetermined grid feature parameter; if the first feature similarity does not reach the predetermined similarity threshold, an initial grid group is formed based on the first grid feature parameter and the initial grid feature parameter;

[0049] The predetermined grid group and the initial grid group are compared to obtain the optimal grid characteristic parameters.

[0050] Further, such as Figure 2 As shown, this step also includes:

[0051] The predetermined grid index sum of the predetermined grid group and the initial grid index sum of the initial grid group are calculated respectively;

[0052] If the predetermined grid index sum is greater than the initial grid index sum, the predetermined grid characteristic parameters are taken as candidate grid characteristic parameters; if the predetermined grid index sum is less than the initial grid index sum, the initial grid characteristic parameters are taken as candidate grid characteristic parameters;

[0053] The candidate grid feature parameters are iteratively optimized to obtain the optimal grid feature parameters.

[0054] Furthermore, this step also includes:

[0055] Acquire a first neighborhood of the candidate grid feature parameters, the first neighborhood including a plurality of groups of neighborhood grid feature parameters;

[0056] Analyzing the simulation data of the plurality of groups of neighborhood grid characteristic parameters based on the predetermined evaluation rules to obtain a plurality of neighborhood grid indexes;

[0057] Determine whether the neighborhood optimal grid index among the plurality of neighborhood grid indexes obtained by comparison is greater than the candidate grid index of the candidate grid characteristic parameter;

[0058] If it is greater than, the neighborhood grid feature parameter corresponding to the neighborhood optimal grid index is recorded as the optimal grid feature parameter; if it is less than, the candidate grid feature parameter is used as the optimal grid feature parameter.

[0059] Continue iterating until a predetermined number of times is reached, and output the optimal grid characteristic parameters at that time.

[0060] In a preferred embodiment, before entering the grid feature optimization process, the predetermined optimization rule randomly selects a feature indicator from the predetermined feature indicator set as the first feature indicator as the direction of optimization analysis. Then, the standard value range of the first feature indicator is queried. For example, if the grid height is 5-12 meters, the first value range is determined; a value is randomly selected within the first value range as the first feature parameter. After that, the parameter values ​​of other feature indicators except the first feature indicator are kept unchanged at the corresponding values ​​in the initial grid feature parameter set to form the predetermined grid feature parameters. Next, in the power distribution database crawled by big data, any set of grid feature parameters is selected and recorded as the first grid feature parameter, and the difference between the first grid feature parameter and the obtained predetermined grid feature parameter in each feature dimension is calculated, and the difference of each dimension is normalized and mapped to the range of 0-1 to obtain the normalized difference of each dimension, and the preset weight coefficient of each dimension set according to experience is retrieved, and weighted processing is performed. The first feature similarity can be obtained by subtracting the weighted processing value from 1, which reflects the degree of fit between the two sets of grid feature parameters. According to the predetermined similarity threshold required for the distribution network grid, the obtained first feature similarity is compared with the predetermined similarity threshold. If the first feature similarity is greater than the predetermined similarity threshold, the first grid feature parameter is integrated with the obtained predetermined grid feature parameter to form a grid parameter group and construct a predetermined grid group. If the first feature similarity is less than or equal to the predetermined similarity threshold, the first grid feature parameter set is combined with the initial grid feature parameter set as the initial grid group. By judging the size relationship between the similarities between parameters, the parameter instances are effectively divided into two groups, providing different object samples for subsequent inter-group evaluation and comparison.

[0061] Afterwards, a grid index evaluation is performed on each set of grid characteristic parameters in the predetermined grid group based on a predetermined evaluation rule, and the grid indexes of all parameter groups in the predetermined grid group are summarized to obtain a predetermined grid index sum. At the same time, an evaluation is performed on each set of grid characteristic parameters in the initial grid group based on a predetermined evaluation rule, and the grid indexes of all parameter groups in the initial grid group are summarized to obtain an initial grid index sum. Next, the predetermined grid index sum is compared with the initial grid index sum. If the predetermined grid index sum is greater than the initial grid index sum, the grid characteristic parameters used when constructing the predetermined grid group are selected as candidate grid characteristic parameters; if the initial grid index sum is greater than the predetermined grid index sum, all grid characteristic parameters used when constructing the initial grid group are selected as candidate grid characteristic parameters.

