Substation planning method and system, storage medium and equipment

Abstract

The invention discloses a transformer substation planning method and system, a storage medium and equipment, and the method comprises the steps: obtaining the historical power utilization data of a user in a target region, and recognizing a typical load time sequence curve through clustering analysis; predicting the peak load of each block in the target area based on a load density method, and generating time sequence load data in combination with a typical load curve; establishing a transformer substation planning optimization model, and outputting the site selection of the transformer substation in the target area by taking a power balance constraint and a power supply radius constraint as constraint conditions and taking the minimization of the total cost as a target; an improved genetic algorithm is adopted to optimize site selection of the transformer substation, and the optimal position of the transformer substation in the target area is screened out from the candidate site set; and dividing the power supply range of each transformer substation in the target area based on the Thiessen polygon, and determining the capacity of the transformer substation according to the total peak load in the power supply range. According to the method, load characteristic complementarity is considered, substation locating and sizing are optimized, the station number and capacity are reduced, loss is reduced, and planning economy is improved.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, specifically to a substation planning method, system, storage medium, and device. Background Technology

[0002] Substation planning is a fundamental aspect of urban power grid development. It not only has a profound impact on the structural integrity and reliability of regional power distribution networks, but also relates to the economic efficiency of urban power infrastructure investment.

[0003] In recent years, driven by emerging factors and the application of new technologies, the landscape of substation planning has undergone significant evolution. Advanced methods leveraging Geographic Information Systems (GIS) and supervised learning techniques have been introduced to optimize substation site selection and capacity determination processes. To address the challenges of planning static distribution substations in large-scale networks, deterministic heuristics and expert selection strategies have been developed. Furthermore, methods suitable for integrating distributed energy resources (DERs) into substation planning are gaining increasing attention. These methods utilize sophisticated spatial analysis techniques, such as weighted Thiessen polygons (Voronoi diagrams), to optimize substation layouts in areas with high DER penetration. Considering the stochasticity of electricity demand and the growing importance of resilience in power system planning, researchers have proposed innovative methods for robust, multi-stage substation expansion planning. These methods employ advanced optimization techniques, including mixed-integer second-order cone programming, to account for uncertainties and ensure the reliability of future substation deployments. In addition, optimization models considering investment costs, losses, maintenance, reliability, and residual value have been developed for long-term capacity management of transformers within substations. Against the backdrop of growing concerns about climate change and the transition to a low-carbon economy, researchers have explored novel substation planning methods aimed at minimizing carbon emissions and maximizing the use of renewable energy. These methods typically employ a combination of sophisticated optimization algorithms and geographic information systems to determine the optimal locations of substations in active distribution networks.

[0004] Despite significant progress in substation site selection and capacity determination methods, challenges remain in effectively addressing the complementarity of load characteristics across different areas. Traditional techniques that heavily rely on simultaneity factors to correct for load variations between plots have proven insufficient, leading to inaccurate load estimations and uneven load distribution. This inefficiency results in underutilization of transformers in areas with complementary load characteristics, while simultaneously overloading transformers in areas with similar load distributions. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide a substation planning method, system, storage medium and equipment, which aims to solve the above-mentioned problems described in the prior art.

[0006] A first aspect of the present invention relates to a substation planning method, the method comprising: Obtain historical electricity consumption data of users within the target area, and use cluster analysis to identify typical load time-series curves, including residential load, commercial load, and industrial load curves; The peak load of each block in the target area is predicted based on the load density method, and the time-series load data of each block in the target area is generated by combining the typical load curve. A substation planning optimization model is established, with power balance constraints and power supply radius constraints as constraints and minimizing total cost as the objective, and the substation site selection within the target area is output. An improved genetic algorithm is used to optimize the substation site selection, and the optimal location of the substation within the target area is selected from the candidate site set; The power supply range of each substation within the target area is divided based on Thiessen polygons, and the substation capacity is determined based on the total peak load within the power supply range.

[0007] According to one aspect of the above technical solution, the method of identifying typical load time series curves using cluster analysis specifically includes: The user's historical electricity consumption data is processed using a deep convolutional embedding clustering method. The user's historical electricity consumption data includes load data at 24 time points every day of the year, in order to identify typical daily load curves for residential load, commercial load, and low-voltage industrial load.

