Medium-voltage grid segment contact and distributed power consumption optimization method based on big data clustering

Abstract

The invention discloses a medium-voltage grid segment contact and distributed power consumption optimization method based on big data clustering, and relates to the technical field of power grid optimization, and the method comprises the steps: S1, power distribution network data collection; s2, carrying out DBSCAN clustering analysis; s3, carrying out segmented contact optimization on the medium-voltage net rack; s4, distributed power supply consumption optimization is carried out; and S5, evaluating and optimizing a result. According to the medium-voltage grid segment contact and distributed power consumption optimization method based on big data clustering, global information provided by DBSCAN clustering can be fully utilized, overall optimization of a power system is considered from a global perspective, the DBSCAN clustering can identify clusters in any shapes and also achieve the purpose of efficiently extracting the global information, and on the basis of a DBSCAN clustering result, the overall optimization of the power system is optimized. Medium-voltage grid segment contact optimization and distributed power consumption optimization are carried out, the power distribution network optimization decision process is effectively simplified, the optimization efficiency is improved, and the method is helpful for adapting to rapidly changing power system requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid optimization, and in particular to a medium voltage grid segmentation connection and distributed power supply consumption optimization method based on big data clustering. Background Art

[0002] The distribution network refers to a power grid that receives electric energy from the transmission network or regional power plants, and distributes it locally or step by step according to voltage to various types of users through distribution facilities. It is the part of the power system that is directly connected to the user and is responsible for transmitting electric energy from the transmission system to the user end. Medium-voltage grid segmentation and distributed power consumption are two other important concepts in the power system. Among them, medium-voltage grid segmentation is an important method in the planning and operation of the distribution network, aiming to improve the reliability and flexibility of the power grid. Distributed power consumption refers to the effective connection and consumption of electric energy generated by distributed power sources (such as solar photovoltaic, wind power, etc.) into the power grid to achieve full utilization of energy and stable operation of the power grid.

[0003] For example, the patent document with publication number CN116845896A discloses a method for optimizing medium-voltage grid segment connections taking into account distributed power sources. The method for optimizing medium-voltage grid segment connections proposed in this invention overcomes the limitation of traditional guideline methods that use load density and power supply radius indicators to segment lines, solves the practical problem of widespread access to distributed power sources, realizes refined segment division based on big data clustering methods, considers power supply reliability and distributed power source absorption requirements for line connection settings, and can provide reasonable suggestions for medium-voltage grid optimization and transformation.

[0004] However, existing optimization methods similar to the above applications often make decisions based on local data, making it difficult to consider the overall optimization of the power system from a global perspective, and failing to fully utilize the global information provided by big data clustering technology, resulting in limitations in the optimization scheme and failure to achieve the optimal effect. At the same time, due to the lack of big data support, the optimization process requires a lot of time and resources to collect and process data, which may make the decision-making process cumbersome and inefficient, making it difficult to adapt to the rapidly changing needs of the power system, and is not conducive to the planning and operation of the distribution network.

[0005] Therefore, it is urgent to improve this shortcoming. The present invention studies and improves the existing technology and its shortcomings, and provides a medium-voltage grid segmented connection and distributed power supply consumption optimization method based on big data clustering. Summary of the invention

[0006] The purpose of the present invention is to provide a medium voltage grid segment connection and distributed power supply consumption optimization method based on big data clustering to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: a medium voltage grid segmented connection and distributed power consumption optimization method based on big data clustering, comprising the following steps:

[0008] S1. Distribution network data collection:

[0009] Collect relevant data from the distribution network, including grid structure, load distribution, location and capacity of distributed generation (DG), and perform necessary preprocessing on the collected raw data;

[0010] S2. DBSCAN cluster analysis:

[0011] Parameter selection: The two main parameters of DBSCAN are determined by domain knowledge and K-distance graph methods, namely, the maximum distance ε between two points that are considered neighbors and the minimum number of points MinPts required to form a dense area. Specifically, a K-distance graph is drawn and the appropriate ε value is selected by observing the inflection points in the graph. The selection of MinPts is based on the dimension of the data or empirical rules.

