Low-voltage distribution area electric energy meter transformation ratio identification method based on particle swarm algorithm

Abstract

This invention discloses a method for identifying the transformer ratio of energy meters in low-voltage distribution transformer areas based on particle swarm optimization (PSO) algorithm, relating to the field of power system metering technology. The method includes: given that a predetermined number of three-phase user energy meters in the transformer area require transformer ratio correction, dividing the transformer area operation dataset into a training set for model training and a validation set for verification; defining the variables to be solved as the user identities requiring correction and their corresponding correct transformer ratio parameters; constructing a single comprehensive evaluation function to evaluate the correction effect of any given user identity and transformer ratio combination by integrating multiple transformer area operation features; employing a swarm intelligence optimization algorithm, using the user identity and transformer ratio combination as optimization variables and the comprehensive evaluation function as the optimization objective, and searching in the solution space by simulating swarm iterative optimization behavior; and outputting the globally optimal solution obtained by the search when the swarm intelligence optimization algorithm meets a preset convergence condition.

Description

Technical Field

[0001] This invention relates to the field of power system metering technology, and in particular to a method for identifying the transformer ratio of low-voltage distribution substation energy meters based on particle swarm optimization algorithm. Background Technology

[0002] As a crucial link at the end of the power system, the accuracy of metering in low-voltage distribution substations directly impacts the level of lean power grid management. Current research on metering parameter correction in the distribution network field mainly focuses on two areas: transformer parameter estimation and topology identification. However, existing research has significant gaps in addressing the identification of unknown transformer ratios on the user side. Most methods assume accurate user metering parameters and only focus on substation-side parameter correction; the few studies involving user-side correction also primarily target single users, lacking systematic solutions for simultaneous identification of multiple users. Especially in low-voltage distribution substation scenarios, where three-phase and single-phase users operate in a mixed configuration, traditional methods struggle to adapt to the combined optimization problem.

[0003] Many methods have emerged for metering calibration in low-voltage distribution transformer areas, including manual verification, threshold analysis, and multiple regression. Manual verification involves checking the nameplate markings on the transformers on-site, manually recording the transformation ratio parameters, and correcting the metering data. Its core reliance is on the accuracy of the physical equipment markings; essentially, it's a manual parameter entry process. Manual verification suffers from low efficiency, inability to detect missing or incorrectly affixed nameplates, and relatively low accuracy in older transformer areas. Threshold analysis sets a threshold for the difference between the transformer's power and the total power of the users, identifying users with excessive differences as having abnormal transformation ratios. Threshold analysis relies on experience to set thresholds, cannot adapt to load fluctuations, is prone to misjudgments, and cannot distinguish errors caused by wiring. Multiple regression constructs a linear regression model between user power and transformer power, identifying users with large residuals as having abnormal transformation ratios. This method assumes a linear relationship between user and transformer power, ignoring factors such as three-phase imbalance and harmonic pollution, resulting in low accuracy in complex load scenarios and the inability to identify multiple users simultaneously. Summary of the Invention

[0004] The main objective of this invention is to provide a method for identifying the transformer ratio of energy meters in low-voltage distribution substations based on particle swarm optimization (PSO). This method addresses the shortcomings of current methods for identifying abnormal transformer ratios of multiple three-phase users' energy meters in low-voltage distribution substations, such as low efficiency, poor adaptability, and one-sided optimization objectives. By constructing a PSO optimization model that integrates multiple electrical feature constraints, this method achieves high-precision and automated identification of the identities of multiple abnormal users and their correct transformer ratios simultaneously, thereby improving the accuracy of power grid metering and the level of lean management.