[0008] According to one aspect of the above technical solution, the generation of time-series load data by combining the typical load curve is achieved through the following formula: P i ( t )= P {i,max} * L {type(i)} ( t ); in, P i ( t ) is the first i Each block at any time t The load, P {i,max} For the first predicted by the load density method i Peak load of each block, L {type(i)} ( t ) is the first i The load type of each block at time t The normalized value of the typical load curve.

[0009] According to one aspect of the above technical solution, with the objective of minimizing the total cost, the formula for calculating the total cost is as follows: ; ; ; ; In the formula, C sub , C line and C loss These are the substation investment cost, the line investment cost, and the network loss cost, respectively. f ( ) is the substation investment function; N The number of substations; S k For substation k The capacity; J k For substation k A collection of neighborhoods with access to electricity; l ik To the substation k To the neighborhood i The length of the line; α The investment cost per unit length of the line; β The network loss cost coefficient is used; the substation investment cost and line investment cost are converted into annual cost, where CRF is the annual cost coefficient.

[0010] According to one aspect of the above technical solution, the constraints for calculating the total cost include power balance constraints and power supply radius constraints, wherein: The power balance constraint satisfies the following formula: ; ; In the formula, W k ( t ) is the first k The substation at time t The active load it carries, S k For the first k The configuration capacity of the substation, cosφ For power factor, e ( n k For consideration N Transformer load rate after the -1 criterion; The power supply radius constraint satisfies the following formula: ; In the formula, R k For substation k The power supply distance; R c This refers to the power supply radius of the substation.

[0011] According to one aspect of the above technical solution, the method of optimizing substation site selection using an improved genetic algorithm includes: The maximum number of substations is determined based on the total load of the target area and the maximum allowable capacity of a single substation. The candidate sites are binary-coded using the maximum number of substations as the chromosome length. A single-point crossover operator and a mutation operator are used for population iteration, and duplicate site genes in the chromosome are removed during decoding to obtain an effective substation site selection scheme. The expression for optimizing substation site selection using the improved genetic algorithm is as follows: ; In the formula, N max This represents the maximum number of substations required. S min ε represents the minimum capacity of a given substation; ε is the capacity-to-load ratio, which is a measure of the ratio of substation capacity to required power.

[0012] According to one aspect of the above technical solution, the division of the power supply range of each substation based on Thiessen polygons specifically involves: Using the selected substation site coordinates as the generation point, a Thiessen polygon is constructed, and the area within each polygon is the power supply range of the substation; wherein, the power supply radius is defined as the maximum distance from the substation site to each vertex of its Thiessen polygon. The determination of the substation capacity specifically involves: Calculate the total peak load within each power supply range, and select the closest standard transformer capacity level as the configuration capacity of the substation based on the total peak load.

[0013] A second aspect of the present invention is to provide a substation planning system, applied to the method described in the above-mentioned technical solution, the system comprising: The data analysis module is used to acquire historical electricity consumption data of users in the target area and to identify typical load time-series curves, including residential load, commercial load and industrial load curves, using cluster analysis. The load forecasting module is used to predict the peak load of each block in the target area based on the load density method, and generate time-series load data of each block in the target area by combining the typical load curve. The location selection output module is used to establish a substation planning optimization model, with power balance constraints and power supply radius constraints as constraints and minimizing total cost as the objective, and output the substation location within the target area; The site selection optimization module is used to optimize the substation site selection using an improved genetic algorithm, and to select the optimal location of the substation within the target area from the candidate site set; The capacity configuration module is used to divide the power supply range of each substation within the target area based on Thiessen polygons, and to determine the substation capacity according to the total peak load within the power supply range.

[0014] A third aspect of the present invention is to provide a readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the steps of the method described in the above-described technical solution.

[0015] A fourth aspect of the present invention is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method described in the above technical solutions.