[0012] Construct neighborhood: For each point in the data set, construct its ε neighborhood, that is, all points in the circular area with a radius of ε around the point, which is used to determine the local density of the point;

[0013] Density accessibility judgment: judge whether each point is a core point, boundary point or noise point;

[0014] Cluster formation: Starting from any unvisited core point, start expanding the cluster until all core points are visited and classified into the corresponding cluster;

[0015] Noise point processing: All points that are not classified into any cluster in the cluster formation process are marked as noise points, that is, outliers or abnormal values ​​in the data set;

[0016] S3, Medium Voltage Grid Sectional Connection Optimization:

[0017] Intra-cluster contact analysis: Based on the DBSCAN clustering results, the grid structure and load distribution within each cluster are analyzed to determine the contact points and contact lines within the cluster;

[0018] Inter-cluster communication optimization: Based on the global information, i.e. the structure and load distribution of the entire medium-voltage grid, the communication requirements between different clusters are analyzed to determine the optimal communication path and method. After determining the communication path and method, the communication configuration is optimized, including selecting appropriate contact switches, adjusting the capacity and direction of the contact lines, etc.

[0019] S4. Distributed power consumption optimization:

[0020] DG location optimization: According to the clustering results (including key areas such as load-intensive areas and grid weak points, the clustering results help determine the potential installation location of DG to better meet load demand and optimize the grid structure) and the grid structure (including the topology of the grid, the location and capacity of substations, the direction and connection relationship of feeders, etc., the grid structure helps determine the bottleneck area and potential upgrade points of the grid, so as to optimize and adjust when installing DG), determine the best installation location of distributed generation to maximize its absorption capacity and the economy of the grid;

[0021] DG capacity planning: Plan the capacity of distributed power sources based on load distribution and DG absorption capacity to ensure stable operation and efficient utilization of the power grid;

[0022] S5. Result evaluation and optimization:

[0023] Select evaluation indicators, such as grid reliability, economy, and consumption rate of distributed generation, to evaluate the optimization results. Based on the evaluation results, adjust and optimize clustering parameters, contact methods, DG location and capacity, etc. to improve the overall effect.

[0024] Furthermore, in step S1, the relevant data is collected as follows:

[0025] Grid structure (reflecting the physical layout and connection mode of the distribution network): query and obtain grid structure data with the help of the Geographic Information System (GIS) platform, or understand the structure and layout of the grid through drawings of the distribution network, such as line maps, substation wiring diagrams, geographic wiring diagrams, etc., and obtain key information such as the grid topology, line direction, substation location and capacity;

[0026] Load distribution (reflecting the load conditions and change trends in various areas of the distribution network): Install load monitoring equipment, such as smart meters and load sensors, at key nodes and user ends of the distribution network to monitor and record load data in real time, including load size, load type, load change, etc.

[0027] Distributed power sources (reflecting the access status, location and capacity of distributed power sources in the distribution network): By establishing a cooperative relationship with the operators or owners of distributed power sources, detailed information on distributed power sources can be obtained, including the type, capacity, access point location, etc. of distributed power sources. This is an important source of DG data. Monitoring equipment, such as power quality monitoring devices and power factor monitoring devices, are installed at the access points of distributed power sources to obtain key parameters such as output power and power quality of distributed power sources in real time, providing strong support for the collection of DG data.

[0028] Furthermore, in step S1, the preprocessing includes data cleaning, data integration and data normalization, which are specifically as follows:

[0029] Data cleaning: remove duplicate and abnormal data, and process missing data. Specifically, for missing grid structure data, use interpolation, regression or data based on adjacent nodes to estimate and fill in the missing data; for missing load data, use time series analysis, interpolation or prediction models based on historical data to fill in the missing data.