[0005] To achieve the above objectives, the first aspect of this application provides a method for identifying the transformer ratio of a low-voltage distribution substation based on a particle swarm optimization algorithm, comprising: Given that a predetermined number of three-phase user energy meter transformation ratios within the transformer area need to be corrected, the transformer area operation dataset is divided into a training set for model training and a validation set for validation; the variables to be solved are defined as the user identities that need to be corrected and their corresponding correct transformation ratio parameters. A single comprehensive evaluation function is constructed, which evaluates the correction effect of any given user identity and ratio combination by integrating multiple transformer area operation characteristics. The comprehensive evaluation function consists of a guiding fitness term and a constraint penalty term, wherein the guiding fitness term is used to drive the solution vector to approach the ideal transformer area operation state, and the constraint penalty term is used to impose a negative evaluation on the solution vector that violates the basic operation constraints of the transformer area. A swarm intelligence optimization algorithm is adopted, with the user identity and the combination of ratios as optimization variables and the comprehensive evaluation function as optimization objectives. By simulating the iterative optimization behavior of the swarm, a search is performed in the solution space. When the swarm intelligence optimization algorithm satisfies the preset convergence condition, it outputs the globally optimal solution obtained from the search.

[0006] Optionally, the operating characteristics of the plurality of transformer substations include at least: The daily line loss rate of the transformer substation, the first correlation coefficient based on amplitude linear correlation, the second correlation coefficient based on sequence monotonic correlation, and the mean ratio of active power at the transformer outlet to the total active power of users.

[0007] Optionally, the guiding fitness term is obtained by mapping the multiple station operation characteristics to a unified scoring interval and then performing a weighted sum. The constraint penalty term is obtained by quantifying and weighting the degree to which the operating characteristics of the multiple transformer substations deviate from their respective preset reasonable ranges.

[0008] Optionally, the swarm intelligence optimization algorithm includes a step of discretizing the optimization variables during the iteration process. The discretization process is used to ensure that the variable components of the generated solution vector all belong to a predefined set of discrete candidate values.

[0009] Optionally, when initializing the optimization algorithm, a hybrid strategy combining deterministic initialization and random initialization that incorporates prior knowledge is used to generate the initial population.

[0010] Optionally, the preset convergence conditions include: The number of iterations reaches a preset maximum value, or the improvement in the fitness value of the global optimal solution over multiple iterations is less than a preset threshold.

[0011] Optionally, during the iterative optimization process, the movement direction and step size of each search individual are dynamically updated based on the individual's historical best solution and the group's historical best solution.

[0012] Secondly, a low-voltage distribution substation energy meter transformation ratio identification device based on particle swarm optimization algorithm is provided, the device comprising: The data acquisition and processing module is used to divide the transformer operation dataset into a training set for model training and a validation set for verification, under the premise that a predetermined number of three-phase user energy meter transformation ratios need to be corrected within the transformer area; the variables to be solved are defined as the user identities that need to be corrected and their corresponding correct transformation ratio parameters. The optimization modeling module is used to construct a single comprehensive evaluation function. This comprehensive evaluation function evaluates the correction effect of any given user identity and ratio combination by integrating multiple transformer area operation characteristics. The comprehensive evaluation function consists of a guiding fitness term and a constraint penalty term. The guiding fitness term is used to drive the solution vector to approach the ideal transformer area operation state, and the constraint penalty term is used to impose a negative evaluation on the solution vector that violates the basic operating constraints of the transformer area. The swarm intelligence optimization solution module is used to employ a swarm intelligence optimization algorithm, taking the user identity and the combination of ratios as optimization variables and the comprehensive evaluation function as optimization objectives, and searching in the solution space by simulating the iterative optimization behavior of the swarm. The result output module is used to output the globally optimal solution obtained by the search when the swarm intelligence optimization algorithm meets the preset convergence conditions.