[0016] Compared with existing technologies, the substation planning method, system, storage medium, and equipment shown in this invention have the following advantages: The method described in this embodiment achieves a significant improvement in overall benefits in substation planning by systematically considering the complementarity of load characteristics. First, the method accurately identifies the temporal characteristic curves of different types of loads using deep convolutional embedding clustering technology, fundamentally changing the load estimation distortion problem caused by traditional reliance on static simultaneity coefficients. Then, it combines spatiotemporal load prediction with a multi-objective optimization model, uses an improved genetic algorithm for substation site selection optimization, and scientifically divides the power supply area using Thiessen polygons, ultimately forming an overall planning scheme. This technical approach allows the planning results to fully utilize the peak-valley complementarity characteristics of loads in different regions, not only simplifying the calculation process but also improving search efficiency, enabling rapid decision-making in complex planning scenarios. Attached Figure Description

[0017] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart illustrating the substation planning method provided in an embodiment of the present invention. Figure 2A schematic diagram of the normalized daily time-series curves of three typical loads in substation planning provided for embodiments of the present invention; Figure 3 A schematic diagram illustrating the division of power supply range based on Thiessen polygons, provided for an embodiment of the present invention; Figure 4 This is a structural block diagram of a substation planning system provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Several embodiments of the present invention are shown in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that the disclosure of the present invention will be more thorough and complete.

[0019] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0021] Example 1 Please see Figure 1 The first aspect of the present invention is to provide a substation planning method, the method comprising steps S10-S50: Step S10: Obtain historical electricity consumption data of users in the target area, and use cluster analysis to identify typical load time series curves, including residential load, commercial load and industrial load curves.

[0022] Specifically, in a typical implementation, daily load data from multiple users in the target area can be collected throughout the year. Each user's daily electricity consumption records include 24 time points to cover daily periodicity and seasonal variations. The dataset should encompass diverse user types, such as residential households, commercial establishments, and low-voltage industrial enterprises, to ensure the diversity of load curves.

[0023] In terms of data processing methods, this embodiment employs advanced clustering analysis techniques, such as deep convolutional embedding clustering, to automatically identify typical load time-series curves. This method extracts deep features from load data through neural networks and performs unsupervised clustering, effectively overcoming the shortcomings of traditional clustering methods that are insensitive to load fluctuations. The clustering process divides the load data into several categories, ultimately extracting typical curves representing residential, commercial, and industrial loads. These curves reveal the time-series characteristics of different load types, such as the evening peak of residential load, the daytime peak of commercial load, and the stable operation mode of industrial load.

[0024] Step S20: Predict the peak load of each block in the target area based on the load density method, and generate time-series load data of each block in the target area by combining the typical load curve.

[0025] In this embodiment, the application of the load density method is fundamental to this step. This method utilizes land use planning data (such as block area, population density, and building volume ratio) to predict the peak saturation load of each block. Specifically, the peak load is calculated according to the formula: ; In the formula, L i Indicates a street i Peak load, s i Let σi be the area of ​​the block, and let σi be the saturation load density index (load per unit area). η i Floor Area Ratio (FAR) reflects building density. Specifically, the load density method is particularly suitable for newly developed areas because it can be combined with urban development planning to provide long-term, stable load estimates. For example, in a new town with a planned area of ​​25.5 square kilometers, the area can be divided into 308 blocks, and the peak load of each block can be calculated, thus forming the basic framework for the spatial distribution of load.

[0026] Secondly, the generation of time-series load data is achieved by combining peak load with the typical load curve obtained in step S10. This process is accomplished using the following formula: ; In the formula, where, w i ( t (This is a neighborhood) i At 24 time points each day t Forecasted load, λ j ( t) represents the normalized value of the typical load curve for the corresponding load type (j=1, 2, 3 representing residential, commercial, or industrial, respectively). Typical load curves are derived from clusters of actual data, such as... Figure 2 As shown, the diurnal variation patterns of different load types are clearly illustrated, such as the evening peak of residential load and the daytime peak of commercial load. This combination makes the predicted load no longer a static peak, but a dynamic sequence containing temporal fluctuations, thus revealing the complementarity of loads. For example, the daytime peak of commercial load and the nighttime peak of residential load may form a peak-shaving effect.