[0030] Data integration: Integrate data from different data sources to form a complete data set;

[0031] Data normalization: Min-Max standardization is used to normalize data of different dimensions and dimensional units so that each indicator is at the same order of magnitude, which is convenient for subsequent cluster analysis.

[0032] Furthermore, in step S2, the core points, boundary points and noise points are defined as follows:

[0033] Core point: There are at least MinPts points in the ε neighborhood;

[0034] Boundary point: in the ε neighborhood of a core point but not a core point itself;

[0035] Noise point: It is neither a core point nor belongs to the neighborhood of any core point.

[0036] Furthermore, in step S2, cluster formation specifically includes the following sub-steps:

[0037] Starting from any unvisited core point, a new cluster is formed;

[0038] Add the core point and the points in its neighborhood (including the core point and the boundary points) to the cluster;

[0039] Recursively expand the newly added points. If there are other unvisited core points in the neighborhood of a point, these core points and the points in their neighborhood are also added to the cluster.

[0040] The above process is repeated until all core points are visited and classified into corresponding clusters.

[0041] Furthermore, in step S3, the specific operation of the intra-cluster contact analysis is as follows: DBSCAN clustering divides the medium-voltage grid into multiple clusters, and the power grid structure and load distribution within each cluster have certain similarities. Within each cluster, potential contact points are determined based on the complexity of the power grid structure and the uniformity of the load distribution. The contact points are located at key positions of the power grid structure, such as branch points, intersections, etc., and then the contact lines within the cluster are determined based on the location of the contact points and the load demand.

[0042] Furthermore, in step S3, the communication requirements include load transfer requirements (referring to the ability to transfer load to other clusters when a fault occurs in a cluster), reliability improvement requirements (referring to improving the overall reliability of the power grid by increasing inter-cluster communication), and flexibility enhancement requirements (referring to adapting to future load changes and power grid development needs by optimizing inter-cluster communication), etc. Specifically, the benefits of each communication requirement are evaluated, including economic benefits (such as investment costs, operation and maintenance costs, etc.) and social benefits (such as improving power supply reliability, reducing power outage losses, etc.), and the benefits of different communication schemes are compared to determine the optimal communication requirement;

[0043] The communication requirements provide a basis for the selection of communication paths and communication methods, and the communication paths give priority to paths that can maximize load transfer capacity and improve grid reliability. The range of communication methods includes but is not limited to hand-in-hand wiring (connecting two feeders from the same substation or different busbars from different substations through a connecting switch to form a ring network, but open-loop operation during normal operation), multi-segment connection wiring (a wiring mode defined according to the number of segments and connections of the distribution network line, such as three sections and three connections, and the line segments generally use circuit breakers to reduce the scope of faults and planned power outages), and ring network wiring (a local area network (LAN) architecture with optical fiber as the main transmission medium).

[0044] Furthermore, in step S4, when determining the DG location, priority is given to load-intensive areas, grid weak points, and areas with good absorption capacity, and it is ensured that the installation location of the DG will not have a negative impact on grid stability.

[0045] Furthermore, in step S4, the capacity planning of DG specifically includes the following sub-steps:

[0046] Load forecasting: forecast future loads, including load growth trends, seasonal changes, daily changes, etc., to determine DG capacity requirements;

[0047] DG absorption capacity assessment: Evaluate the grid's absorption capacity for DG, including the grid's capacity, stability, and regulation capabilities, to determine the DG capacity range that the grid can accommodate, thereby avoiding overload and instability;

[0048] DG capacity planning: According to the load forecast and DG absorption capacity assessment results, the DG capacity is planned. The planning principle is that the DG capacity can meet the load demand without having a negative impact on the stability of the power grid. At the same time, the redundant capacity of DG and the configuration of backup power supply are considered to improve the reliability and resilience of the power grid.