[0013] This application provides a particle swarm optimization (PSO) algorithm-based method for identifying the transformer ratio of energy meters in low-voltage distribution substations. This method aims to overcome the shortcomings of existing methods for identifying the transformer ratio of three-phase users in low-voltage distribution substations, such as low efficiency, poor adaptability, and one-sided optimization objectives. Its core objective is to construct a PSO optimization model that integrates multiple electrical characteristic constraints, thereby achieving high-precision, automated, and synchronous identification of multiple abnormal user identities and their correct transformer ratios. This scheme standardizes multiple indicators such as line loss rate and correlation coefficient and introduces penalty terms to form an objective function that combines guidance and constraints. It utilizes the global search capability of swarm intelligence algorithms to solve the discrete combinatorial optimization problem of "multiple user candidates + finite transformer ratio set," ultimately significantly improving the accuracy of power grid metering and the level of lean management. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] in: Figure 1A flowchart illustrating a method for identifying the transformer ratio of a low-voltage distribution substation energy meter based on particle swarm optimization algorithm, provided in an embodiment of this application. Figure 2 This is a schematic diagram of an iterative fitness curve provided in an embodiment of this application; Figure 3 A schematic diagram showing the comparison of line loss rate for each cycle before and after correction, provided as an embodiment of this application; Figure 4 A schematic diagram showing the comparison of correlation coefficients for each period before and after correction, provided as an embodiment of this application; Figure 5 This application provides a schematic diagram comparing the ratio of each mean before and after correction, as part of an embodiment of the present application. Figure 6 A schematic diagram showing the comparison between the active power of the transformer in each cycle and the total active power of the user before and after correction, provided for an embodiment of this application; Figure 7 A schematic diagram of a low-voltage distribution substation energy meter transformation ratio identification device based on particle swarm optimization algorithm provided in this application embodiment; Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0016] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0017] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0018] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0019] The embodiments of this application are described below with reference to the accompanying drawings.

[0020] Figure 1 This is a flowchart illustrating a method for identifying the transformer ratio of a low-voltage distribution substation energy meter based on the particle swarm optimization algorithm, provided in an embodiment of this application.

[0021] like Figure 1 As shown, the method includes: 101. Given that there is a predetermined number of three-phase users in the transformer area whose energy meter transformation ratios need to be corrected, the transformer area operation dataset is divided into a training set for model training and a validation set for validation; the variables to be solved are defined as the user identities that need to be corrected and their corresponding correct transformation ratio parameters.

[0022] Specifically, when the energy meters of some three-phase users in a low-voltage distribution substation require transformation ratio calibration, but the specific user numbers and transformation ratios are unknown, firstly, in cases where it is known that... When the electricity meters of individual users have different transformation ratios, the dataset collected on-site can be divided into a training set and a validation set. The user ID to be determined can be defined as... The variable to be determined is .

[0023] In one optional implementation, the above-mentioned multiple transformer area operating characteristics include at least: The daily line loss rate of the transformer substation, the first correlation coefficient based on amplitude linear correlation, the second correlation coefficient based on sequence monotonic correlation, and the mean ratio of active power at the transformer outlet to the total active power of users.

[0024] Among them, the first correlation coefficient can be the Pearson correlation coefficient, and the second correlation coefficient can be the Spearman correlation coefficient.

[0025] 102. Construct a single comprehensive evaluation function. The comprehensive evaluation function evaluates the correction effect of any given user identity and ratio combination by integrating multiple transformer area operation characteristics. The comprehensive evaluation function consists of a guiding fitness term and a constraint penalty term. The guiding fitness term is used to drive the solution vector to approach the ideal transformer area operation state, and the constraint penalty term is used to impose a negative evaluation on the solution vector that violates the basic operation constraints of the transformer area.

[0026] In this embodiment of the application, the above-mentioned guiding fitness term can be obtained by using the above-mentioned basic fitness or by using the basic fitness.

[0027] In one optional implementation, the aforementioned guiding fitness term is obtained by weighted summation after mapping the above-mentioned multiple station operation characteristics to a unified scoring interval. The aforementioned restrictive penalty items are obtained by quantifying and weighting the degree to which the operating characteristics of the aforementioned multiple transformer substations deviate from their respective preset reasonable ranges.