[0062] Subsequently, a parameter variation range interval is set near the current candidate grid characteristic parameter value. Within this interval, by fine-tuning and perturbing the values ​​of each characteristic parameter, multiple newly combined grid characteristic parameter instances are generated. The set of these instance parameters constitutes the first neighborhood, wherein each group of parameter instances is a group of neighborhood grid characteristic parameters, and the first neighborhood contains multiple groups of neighborhood grid characteristic parameters. Afterwards, for each group of neighborhood characteristic parameters in the first neighborhood, the voltage and power supply data during operation are simulated by the simulation platform, and then the first voltage index and the first power supply index of the group of parameters are calculated according to the simulation results according to the predetermined evaluation rules, and then the weighted method is used to integrate them into the neighborhood grid index corresponding to the group of neighborhood parameters. The simulation data of all neighborhood grid characteristic parameters in the first neighborhood are analyzed to obtain multiple neighborhood grid indexes. Then, the multiple neighborhood grid indexes are sorted from large to small to obtain the largest neighborhood grid index as the neighborhood optimal grid index. Subsequently, the neighborhood optimal grid index is compared with the candidate grid index corresponding to the current candidate grid characteristic parameter.

[0063] If the neighborhood optimal grid index is greater than the candidate grid index, it means that a more optimized solution has been found in the first neighborhood. At this time, the neighborhood grid feature parameters corresponding to the neighborhood optimal grid index are recorded as the updated optimal grid feature parameters for the next round of iterative calculations; if the neighborhood optimal grid index is less than or equal to the candidate grid index, the current candidate grid feature parameters are still optimal and continue to be used as the optimal grid feature parameters. Subsequently, in order to prevent the parameter optimization process from falling into the local optimum and being unable to continue to improve, the total number of optimization iterations, that is, the predetermined number of times, is set based on the complexity of the problem as one of the optimization termination conditions. Then, in the optimization iteration process, each time the optimal grid feature parameters calculated in the previous round are used as the current candidate grid feature parameters, its first neighborhood is obtained and searched and updated, and when the continuous iteration reaches the predetermined number of times, the iteration is terminated. At this time, the latest optimal grid feature parameters are the global optimal solution, which are used as the final optimal grid feature parameters.

[0064] The target grid is transformed and optimized according to the optimal grid characteristic parameters.

[0065] In the embodiment of the present application, after multiple rounds of iterative optimization, the optimal grid characteristic parameters including the best values ​​of grid size, grid span, grid height, grid depth and cross-sectional shape are obtained. Then, the target grid to be optimized in the target distribution network is specifically modified by the optimal grid characteristic parameters. For example, the pole configuration spacing and height are adjusted according to the optimal grid characteristic parameters; poles are added or replaced according to the grid span in the optimal grid characteristic parameters. By modifying and optimizing the target grid through the optimal grid characteristic parameters, the power supply quality and transmission efficiency of the distribution network can be comprehensively improved, and the distribution network grid upgrade can be completed.

[0066] In summary, the distribution network optimization method provided by the embodiment of the present application has the following technical effects:

[0067] Obtain the initial grid characteristic parameters to provide basic input for subsequent evaluation and optimization. Retrieve the target power distribution database to ensure the comprehensive accuracy of the data on which the evaluation and optimization are based. Randomly obtain the first distribution record from multiple groups of distribution records, the first distribution record includes the first distribution network voltage record and the first distribution network power supply record, and obtain the starting point for iterative optimization; analyze the first distribution network voltage record and the first distribution network power supply record based on predetermined evaluation rules to obtain the initial grid index and judge the pros and cons of the grid scheme. Iterate and optimize the initial grid characteristic parameters based on predetermined optimization rules to obtain the optimal grid characteristic parameters, and provide the optimal parameters for the transformation and optimization of the target grid; transform and optimize the target grid according to the optimal grid characteristic parameters, improve the actual operation quality of the distribution system, and improve the stability of the distribution network and the reliability of power supply.