[0049] The present invention provides a medium voltage grid segmentation connection and distributed power supply consumption optimization method based on big data clustering, which has the following beneficial effects:

[0050] The present invention can make full use of the global information provided by DBSCAN clustering and consider the overall optimization of the power system from a global perspective. DBSCAN clustering forms clusters by judging the density accessibility of points and excludes noise points. It can not only identify clusters of any shape, but also achieve the purpose of efficiently extracting global information. Based on the DBSCAN clustering results, medium-voltage grid segmented connection optimization and distributed power supply consumption optimization are carried out, which effectively simplifies the distribution network optimization decision-making process, improves the optimization efficiency, and helps to adapt to the rapidly changing power system needs. At the same time, it also avoids the limitations of decision-making based on local data, which is conducive to achieving the best effect and realizing the optimization of the power grid structure and the efficient use of distributed power sources. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A schematic diagram of the steps of a medium voltage grid segmented connection and distributed power consumption optimization method based on big data clustering according to the present invention;

[0052] Figure 2 The present invention is a schematic diagram of the DBSCAN clustering analysis process of a medium-voltage grid segmented connection and distributed power supply consumption optimization method based on big data clustering. DETAILED DESCRIPTION

[0053] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0054] like Figure 1-Figure 2 As shown, a medium voltage grid segmented connection and distributed power consumption optimization method based on big data clustering includes the following steps:

[0055] S1. Distribution network data collection:

[0056] Collect relevant data from the distribution network, including grid structure, load distribution, location and capacity of distributed generation (DG), and perform necessary preprocessing on the collected raw data;

[0057] In this embodiment, the relevant data is collected as follows:

[0058] Grid structure (reflecting the physical layout and connection mode of the distribution network): query and obtain grid structure data with the help of the Geographic Information System (GIS) platform, or understand the structure and layout of the grid through distribution network drawings, such as line maps, substation wiring diagrams, geographic wiring diagrams, etc., and obtain key information such as the grid topology, line direction, substation location and capacity;

[0059] Load distribution (reflecting the load conditions and change trends in various areas of the distribution network): Install load monitoring equipment, such as smart meters and load sensors, at key nodes and user ends of the distribution network to monitor and record load data in real time, including load size, load type, load change, etc.

[0060] Distributed power sources (reflecting the access status, location, capacity and other information of distributed power sources in the distribution network): By establishing a cooperative relationship with the operators or owners of distributed power sources, detailed information of distributed power sources can be obtained, including the type, capacity, access point location, etc. of distributed power sources. This is an important source of DG data. Monitoring equipment, such as power quality monitoring devices and power factor monitoring devices, are installed at the access points of distributed power sources to obtain key parameters such as output power and power quality of distributed power sources in real time, providing strong support for the collection of DG data.

[0061] In this embodiment, the preprocessing operation is specifically as follows:

[0062] Data cleaning: remove duplicate and abnormal data, and process missing data. Specifically, for missing grid structure data, use interpolation, regression or data based on adjacent nodes to estimate and fill in the missing data; for missing load data, use time series analysis, interpolation or prediction models based on historical data to fill in the missing data.

[0063] Data integration: Integrate data from different data sources to form a complete data set;

[0064] Data normalization: Min-Max standardization is used to normalize data of different dimensions and dimensional units so that each indicator is at the same order of magnitude, which is convenient for subsequent cluster analysis.

[0065] S2. DBSCAN cluster analysis:

[0066] Parameter selection: two main parameters of DBSCAN are determined by domain knowledge and K-distance graph methods, namely, the maximum distance ε between two points that are considered neighbors and the minimum number of points MinPts required to form a dense area. In this embodiment, a K-distance graph is drawn, and the appropriate ε value is selected by observing the inflection points in the graph. The selection of MinPts is based on the dimension of the data or empirical rules.