[0028] Specifically, the line loss rate can be... Pearson correlation coefficient Spearman correlation coefficient Ratio to mean The four features are standardized and mapped to a unified score range, and the weighted sum of the features yields the basic fitness. .

[0029] In one implementation, the standardized formulas corresponding to each feature are calculated using the following formulas (1)-(4). The weights of each feature are determined by combining the timing requirements of the low-voltage distribution area: (The most fundamental physical constraints directly reflect the rationality of power loss.) (Reflects the linearity of the power curve, less important) (Reflects the monotonicity of the power curve, equally important as the Pearson curve.) (Reflects the power amplitude matching degree, its importance is comparable to the correlation coefficient). The basic fitness is obtained by weighted summation of each feature. .

[0030] (1) (2) (3) (4) Basic fitness can only guide features toward the ideal value, but it cannot enforce constraints. Therefore, penalty terms need to be designed to impose additional penalties on features that violate the constraints.

[0031] In one implementation, the penalty terms corresponding to each feature can be obtained from the following formulas (5)-(8). , , , ,in The penalty coefficients for each feature are used as the basis for summing the penalty terms for each feature to obtain the total penalty term. The basic fitness With total penalty item The overall objective function is obtained after fusion. As shown in equation (9), the final result is achieved by maximizing Find the optimal solution.

[0032] (5) (6) (7) (8) (9) 103. Using a swarm intelligence optimization algorithm, the user identity and ratio combination mentioned above are used as optimization variables, and the comprehensive evaluation function mentioned above is used as the optimization objective. By simulating the iterative optimization behavior of the swarm, a search is performed in the solution space.

[0033] The swarm intelligence optimization algorithm in this application embodiment can be the particle swarm optimization algorithm (PSO), or it can be selected from genetic algorithms, ant colony algorithms, etc. as needed, without any restrictions.

[0034] In one alternative implementation, when initializing the optimization algorithm, a hybrid strategy combining deterministic initialization with random initialization that incorporates prior knowledge is used to generate the initial population.

[0035] This can be understood as allowing the setting of key PSO parameters and initial solutions, with a small number of custom initial solutions and a large number of random solutions.

[0036] Optionally, during the iterative optimization process, the movement direction and step size of each search individual are dynamically updated based on the individual's historical best solution and the group's historical best solution. This summarizes the core iterative principle of the swarm intelligence optimization algorithm (especially PSO) in this application, namely, a dynamic update mechanism based on "individual experience" and "social experience".

[0037] In one alternative implementation, the above-mentioned swarm intelligence optimization algorithm includes a step of discretizing the optimization variables during the iteration process. The discretization process is used to ensure that the variable components of the generated solution vector all belong to a predefined set of discrete candidate values.

[0038] The aforementioned predefined set of discrete candidate values ​​includes a set of all three-phase user identifiers that may be abnormal, as well as a finite set of standard transformer ratio parameters of the electricity meter.

[0039] 104. When the above swarm intelligence optimization algorithm satisfies the preset convergence condition, the globally optimal solution obtained by the search is output.

[0040] The optimal solution represents the identified user identity to be corrected and its correct ratio parameter.

[0041] Specifically, a series of key PSO parameters, such as the number of particles, maximum number of iterations, learning factor, and inertia weight, can be set as needed. The predefined variables and objective function are substituted into the model for solution and verification. When setting the initial solution, a strategy of "a few custom initial solutions + a large number of random solutions" is adopted. The PSO algorithm enters a loop iteration, updating the particle velocity and position in each iteration.

[0042] Specifically, in each iteration, the velocity and position of each particle can be updated according to formulas (10)-(11). Since the variables to be determined are discrete values, the validity of the position needs to be corrected after the update, the fitness value of the updated particle is recalculated, and the individual optimal value is updated according to the fitness value. with global optimal .

[0043] (10) (11) In one optional implementation, the aforementioned preset convergence conditions include: The number of iterations reaches the preset maximum value, or the improvement in the fitness value of the global optimal solution over multiple iterations is less than the preset threshold.