[0027] In addition, this step involves calculating the average load to assess the block's average daily energy consumption, as shown in the following expression: ; This indicator provides planners with an overview of the basic energy demand of a neighborhood, helping them determine the rationality of substation capacity allocation. Through time-series load data, planners can further analyze load peak-valley characteristics, seasonal variations, and potential renewable energy integration opportunities, such as matching solar power generation with load curves, thereby optimizing the deployment of energy storage systems or reactive power compensation equipment.

[0028] The generation of time-series load data by combining the typical load curve is achieved through the following formula: P i ( t )= P {i,max} * L {type(i)} ( t ); in, P i ( t ) is the first i Each block at any time t The load, P {i,max} For the first predicted by the load density method i Peak load of each block, L {type(i)} ( t ) is the first i The load type of each block at time t The normalized value of the typical load curve.

[0029] Step S30: Establish a substation planning optimization model, with power balance constraints and power supply radius constraints as constraints and minimizing total cost as the objective, and output the substation site selection within the target area.

[0030] In this embodiment, the objective is to minimize the total cost, and the formula for calculating the total cost is as follows: ; ; ; ; In the formula, C sub , C line and C loss These are the substation investment cost, the line investment cost, and the network loss cost, respectively. f ( ) is the substation investment function; N The number of substations; S k For substation k The capacity; J k For substation k A collection of neighborhoods with access to electricity; l ik To the substation k To the neighborhood i The length of the line; α The investment cost per unit length of the line; β This represents the network loss cost coefficient; substation investment costs and line investment costs are converted into annual costs, of which... C RF This represents the annual cost coefficient.

[0031] Among them, the investment cost of substations C sub It is directly related to the substation capacity, and the calculation formula is: ,in C RF This is the annual cost coefficient. f ( S k ) is a capacity of S k The substation investment function, N This refers to the number of substations.

[0032] Among them, the cost of line investment C line Depending on the line length from the substation to each power supply street, the expression is: ,in α It is the investment cost per unit length of line. l ik For substation k To the neighborhood i The length of the line, J k It is a substationk The collection of power supply blocks.

[0033] Among them, network loss cost C loss The formula for representing energy loss during operation is as follows: ,in β This is the network loss cost coefficient. w i Let be the average load of block i. This cost highlights the importance of energy efficiency optimization, especially when load fluctuates significantly.

[0034] In summary, by minimizing the total cost, the model achieves a balance between economy and reliability, avoiding over-investment or inefficient operation.

[0035] In the process of calculating the total cost minimization, the constraints for calculating the total cost include power balance constraints and power supply radius constraints, wherein: The power balance constraint satisfies the following formula: ; ; In the formula, W k ( t ) is the first k The substation at time t The active load it carries, S k For the first k The configuration capacity of the substation, cosφ The power factor is typically taken as 0.9. e ( n k For consideration N The transformer load rate after the N-1 criterion (related to the number of transformers, such as a limit of 65% for a dual-transformer station); this constraint is based on the N-1 criterion to ensure that the system can still operate reliably under a single fault.

[0036] The power supply radius constraint satisfies the following formula: ; In the formula, R k For substation k The power supply distance (defined as the distance from the station to the farthest point in the power supply area); R c The power supply radius of the substation is also the maximum allowable value (usually 3 kilometers for urban power distribution networks).

[0037] Step S40: An improved genetic algorithm is used to optimize the substation site selection, and the optimal location of the substation within the target area is selected from the candidate site set.

[0038] It should be noted that the substation site selection problem is essentially a complex combinatorial optimization problem, involving a large number of candidate sites and various constraints. If an exhaustive search method is used, the computational cost increases exponentially with the problem size, making it impractical. While genetic algorithms, as a heuristic optimization method, avoid local optima through population evolution, their standard version may produce invalid solutions in substation planning (such as duplicate site selection or violations of site conditions). Therefore, this method makes targeted improvements to the genetic algorithm, including chromosome encoding optimization and constraint handling mechanisms, to enhance search efficiency and feasibility.

[0039] In this embodiment, the optimization of substation site selection using an improved genetic algorithm includes: The maximum number of substations is determined based on the total load of the target area and the maximum allowable capacity of a single substation. The candidate sites are binary-coded using the maximum number of substations as the chromosome length. A single-point crossover operator and a mutation operator are used for population iteration, and duplicate site genes in the chromosome are removed during decoding to obtain an effective substation site selection scheme.