[0067] Construct neighborhood: For each point in the data set, construct its ε neighborhood, that is, all points in the circular area with a radius of ε around the point, which is used to determine the local density of the point;

[0068] Density accessibility judgment: Determine whether each point is a core point (there are at least MinPts points in the ε neighborhood), a boundary point (in the ε neighborhood of the core point but not a core point itself), or a noise point (neither a core point nor belonging to the neighborhood of any core point);

[0069] Cluster formation: Starting from any unvisited core point, a new cluster is formed; the core point and the points in its neighborhood (including core points and boundary points) are added to the cluster; the newly added points are recursively expanded, and if there are other unvisited core points in the neighborhood of a point, these core points and the points in their neighborhood are also added to the cluster; the above process is repeated until all core points are visited and classified into the corresponding clusters;

[0070] Noise point processing: All points that are not classified into any cluster in the clustering process are marked as noise points, that is, outliers or abnormal values ​​in the data set.

[0071] S3, Medium Voltage Grid Sectional Connection Optimization:

[0072] Intra-cluster contact analysis: Based on the DBSCAN clustering results, the grid structure and load distribution within each cluster are analyzed to determine the contact points and contact lines within the cluster;

[0073] Inter-cluster communication optimization: Based on the global information, i.e. the structure and load distribution of the entire medium-voltage grid, the communication requirements between different clusters are analyzed to determine the optimal communication path and method. After determining the communication path and method, the communication configuration is optimized, including selecting appropriate contact switches, adjusting the capacity and direction of the contact lines, etc.

[0074] In this embodiment, the specific operation of the intra-cluster contact analysis is as follows: DBSCAN clustering divides the medium voltage grid into multiple clusters, and the power grid structure and load distribution in each cluster have certain similarities. In each cluster, according to the complexity of the power grid structure and the uniformity of the load distribution, potential contact points are determined. The contact points are located at key positions of the power grid structure, such as branch points, intersections, etc. Then, according to the location of the contact points and the load demand, the contact lines in the cluster are determined;

[0075] In this embodiment, the communication requirements include load transfer requirements (referring to the ability to transfer load to other clusters when a fault occurs in a cluster), reliability improvement requirements (referring to improving the overall reliability of the power grid by increasing inter-cluster communication), and flexibility enhancement requirements (referring to adapting to future load changes and power grid development needs by optimizing inter-cluster communication), etc. Specifically, the benefits of each communication requirement are evaluated, including economic benefits (such as investment costs, operation and maintenance costs, etc.) and social benefits (such as improving power supply reliability, reducing power outage losses, etc.), and the benefits of different communication solutions are compared to determine the optimal communication requirement;

[0076] The contact demand provides a basis for the selection of contact paths and contact methods, and the contact paths give priority to paths that can maximize the load transfer capacity and improve the reliability of the power grid. The range of contact methods includes but is not limited to hand-in-hand connection (connecting two feeders from different buses of the same substation or different substations through a contact switch to form a ring network, but open-loop operation during normal operation), multi-segment contact connection (a connection mode defined according to the number of segments and contacts of the distribution network line, such as three-segment three-contact, and the line segments generally use circuit breakers to reduce the scope of faults and planned power outages), and ring network connection (a local area network (LAN) architecture with optical fiber as the main transmission medium);

[0077] In this embodiment, after the connection setting is completed, it is necessary to check the reverse load rate and line penetration rate of each segment. The reverse load rate is used to evaluate the absorption capacity of distributed power sources, and the line penetration rate is used to evaluate the load transfer capacity of the power grid. By checking the reverse load rate and line penetration rate, the reliability and flexibility of the power grid can be ensured after the distributed power sources are connected.

[0078] S4. Distributed power consumption optimization:

[0079] DG location optimization: According to the clustering results (including key areas such as load-intensive areas and grid weak points, the clustering results help determine the potential installation location of DG to better meet load demand and optimize the grid structure) and the grid structure (including the topology of the grid, the location and capacity of substations, the direction and connection relationship of feeders, etc., the grid structure helps to determine the bottleneck area and potential upgrade points of the grid, so as to optimize and adjust when installing DG), determine the best installation location of distributed generation to maximize its absorption capacity and the economy of the grid. When determining the DG location, priority is given to load-intensive areas, grid weak points and areas with good absorption capacity, and ensure that the installation location of DG will not have a negative impact on the stability of the grid;