[0068] Embodiment 2

[0069] Based on the same inventive concept as the distribution network optimization method in the above-mentioned embodiment, Figure 3 As shown, an embodiment of the present application provides a distribution network framework optimization system, the system comprising:

[0070] A characteristic parameter acquisition module 11 is used to acquire initial grid characteristic parameters, wherein the initial grid characteristic parameters are parameters obtained by collecting characteristics of a target grid of a target distribution network based on predetermined characteristic indicators;

[0071] A database retrieval module 12 is used to retrieve a target power distribution database, wherein the target power distribution database refers to a historical record database of the target power distribution network, and the target power distribution database includes multiple groups of power distribution records;

[0072] A power distribution record random module 13, used for randomly acquiring a first power distribution record from the plurality of groups of power distribution records, wherein the first power distribution record includes a first distribution network voltage record and a first distribution network power supply record;

[0073] A power supply record analysis module 14, configured to analyze the first distribution network voltage record and the first distribution network power supply record based on a predetermined evaluation rule to obtain an initial grid index;

[0074] The optimal parameter acquisition module 15 is used to iteratively optimize the initial grid characteristic parameters based on a predetermined optimization rule to obtain the optimal grid characteristic parameters, wherein the optimal grid characteristic parameters have an optimal grid index identifier, and the optimal grid index is greater than the initial grid index;

[0075] The grid transformation optimization module 16 is used to transform and optimize the target grid according to the optimal grid characteristic parameters.

[0076] Furthermore, the characteristic parameter acquisition module 11 includes the following contents:

[0077] The predetermined characteristic indicators include at least grid size, grid span, grid height, grid depth and cross-sectional shape.

[0078] Furthermore, the power supply record analysis module 14 includes the following execution steps:

[0079] The first distribution network voltage record includes a plurality of voltage time series at multiple points in the target distribution network;

[0080] Matching a first voltage timing of a first point in the plurality of voltage timings, wherein the first point is any one of the plurality of points;

[0081] Performing multi-feature collection on a first voltage curve generated based on the first voltage time series to obtain a first voltage curve feature set;

[0082] Analyzing the first voltage curve feature set based on the predetermined voltage evaluation rule to obtain a first voltage index;

[0083] The first power distribution network power supply record includes a first total power supply and a first power loss;

[0084] Based on the predetermined power supply evaluation rule, a normalized result of the ratio of the first power loss to the first total power supply is obtained and recorded as a first power supply index;

[0085] An average value of a first grid index obtained by weighting the first voltage index and the first power supply index is taken as the initial grid index.

[0086] Furthermore, the power supply record analysis module 14 also includes the following contents:

[0087] The first voltage curve feature set includes a first voltage qualification rate, a first voltage flicker frequency, and a first voltage fluctuation amplitude.

[0088] Furthermore, the optimal parameter acquisition module 15 includes the following execution steps:

[0089] Randomly extracting a first characteristic index from the predetermined characteristic indexes based on the predetermined optimization rule;

[0090] Obtaining a predetermined grid characteristic parameter according to a first characteristic parameter randomly obtained in combination with a first value range of the first characteristic index;

[0091] Calculate the similarity between a first grid characteristic parameter and a first characteristic of the predetermined grid characteristic parameter in a plurality of groups of grid characteristic parameters of a plurality of distribution networks in a distribution database, wherein the distribution database is a distribution network historical distribution database established based on big data crawling;

[0092] If the first feature similarity reaches a predetermined similarity threshold, a predetermined grid group is formed based on the first grid feature parameter and the predetermined grid feature parameter; if the first feature similarity does not reach the predetermined similarity threshold, an initial grid group is formed based on the first grid feature parameter and the initial grid feature parameter;

[0093] The predetermined grid group and the initial grid group are compared to obtain the optimal grid characteristic parameters.