[0040] Specifically, the implementation of improved genetic algorithms includes the following key steps: First, based on the total load of the target area and the minimum capacity of a single substation, calculate the maximum number of substations required. The calculation expression is as follows: ; In the formula, N max This represents the maximum number of substations required. S min ε represents the minimum capacity of a given substation; ε is the capacity-to-load ratio, which is a measure of the ratio of substation capacity to required power.

[0041] Secondly, the candidate sites are numbered from 1 to M natural numbers, and each chromosome is composed of... N max A chromosome consists of site numbers, representing a potential substation site selection scheme. For example, chromosome "3-5-6" indicates the selection of candidate sites numbered 3, 5, and 6. The initial population is composed of multiple randomly generated chromosomes to ensure diversity.

[0042] Then, genetic operators are designed, including: Selection operation: Based on the fitness function (i.e., the reciprocal of the objective function Ctotal), select high-quality individuals and prioritize retaining the solutions with lower costs.

[0043] Crossover operation: A single-point crossover operator is used to randomly select the crossover point of two parent chromosomes, exchange some gene sequences, and generate new offspring. For example, parent chromosomes "3-5-6" and "2-4-7" may generate offspring "3-4-6".

[0044] Mutation operation: Randomly change the number of a site in the chromosome with a small probability to introduce new features and avoid premature convergence.

[0045] Finally, in each iteration, the chromosome is decoded to eliminate duplicate site numbers (e.g., "3-3-5-6" is simplified to "3-5-6"), ensuring that the location of each substation is unique. Simultaneously, the scheme is checked to ensure it meets constraints such as power supply radius; individuals violating these constraints are penalized with a penalty function to reduce their fitness.

[0046] The improved genetic algorithm shown in this embodiment has significant advantages over traditional methods: Crossover and mutation are used to avoid local optima and ensure that the solution is close to the global optimum; chromosome encoding simplifies the search space and the feasibility verification mechanism reduces invalid calculations, resulting in high computational efficiency.

[0047] Step S50: Divide the power supply range of each substation within the target area based on Thiessen polygons, and determine the substation capacity according to the total peak load within the power supply range.

[0048] This step is the final decision-making stage in the substation planning methodology, aiming to transform the optimized site selection scheme into specific and implementable power supply zones and capacity configurations. By applying Thiessen polygons, a mature geometric tool, the power supply area is scientifically and automatically divided, and the substation capacity is accurately determined in conjunction with load analysis, ensuring the spatial rationality and technical and economic efficiency of the planning scheme.

[0049] The Thiessen polygon, also known as the Voronoi diagram, is a classic method for regional division based on spatial distance. Its core principle is that for a given set of points on a plane (i.e., the generating points, which in this method are the selected substation sites), the entire plane is divided into several non-overlapping polygonal regions. Each polygonal region contains one and only one generating point, and the distance from any point within that region to its generating point is less than its distance to any other generating point. This characteristic perfectly matches the engineering requirements for substation power supply range division, namely, each geographical area should be powered by the nearest substation, which helps minimize line investment and network losses. The mathematical definition of the Thiessen polygon is shown in the formula: ; In the formula, p i It is the first i The coordinates of the substation xIt is any point within the polygonal region. d ( () is the Euclidean distance.

[0050] In practical operation, once the optimal set of substation locations is determined using the genetic algorithm in step S40, a Thiessen polygon map of the entire target area is automatically constructed using the geographical coordinates of these stations as generation points. The Thiessen polygon (V-polygon) corresponding to each substation represents its theoretical power supply range. The power supply radius of the substation... R k It is defined as the maximum distance from the substation site to each vertex of its V-polygon. This division method ensures the uniqueness and spatial rationality of the power supply range.

[0051] like Figure 3 The Thiessen polygon diagram shown is composed of polygon A1. A2 A3 A4 A5 The area enclosed by A6 is the power supply range of substation S1, and its power supply radius is [missing information]. max { S 1 A i This classification is automatic and objective, avoiding the subjectivity of human-made classifications.