[0044] The aforementioned preset maximum value and preset threshold can be set and adjusted as needed. Specifically, for example, when the PSO algorithm reaches a preset maximum number of iterations or the global optimum... The fitness did not improve significantly over 20 consecutive iterations (change amount) When the algorithm converges, the global optimal solution is found. That is, the solution that satisfies all constraints.

[0045] The method in this application embodiment has at least the following advantages compared with the prior art: (1) It has a more adaptable optimization framework. For discrete optimization problems, it adopts the particle swarm optimization algorithm (PSO) to simulate the group search behavior. Compared with continuous optimization methods such as linear regression, the solution accuracy is greatly improved.

[0046] (2) It has a multi-constraint fusion objective function, which solves the one-sided problem of the traditional method of "single index optimization".

[0047] (3) It has higher efficiency and accuracy than manual verification.

[0048] The method proposed in this application will be further explained below in conjunction with specific application scenarios.

[0049] Taking a low-voltage distribution area in a certain region as an example, this area has one transformer and 38 users. Due to equipment upgrades and the introduction of high-power equipment by some users, some meters needed to be replaced, resulting in inaccurate metering data. Upon inspection, it was found that the three-phase transformation ratios of two of the 38 three-phase users were unknown, and uncorrected active power data for five cycles (August 24-August 28) were collected on-site.

[0050] (1) When it is known that there are two users whose electricity meters have transformation ratios but the specific user numbers and transformation ratios are unknown, the first three cycles are used as the training set and the last two cycles are used as the validation set.

[0051] (2) A survey of transformer substations with the same voltage level and similar user types revealed that the daily line loss rate of this type of substation was... Should be in Between, Pearson correlation coefficient The Spearman correlation coefficient should be no less than 0.8. It should be no less than 0.85, the mean ratio Should be in Between. The above conditions are taken as... Constraints of the algorithm.

[0052] (3) Calculate the line loss rate for each cycle before correction. Pearson correlation coefficient Spearman correlation coefficient Ratio to mean After standardizing the four features, we obtain , , , Assign values ​​to each penalty coefficient in the penalty term, where... , , , .

[0053] (4) Define the range of the three-phase users to be determined, and determine the possible transformer ratios based on the voltage level and load level of the transformer area. Set the key parameters and initial solution of PSO, where the number of particles is 300, the maximum number of iterations is 1000, and the inertia weight is... Cognitive coefficient Social coefficient Three initial solutions are defined, and the rest are automatically generated randomly.

[0054] Three-phase user ID:

[0055] Possible ratios:

[0056] (5) Preliminary optimization yielded the following results: (User 22, Variation Ratio 5; User 37, Variation Ratio 40). Substituting this solution into the training and validation sets revealed that some indicators did not meet the constraints, and the optimal fitness no longer changed after the 15th iteration. This indicates that the algorithm has fallen into local convergence, and the social coefficient needs to be increased. This increases the intensity of particle learning towards the global optimal solution. Social coefficient As the value increases from 1.8 to 2.0, as shown in Table 1, the globally optimal solution (user 37, ratio 30; user 22, ratio 60) no longer changes, and the iterative fitness curve is as follows. Figure 2 As shown, the iterative fitness converges with each iteration, indicating that the optimization process is effective. The social coefficient... Substituting the globally optimal solution obtained at that time into the equation yields the line loss rates before and after correction. like Figure 3 As shown, Pearson correlation coefficient Correlation coefficient with Spearman like Figure 4 As shown, the mean ratio like Figure 5 As shown, both the training and validation sets meet the requirements after verification. The active power of the transformer in each cycle is compared with the total active power of the three-phase users before and after correction. Figure 6 As shown.