[0080] DG capacity planning: According to the load distribution and the absorption capacity of DG, the capacity of distributed generation is planned to ensure the stable operation and efficient utilization of the power grid; it includes the following sub-steps:

[0081] Load forecasting: forecast future loads, including load growth trends, seasonal changes, daily changes, etc., to determine DG capacity requirements;

[0082] DG absorption capacity assessment: Evaluate the grid's absorption capacity for DG, including the grid's capacity, stability, and regulation capabilities, to determine the DG capacity range that the grid can accommodate, thereby avoiding overload and instability;

[0083] DG capacity planning: According to the load forecast and DG absorption capacity assessment results, the DG capacity is planned. The planning principle is that the DG capacity can meet the load demand without having a negative impact on the stability of the power grid. At the same time, the redundant capacity of DG and the configuration of backup power supply are considered to improve the reliability and resilience of the power grid.

[0084] S5. Result evaluation and optimization:

[0085] Select evaluation indicators, such as grid reliability, economy, and consumption rate of distributed generation, to evaluate the optimization results. Based on the evaluation results, adjust and optimize clustering parameters, contact methods, DG location and capacity, etc. to improve the overall effect.

[0086] The embodiments of the present invention are given for the purpose of illustration and description, and are not intended to be exhaustive or to limit the invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. A medium voltage grid segmentation connection and distributed power supply consumption optimization method based on big data clustering, characterized in that: The following steps are involved: S1. Distribution network data collection: Collect relevant data from the distribution network, including grid structure, load distribution, location and capacity of distributed power sources, and perform necessary preprocessing on the collected raw data; S2. DBSCAN cluster analysis: Parameter selection: The two main parameters of DBSCAN are determined by domain knowledge and K distance graph method, namely, the maximum distance ε between two points considered as neighbors and the minimum number of points MinPts required to form a dense area; Construct neighborhood: For each point in the data set, construct its ε neighborhood, that is, all points in the circular area with a radius of ε around the point; Density accessibility judgment: judge whether each point is a core point, boundary point or noise point; Cluster formation: Starting from any unvisited core point, start expanding the cluster until all core points are visited and classified into the corresponding cluster; Noise point processing: all points that are not classified into any cluster are marked as noise points; S3, Medium Voltage Grid Sectional Connection Optimization: Intra-cluster contact analysis: Based on the DBSCAN clustering results, the grid structure and load distribution within each cluster are analyzed to determine the contact points and contact lines within the cluster; Inter-cluster communication optimization: Based on the global information, i.e. the structure and load distribution of the entire medium voltage grid, the communication requirements between different clusters are analyzed to determine the optimal communication path and communication method. After the communication path and communication method are determined, the communication configuration is optimized. S4. Distributed power consumption optimization: DG location optimization: Determine the best installation location of distributed generation based on clustering results and grid structure to maximize its absorption capacity and the economy of the grid; DG capacity planning: According to the load distribution and DG absorption capacity, the capacity of distributed generation is planned to ensure the stable operation and efficient utilization of the power grid.

2. According to the method of optimizing medium voltage grid segmentation and distributed power consumption based on big data clustering in claim 1, it is characterized in that: The method further comprises the steps of: S5. Result evaluation and optimization: Select evaluation indicators, evaluate the optimization results, and adjust and optimize clustering parameters, contact methods, DG locations and capacities based on the evaluation results.

3. According to the method of optimizing medium voltage grid segmentation and distributed power consumption based on big data clustering in claim 1, it is characterized in that: In step S1, the relevant data is collected as follows: Grid structure: query and obtain grid structure data with the help of geographic information system platform, or understand the structure and layout of the grid through distribution network drawings, and obtain key information of the grid; Load distribution: Install load monitoring equipment at key nodes and user ends of the distribution network to monitor and record load data in real time, including load size, load type, and load changes; Distributed power generation: By establishing a cooperative relationship with the operator or owner of the distributed power generation, detailed information on the distributed power generation can be obtained, including the type, capacity, and access point location of the distributed power generation. Monitoring equipment can also be installed at the access point of the distributed power generation to obtain key parameters of the distributed power generation in real time.