[0094] Furthermore, the optimal parameter acquisition module 15 also includes the following execution steps:

[0095] The predetermined grid index sum of the predetermined grid group and the initial grid index sum of the initial grid group are calculated respectively;

[0096] If the predetermined grid index sum is greater than the initial grid index sum, the predetermined grid characteristic parameters are taken as candidate grid characteristic parameters; if the predetermined grid index sum is less than the initial grid index sum, the initial grid characteristic parameters are taken as candidate grid characteristic parameters;

[0097] The candidate grid feature parameters are iteratively optimized to obtain the optimal grid feature parameters.

[0098] Furthermore, the optimal parameter acquisition module 15 also includes the following execution steps:

[0099] Acquire a first neighborhood of the candidate grid feature parameters, the first neighborhood including a plurality of groups of neighborhood grid feature parameters;

[0100] Analyzing the simulation data of the plurality of groups of neighborhood grid characteristic parameters based on the predetermined evaluation rules to obtain a plurality of neighborhood grid indexes;

[0101] Determine whether the neighborhood optimal grid index among the plurality of neighborhood grid indexes obtained by comparison is greater than the candidate grid index of the candidate grid characteristic parameter;

[0102] If it is greater than, the neighborhood grid feature parameter corresponding to the neighborhood optimal grid index is recorded as the optimal grid feature parameter; if it is less than, the candidate grid feature parameter is used as the optimal grid feature parameter.

[0103] Continue iterating until a predetermined number of times is reached, and output the optimal grid characteristic parameters at that time.