[0052] After clearly defining the power supply area, the capacity of each substation is further determined. First, it is necessary to calculate the total load time-series data for all blocks falling within the power supply area of ​​each substation. W k ( t ), and find its maximum value. max t { W k ( t The peak load of the power supply area is defined as follows: Then, the minimum required transformer capacity is calculated using the following formula. P k : ; In the formula, e ( n k The value is the transformer load factor after considering the N-1 criterion (e.g., 65% for a dual transformer), and cosφ is the power factor (e.g., 0.9). Finally, based on the calculated... P k The value is used to select the closest standard transformer capacity level as the final configuration capacity for the substation. This ensures that the substation's capacity configuration is neither excessive nor insufficient while meeting power supply reliability requirements, thus achieving optimal equipment utilization.

[0053] Compared with existing technologies, the substation planning method shown in this embodiment has the following advantages: The method described in this embodiment achieves a significant improvement in overall benefits in substation planning by systematically considering the complementarity of load characteristics. First, the method accurately identifies the temporal characteristic curves of different types of loads using deep convolutional embedding clustering technology, fundamentally changing the load estimation distortion problem caused by traditional reliance on static simultaneity coefficients. Then, it combines spatiotemporal load prediction with a multi-objective optimization model, uses an improved genetic algorithm for substation site selection optimization, and scientifically divides the power supply area using Thiessen polygons, ultimately forming an overall planning scheme. This technical approach allows the planning results to fully utilize the peak-valley complementarity characteristics of loads in different regions, not only simplifying the calculation process but also improving search efficiency, enabling rapid decision-making in complex planning scenarios.

[0054] Example 2 Please see Figure 4 A second embodiment of the present invention provides a substation planning system applied to the method described in the above embodiments, the system comprising: Data analysis module 10 is used to acquire historical electricity consumption data of users in the target area and to identify typical load time-series curves, including residential load, commercial load and industrial load curves, using cluster analysis. The load forecasting module 20 is used to predict the peak load of each block in the target area based on the load density method, and generate time-series load data of each block in the target area in combination with the typical load curve. The location output module 30 is used to establish a substation planning optimization model, with power balance constraints and power supply radius constraints as constraints and minimizing total cost as the objective, and output the substation location within the target area; The site selection optimization module 40 is used to optimize the substation site selection using an improved genetic algorithm, and to select the optimal location of the substation within the target area from the candidate site set; The capacity configuration module 50 is used to divide the power supply range of each substation within the target area based on Thiessen polygons, and to determine the substation capacity according to the total peak load within the power supply range.

[0055] Compared with existing technologies, the substation planning system shown in this embodiment has the following advantages: The system described in this embodiment achieves significant improvements in overall efficiency in substation planning by systematically considering the complementarity of load characteristics. First, the system accurately identifies the temporal characteristic curves of different types of loads using deep convolutional embedding clustering technology, fundamentally changing the load estimation distortion problem caused by traditional reliance on static simultaneity coefficients. Then, it combines spatiotemporal load prediction with a multi-objective optimization model, uses an improved genetic algorithm for substation site selection optimization, and scientifically divides the power supply area using Thiessen polygons, ultimately forming an overall planning scheme. This technical approach allows the planning results to fully utilize the peak-valley complementarity characteristics of loads in different regions, simplifying the calculation process, improving search efficiency, and enabling rapid decision-making in complex planning scenarios.

[0056] Example 3 A third embodiment of the present invention provides a readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the steps of the methods described in the above embodiments.

[0057] Example 4 A fourth embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method described in the above embodiments.

[0058] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0059] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A substation planning method, characterized in that, The method includes: Obtain historical electricity consumption data of users within the target area, and use cluster analysis to identify typical load time-series curves, including residential load, commercial load, and industrial load curves; The peak load of each block in the target area is predicted based on the load density method, and the time-series load data of each block in the target area is generated by combining the typical load curve. A substation planning optimization model is established, with power balance constraints and power supply radius constraints as constraints and minimizing total cost as the objective, and the substation site selection within the target area is output. An improved genetic algorithm is used to optimize the substation site selection, and the optimal location of the substation within the target area is selected from the candidate site set; The power supply range of each substation within the target area is divided based on Thiessen polygons, and the substation capacity is determined based on the total peak load within the power supply range.