4. According to a method for optimizing medium voltage grid segmentation and distributed power consumption based on big data clustering according to claim 1, it is characterized in that: In step S1, the preprocessing includes data cleaning, data integration and data normalization, which are specifically as follows: Data cleaning: remove duplicate and abnormal data, and process missing data. Specifically, for missing grid structure data, use interpolation, regression or data based on adjacent nodes to estimate and fill in the missing data; for missing load data, use time series analysis, interpolation or prediction models based on historical data to fill in the missing data. Data integration: Integrate data from different data sources to form a complete data set; Data normalization: Min-Max standardization is used to normalize data of different dimensions and dimensional units so that each indicator is at the same order of magnitude.

5. The method for optimizing medium voltage grid segmentation and distributed power consumption based on big data clustering according to claim 1 is characterized in that: In step S2, the core points, boundary points and noise points are defined as follows: Core point: There are at least MinPts points in the ε neighborhood; Boundary point: in the ε neighborhood of a core point but not a core point itself; Noise point: It is neither a core point nor belongs to the neighborhood of any core point.

6. The method for optimizing medium voltage grid segmentation and distributed power consumption based on big data clustering according to claim 1 is characterized in that: In step S2, cluster formation specifically includes the following sub-steps: Starting from any unvisited core point, a new cluster is formed; Add the core point and the points in its neighborhood to the cluster; Recursively expand the newly added points. If there are other unvisited core points in the neighborhood of a point, these core points and the points in their neighborhood are also added to the cluster. The above process is repeated until all core points are visited and classified into corresponding clusters.

7. The method for optimizing medium voltage grid segmentation and distributed power consumption based on big data clustering according to claim 1 is characterized in that: In step S3, the specific operation of the intra-cluster contact analysis is as follows: DBSCAN clustering divides the medium voltage grid into multiple clusters, and in each cluster, potential contact points are determined according to the complexity of the grid structure and the uniformity of the load distribution. The contact points are located at key positions of the grid structure, and then the contact lines within the cluster are determined according to the positions of the contact points and the load requirements.

8. The method for optimizing medium voltage grid segmentation and distributed power consumption based on big data clustering according to claim 1 is characterized in that: In step S3, the communication demand includes load transfer demand, reliability improvement demand and flexibility enhancement demand. Specifically, the optimal communication demand is determined by evaluating the benefits of each communication demand, including economic benefits and social benefits, and comparing the benefits of different communication solutions. The communication requirements provide a basis for the selection of communication paths and communication methods, and the communication paths give priority to paths that can maximize load transfer capabilities and improve grid reliability. The range of communication methods includes but is not limited to hand-in-hand wiring, multi-segment communication wiring, and ring network wiring.

9. The method for optimizing medium voltage grid segmentation and distributed power consumption based on big data clustering according to claim 1 is characterized in that: In step S4, when determining the DG location, priority is given to load-intensive areas, grid weak points, and areas with good absorption capacity, and it is ensured that the installation location of the DG will not have a negative impact on the stability of the grid.

10. The method for optimizing medium voltage grid segmentation and distributed power consumption based on big data clustering according to claim 1, characterized in that: In step S4, the capacity planning of DG specifically includes the following sub-steps: Load forecasting: forecast future loads, including load growth trends, seasonal changes, and daily changes, to determine DG capacity requirements; DG absorption capacity assessment: Evaluate the grid's absorption capacity for DG, including the capacity, stability, and regulation capabilities of the grid, to determine the DG capacity range that the grid can accommodate; DG capacity planning: According to the load forecast and DG absorption capacity assessment results, the DG capacity is planned. The planning principle is that the DG capacity can meet the load demand without having a negative impact on the stability of the power grid. At the same time, the redundant capacity of DG and the configuration of backup power supply are considered.

Images