[0057]

[0058] Table 1 This example verifies the effectiveness, robustness, and engineering practical value of the technical solution presented in this application under complex real-world scenarios. In the case study, given that two users have abnormal ratios but their specific information is unknown, this embodiment successfully combines the PSO algorithm with social coefficient adjustment, overcoming local optima and ultimately accurately identifying users 37 (ratio 30) and 22 (ratio 60) as abnormal users. After correction, as shown... Figure 2 , Figure 3 , Figure 4 As shown, the line loss rates of both the training and validation sets remained stable within a reasonable range, the correlation coefficient significantly improved, and the mean ratio approached 1.0. This comprehensively verifies the correctness of the algorithm model, its excellent generalization ability, and its direct effectiveness in solving the problem of inaccurate measurement in actual transformer substations.

[0059] Based on the description of the foregoing method embodiments, this application also provides a low-voltage distribution substation energy meter transformation ratio identification device based on particle swarm optimization algorithm.

[0060] like Figure 7 As shown, the low-voltage distribution substation energy meter transformation ratio identification device 700 based on particle swarm optimization algorithm may include: The data acquisition and processing module 710 is used to divide the transformer operation dataset into a training set for model training and a validation set for validation, under the premise that a predetermined number of three-phase user energy meter transformation ratios need to be corrected within the transformer area; and to define the variables to be solved as the user identities that need to be corrected and their corresponding correct transformation ratio parameters. The optimization modeling module 720 is used to construct a single comprehensive evaluation function. This comprehensive evaluation function evaluates the correction effect of any given user identity and ratio combination by integrating multiple transformer area operation characteristics. The comprehensive evaluation function consists of a guiding fitness term and a constraint penalty term. The guiding fitness term is used to drive the solution vector to approach the ideal transformer area operation state, and the constraint penalty term is used to impose a negative evaluation on the solution vector that violates the basic operating constraints of the transformer area. The swarm intelligence optimization solution module 730 is used to employ a swarm intelligence optimization algorithm, taking the above-mentioned user identity and ratio combination as optimization variables and the above-mentioned comprehensive evaluation function as optimization objective, and searching in the solution space by simulating the iterative optimization behavior of the swarm. The result output module 740 is used to output the globally optimal solution obtained by the search when the above swarm intelligence optimization algorithm meets the preset convergence conditions.

[0061] It is understood that the relevant content concerning each module in the above-mentioned device has been described in detail in the foregoing method embodiments, and specific details can be found in the method embodiments; that is, the low-voltage distribution substation energy meter transformation ratio identification device 700 based on particle swarm optimization algorithm provided in this application can perform the following... Figure 1 Any steps in the illustrated embodiments will not be described in detail here.

[0062] In one embodiment of this application, an electronic device is also provided. See also... Figure 8 , Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 8 As shown, the electronic device 800 includes a processor 801 and a memory 802. The memory 802 stores a computer program, which, when executed by the processor 801, will perform actions such as... Figure 1 Any step in the method embodiment shown can be a control method step in the experimental process, such as controlling and adjusting the transformer ratio of the voltage regulator, monitoring the phase difference, etc. The electronic device 800 may also include input / output devices, etc. In a specific embodiment, the electronic device can be a terminal device, etc.

[0063] In one embodiment, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor 801, causes the processor 801 to perform any of the steps in the above method embodiments.

[0064] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0065] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0066] The embodiments described above are merely examples of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these modifications and improvements all fall within the protection scope of this application.

Claims

1. A method for identifying the transformer ratio of energy meters in low-voltage distribution substations based on particle swarm optimization, characterized in that, include: Given that a predetermined number of three-phase user energy meter transformation ratios within the transformer area need to be corrected, the transformer area operation dataset is divided into a training set for model training and a validation set for validation; the variables to be solved are defined as the user identities that need to be corrected and their corresponding correct transformation ratio parameters. A single comprehensive evaluation function is constructed, which evaluates the correction effect of any given user identity and ratio combination by integrating multiple transformer area operation characteristics. The comprehensive evaluation function consists of a guiding fitness term and a constraint penalty term, wherein the guiding fitness term is used to drive the solution vector to approach the ideal transformer area operation state, and the constraint penalty term is used to impose a negative evaluation on the solution vector that violates the basic operation constraints of the transformer area. A swarm intelligence optimization algorithm is adopted, with the user identity and the combination of ratios as optimization variables and the comprehensive evaluation function as optimization objectives. By simulating the iterative optimization behavior of the swarm, a search is performed in the solution space. When the swarm intelligence optimization algorithm satisfies the preset convergence condition, it outputs the globally optimal solution obtained from the search.