[0104] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0105] Furthermore, the first or second mentioned above may not only represent an order relationship, but may also represent a specific concept, and / or refer to multiple elements that can be selected individually or in full. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A distribution network optimization method, characterized in that: include: Acquire initial grid characteristic parameters, wherein the initial grid characteristic parameters are parameters obtained by collecting characteristics of a target grid of a target distribution network based on predetermined characteristic indicators; Retrieving a target power distribution database, wherein the target power distribution database refers to a historical record database of the target power distribution network, and the target power distribution database includes multiple groups of power distribution records; Randomly acquiring a first power distribution record from the plurality of groups of power distribution records, wherein the first power distribution record includes a first distribution network voltage record and a first distribution network power supply record; Analyze the first distribution network voltage record and the first distribution network power supply record based on a predetermined evaluation rule to obtain an initial grid index; Iteratively optimizing the initial grid characteristic parameters based on a predetermined optimization rule to obtain optimal grid characteristic parameters, wherein the optimal grid characteristic parameters have an identifier of an optimal grid index, and the optimal grid index is greater than the initial grid index; Transforming and optimizing the target grid according to the optimal grid characteristic parameters; The predetermined evaluation rules include a predetermined voltage evaluation rule and a predetermined power supply evaluation rule. The first distribution network voltage record and the first distribution network power supply record are analyzed based on the predetermined evaluation rules to obtain an initial grid index, including: The first distribution network voltage record includes a plurality of voltage time series at multiple points in the target distribution network; Matching a first voltage timing of a first point in the plurality of voltage timings, wherein the first point is any one of the plurality of points; Performing multi-feature collection on a first voltage curve generated based on the first voltage time series to obtain a first voltage curve feature set; Analyzing the first voltage curve feature set based on the predetermined voltage evaluation rule to obtain a first voltage index; The first power distribution network power supply record includes a first total power supply and a first power loss; Based on the predetermined power supply evaluation rule, a normalized result of the ratio of the first power loss to the first total power supply is obtained and recorded as a first power supply index; Obtaining the initial grid index by weighted averaging the first voltage index and the first power supply index; The initial grid characteristic parameters are iteratively optimized based on a predetermined optimization rule to obtain the optimal grid characteristic parameters, including: Randomly extracting a first characteristic index from the predetermined characteristic indexes based on the predetermined optimization rule; Obtaining a predetermined grid characteristic parameter according to a first characteristic parameter randomly obtained in combination with a first value range of the first characteristic index; Calculate the similarity between a first grid characteristic parameter and a first characteristic of the predetermined grid characteristic parameter in a plurality of groups of grid characteristic parameters of a plurality of distribution networks in a distribution database, wherein the distribution database is a distribution network historical distribution database established based on big data crawling; If the first feature similarity reaches a predetermined similarity threshold, a predetermined grid group is formed based on the first grid feature parameter and the predetermined grid feature parameter; if the first feature similarity does not reach the predetermined similarity threshold, an initial grid group is formed based on the first grid feature parameter and the initial grid feature parameter; Comparing the predetermined grid group with the initial grid group to obtain the optimal grid characteristic parameters; Comparing the predetermined grid group with the initial grid group to obtain the optimal grid characteristic parameters includes: The predetermined grid index sum of the predetermined grid group and the initial grid index sum of the initial grid group are calculated respectively; If the predetermined grid index sum is greater than the initial grid index sum, the predetermined grid characteristic parameters are taken as candidate grid characteristic parameters; if the predetermined grid index sum is less than the initial grid index sum, the initial grid characteristic parameters are taken as candidate grid characteristic parameters; Iteratively optimizing the candidate grid characteristic parameters to obtain the optimal grid characteristic parameters; Iteratively optimizing the candidate grid feature parameters to obtain the optimal grid feature parameters includes: Acquire a first neighborhood of the candidate grid feature parameters, the first neighborhood including a plurality of groups of neighborhood grid feature parameters; Analyzing the simulation data of the plurality of groups of neighborhood grid characteristic parameters based on the predetermined evaluation rules to obtain a plurality of neighborhood grid indexes; Determine whether the neighborhood optimal grid index among the plurality of neighborhood grid indexes obtained by comparison is greater than the candidate grid index of the candidate grid characteristic parameter; If it is greater than, the neighborhood grid feature parameter corresponding to the neighborhood optimal grid index is recorded as the optimal grid feature parameter; if it is less than, the candidate grid feature parameter is used as the optimal grid feature parameter. Continue iterating until a predetermined number of times is reached, and output the optimal grid characteristic parameters at that time.

2. The method according to claim 1, characterized in that: The predetermined characteristic indicators include at least grid size, grid span, grid height, grid depth and cross-sectional shape.

3. The method according to claim 1, characterized in that: The first voltage curve feature set includes a first voltage qualification rate, a first voltage flicker frequency, and a first voltage fluctuation amplitude.

4. A distribution network optimization system, characterized in that: A method for optimizing a distribution network structure according to any one of claims 1 to 3, the system comprising: A characteristic parameter acquisition module, the characteristic parameter acquisition module is used to acquire initial grid characteristic parameters, the initial grid characteristic parameters are parameters obtained by collecting characteristics of a target grid of a target distribution network based on predetermined characteristic indicators; A database retrieval module, the database retrieval module is used to retrieve a target power distribution database, the target power distribution database refers to a historical record database of the target power distribution network, and the target power distribution database includes multiple groups of power distribution records; A power distribution record random module, the power distribution record random module is used to randomly obtain a first power distribution record from the multiple groups of power distribution records, the first power distribution record includes a first distribution network voltage record and a first distribution network power supply record; A power supply record analysis module, the power supply record analysis module is used to analyze the first distribution network voltage record and the first distribution network power supply record based on a predetermined evaluation rule to obtain an initial grid index; An optimal parameter acquisition module, the optimal parameter acquisition module is used to iteratively optimize the initial grid characteristic parameters based on a predetermined optimization rule to obtain optimal grid characteristic parameters, the optimal grid characteristic parameters have an optimal grid index identifier, and the optimal grid index is greater than the initial grid index; A grid transformation optimization module is used to transform and optimize the target grid according to the optimal grid characteristic parameters.

Images