2. The substation planning method according to claim 1, characterized in that, The method of using cluster analysis to identify typical load time series curves is as follows: The user's historical electricity consumption data is processed using a deep convolutional embedding clustering method. The user's historical electricity consumption data includes load data at 24 time points every day of the year, in order to identify typical daily load curves for residential load, commercial load, and low-voltage industrial load.

3. The substation planning method according to claim 2, characterized in that, The generation of time-series load data by combining the typical load curve is achieved through the following formula: P i ( t )= P {i,max} * L {type(i)} ( t )； in, P i ( t ) is the first i Each block at any time t The load, P {i,max} For the first predicted by the load density method i Peak load of each block, L {type(i)} ( t ) is the first i The load type of each block at time t The normalized value of the typical load curve.

4. The substation planning method according to claim 1, characterized in that, The objective is to minimize the total cost, and the formula for calculating the total cost is as follows: ； ； ； ； In the formula, C sub , C line and C loss These are the substation investment cost, the line investment cost, and the network loss cost, respectively. f ( ) is the substation investment function; N The number of substations; S k For substation k The capacity; J k For substation k A collection of neighborhoods with access to electricity; l ik To the substation k To the neighborhood i The length of the line; α The investment cost per unit length of the line; β The network loss cost coefficient is used; the substation investment cost and line investment cost are converted into annual cost, where CRF is the annual cost coefficient.

5. The substation planning method according to claim 4, characterized in that, The constraints for calculating the total cost include power balance constraints and power supply radius constraints, wherein: The power balance constraint satisfies the following formula: ； ； In the formula, W k ( t ) is the first k The substation at time t The active load it carries, S k For the first k The configuration capacity of the substation, cos φ For power factor, e ( n k For consideration N Transformer load rate after the -1 criterion; The power supply radius constraint satisfies the following formula: ； In the formula, R k For substation k The power supply distance; R c This refers to the power supply radius of the substation.

6. The substation planning method according to claim 1, characterized in that, The method of optimizing substation site selection using an improved genetic algorithm includes: The maximum number of substations is determined based on the total load of the target area and the maximum allowable capacity of a single substation. The candidate sites are binary-coded using the maximum number of substations as the chromosome length. A single-point crossover operator and a mutation operator are used for population iteration, and duplicate site genes in the chromosome are removed during decoding to obtain an effective substation site selection scheme. The expression for optimizing substation site selection using the improved genetic algorithm is as follows: ； In the formula, N max This represents the maximum number of substations required. S min ε represents the minimum capacity of a given substation; ε is the capacity-to-load ratio, which is a measure of the ratio of substation capacity to required power.

7. The substation planning method according to claim 1, characterized in that, The division of the power supply range of each substation based on Thiessen polygons is specifically as follows: Using the selected substation site coordinates as the generation point, a Thiessen polygon is constructed, and the area within each polygon is the power supply range of the substation; wherein, the power supply radius is defined as the maximum distance from the substation site to each vertex of its Thiessen polygon. The determination of the substation capacity specifically involves: Calculate the total peak load within each power supply range, and select the closest standard transformer capacity level as the configuration capacity of the substation based on the total peak load.

8. A substation planning system, characterized in that, The system, applicable to the method of any one of claims 1-7, comprises: The data analysis module is used to acquire historical electricity consumption data of users in the target area and to identify typical load time-series curves, including residential load, commercial load and industrial load curves, using cluster analysis. The load forecasting module is used to predict the peak load of each block in the target area based on the load density method, and generate time-series load data of each block in the target area by combining the typical load curve. The location selection output module is used to establish a substation planning optimization model, with power balance constraints and power supply radius constraints as constraints and minimizing total cost as the objective, and output the substation location within the target area; The site selection optimization module is used to optimize the substation site selection using an improved genetic algorithm, and to select the optimal location of the substation within the target area from the candidate site set; The capacity configuration module is used to divide the power supply range of each substation within the target area based on Thiessen polygons, and to determine the substation capacity according to the total peak load within the power supply range.

9. A readable storage medium having computer instructions stored thereon, characterized in that, When executed by the processor, this instruction implements the steps of the method as described in any one of claims 1-7.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method as described in any one of claims 1-7.

Images