2. The method for identifying the transformer ratio of low-voltage distribution substation energy meters based on particle swarm optimization algorithm according to claim 1, characterized in that, The operational characteristics of the multiple transformer areas include at least the following: The daily line loss rate of the transformer substation, the first correlation coefficient based on amplitude linear correlation, the second correlation coefficient based on sequence monotonic correlation, and the average ratio of active power at the transformer substation outlet to the total active power of users.

3. The method for identifying the transformer ratio of low-voltage distribution substation energy meters based on particle swarm optimization algorithm according to claim 1, characterized in that, The guiding fitness term is obtained by mapping the multiple station operation characteristics to a unified scoring interval and then performing a weighted summation. The constraint penalty term is obtained by quantifying and weighting the degree to which the operating characteristics of the multiple transformer substations deviate from their respective preset reasonable ranges.

4. The method for identifying the transformer ratio of low-voltage distribution substation energy meters based on particle swarm optimization algorithm according to claim 1, characterized in that, The swarm intelligence optimization algorithm includes a step of discretizing the optimization variables during the iteration process. The discretization process is used to ensure that the variable components of the generated solution vector all belong to a predefined set of discrete candidate values.

5. The method for identifying the transformer ratio of low-voltage distribution substation energy meters based on particle swarm optimization algorithm according to claim 1, characterized in that, When initializing the optimization algorithm, a hybrid strategy combining deterministic initialization with random initialization based on prior knowledge is used to generate the initial population.

6. The method for identifying the transformer ratio of a low-voltage distribution substation energy meter based on particle swarm optimization algorithm according to claim 1, characterized in that, The preset convergence conditions include: The number of iterations reaches a preset maximum value, or the improvement in the fitness value of the global optimal solution over multiple iterations is less than a preset threshold.

7. The method for identifying the transformer ratio of low-voltage distribution substation energy meters based on particle swarm optimization algorithm according to claim 1, characterized in that, During the iterative optimization process, the movement direction and step size of each search individual are dynamically updated based on the individual's historical best solution and the group's historical best solution.

8. A low-voltage distribution substation energy meter transformer ratio identification device based on particle swarm optimization algorithm, characterized in that, The device includes: The data acquisition and processing module is used to divide the transformer operation dataset into a training set for model training and a validation set for verification, under the premise that a predetermined number of three-phase user energy meter transformation ratios need to be corrected within the transformer area; the variables to be solved are defined as the user identities that need to be corrected and their corresponding correct transformation ratio parameters. The optimization modeling module is used to construct a single comprehensive evaluation function. This comprehensive evaluation function evaluates the correction effect of any given user identity and ratio combination by integrating multiple transformer area operation characteristics. The comprehensive evaluation function consists of a guiding fitness term and a constraint penalty term. The guiding fitness term is used to drive the solution vector to approach the ideal transformer area operation state, and the constraint penalty term is used to impose a negative evaluation on the solution vector that violates the basic operating constraints of the transformer area. The swarm intelligence optimization solution module is used to employ a swarm intelligence optimization algorithm, taking the user identity and the combination of ratios as optimization variables and the comprehensive evaluation function as optimization objectives, and searching in the solution space by simulating the iterative optimization behavior of the swarm. The result output module is used to output the globally optimal solution obtained by the search when the swarm intelligence optimization algorithm meets the preset convergence conditions.

9. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, When the computer program is executed by a processor, the processor performs the steps of the method as described in any one of claims 1-7.

Images