Power protection method and system for medium-voltage generator car based on distribution automation graph model positioning access point

Abstract

The invention discloses a medium-voltage generator car power protection method and system based on power distribution automation graph model positioning access point, and the method comprises the steps: obtaining power distribution network automation graph model data, carrying out the processing of the data, obtaining the processed data, and building a power distribution network automation graph model database according to the processed data, acquiring real-time operation data and historical load data from the power distribution network automation graph model database, performing feature extraction on the real-time operation data and the historical load data, analyzing the extracted features, and predicting the load of each region of the power distribution network in a future time period by adopting a trained load prediction model to obtain a prediction result; and searching and determining a plurality of candidate access points in the power distribution network automation graph model, screening out an optimal access point from the plurality of candidate access points, and sending the information of the optimal access point to the medium-voltage power generation vehicle access control module. According to the method, the medium-voltage generator car can be quickly, accurately and efficiently connected to the power distribution network for power protection, and the reliability and stability of power supply are improved.

Description

Technical Field

[0001] This invention relates to the field of power supply security technology, specifically to a method and system for power supply security of medium-voltage generator vehicles based on the location of access points in distribution automation diagrams. Background Technology

[0002] With socio-economic development and improved living standards, higher demands are placed on the reliability and stability of power supply. In the face of power outages caused by natural disasters, equipment failures, or planned maintenance, rapid and effective power protection measures are needed to minimize the impact on social production and daily life. Medium-voltage generator trucks, as an important emergency power supply device, can quickly connect to the distribution network in emergencies, providing temporary power support to critical areas or important users.

[0003] Currently, numerous problems exist in the process of connecting medium-voltage generators to the distribution network for power protection. On the one hand, the selection of connection points often relies on manual experience and judgment, lacking a scientifically precise positioning method. Manual judgment is easily influenced by subjective factors and cannot comprehensively consider complex factors such as the real-time operating status of the distribution network, load distribution, and network topology. This may lead to unreasonable connection point selection, failing to fully utilize the power protection efficiency of the medium-voltage generators, and even negatively impacting the safe and stable operation of the distribution network. On the other hand, the lack of effective coordination and automated control between various stages of the connection process results in cumbersome procedures, long processing times, and difficulty in meeting the requirements for rapid response in emergency situations. For example, after determining the connection point, a large amount of manual work is required for line connections, parameter settings, and equipment debugging. Furthermore, delays or errors are prone to occur in the coordination between steps, seriously affecting the timeliness and reliability of power protection work.

[0004] In addition, existing power supply protection schemes are insufficient in real-time monitoring and analysis of the distribution network's operating status. They cannot adjust the operating parameters and access strategies of medium-voltage generators in a timely and accurate manner according to changes in the distribution network, making it difficult to achieve efficient and intelligent power supply protection goals. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for power supply protection of medium-voltage generator vehicles based on distribution automation diagrams to locate access points, thereby solving problems such as inaccurate access point selection, low efficiency of the access process, and lack of real-time intelligent control when medium-voltage generator vehicles are connected to the distribution network for power supply protection.

[0006] This invention is achieved through the following technical solution:

[0007] In a first aspect, the first embodiment of the present invention provides a method for ensuring power supply for a medium-voltage generator vehicle based on a distribution automation map model for locating access points, comprising: The distribution network automation model data is acquired and processed to obtain processed data. The processed data is then used to establish a distribution network automation model database. The distribution network automation model data includes the distribution network topology, equipment parameters, geographic information, real-time operation data, and historical load data. Real-time operation data and historical load data are obtained from the distribution network automation diagram database. Features are extracted from the real-time operation data and historical load data, and data mining algorithms are used to analyze the extracted features. The trained load forecasting model is used to predict the load of each area of ​​the distribution network in the future time period, and the prediction results are obtained. The load prediction model includes a CNN module, a Seq2Seq module, and an Attention module. The CNN module performs convolution operations on the input spatiotemporal load data to extract spatial and temporal features of load between regions. The encoder of the Seq2Seq module encodes the spatial and temporal features extracted by the CNN module into a context vector to capture long-term temporal dependencies. The decoder generates load prediction values ​​for future time steps based on the context vector and the weights assigned by the Attention module. The Attention module calculates the correlation between each time step of the decoder and all time steps of the encoder and dynamically assigns weights. Based on the distribution network topology, equipment parameters, and prediction results, multiple candidate access points are searched and determined in the distribution network automation model. A genetic algorithm is used to select the optimal access point from the multiple candidate access points, and the optimal access point is marked in the distribution network automation model. The optimal access point information is then sent to the medium-voltage generator vehicle access control module.

[0008] Furthermore, the method also includes: real-time monitoring of the operating status of the distribution network after the medium-voltage generator is connected and the operating indicators of the medium-voltage generator, and using optimization algorithms to dynamically adjust the operating indicators of the medium-voltage generator and optimize the operating status of the distribution network.

[0009] Furthermore, the method also includes: collecting the power generation of the medium-voltage generator in real time, transmitting the power generation to the power consumption acquisition system and the synchronous line loss system, and correcting the line loss and the power consumption line loss of the transformer area based on the power generation.

[0010] Furthermore, the specific method for processing the distribution network automation diagram data to obtain processed data, and establishing a distribution network automation diagram database based on the processed data, includes: Data cleaning is performed on the distribution network automation diagram data to identify and process missing values, outliers, and duplicate data. Convert data from different sources into a unified format and coordinate system, and standardize device IDs, attribute fields, and power grid topology. Geographic coordinates, equipment names, and asset numbers are used as key fields to associate geographic data with the power grid topology; A relational database is used to store the power grid topology, and equipment status tables, topology connection tables, and geographic information tables are established to obtain a distribution network automation map model database.

[0011] Furthermore, the specific method for selecting the optimal access point from multiple candidate access points using a genetic algorithm includes: Voltage stability, line loss, and economic cost are used as target optimization indicators. The indicators are normalized to obtain the normalized values ​​of voltage stability (V), line loss (P), and economic cost (D). Different weights, W1, W2, and W3, are assigned to voltage stability, line loss, and economic cost, respectively. The fitness function F is calculated using the following formula: ; Where A is the penalty term used to handle constraints, W1 is the weight of voltage stability, W2 is the weight of line loss, and W3 is the weight of economic cost.

[0012] Furthermore, specific methods for searching and determining multiple candidate access points in the distribution network automation model based on the distribution network topology, equipment parameters, and prediction results include: Initially select nodes with larger conductor cross-sectional areas or nodes close to the load center. Among the initially selected nodes, select nodes where the rated current of the switchgear is greater than the maximum output current of the generator car and the bus voltage level must match the output voltage of the generator car to obtain the nodes after the first screening. After the first round of filtering, nodes in areas with a rapid future load growth trend are selected to obtain nodes after the second round of filtering. The nodes after the second screening are sorted according to the priority of the power supply task, resulting in multiple candidate access points.

[0013] Secondly, another embodiment of the present invention provides a power supply protection system for a medium-voltage generator vehicle based on a distribution automation map model for locating access points, used to implement the method described in the above embodiments, including: The automated diagram management module is used to acquire automated diagram data of the distribution network, process the automated diagram data of the distribution network to obtain processed data, and establish a distribution network automated diagram database based on the processed data. The automated diagram data of the distribution network includes the topology of the distribution network, equipment parameters, geographical information, real-time operation data and historical load data. The load forecasting module is used to obtain real-time operation data and historical load data from the distribution network automation model database, extract features from the real-time operation data and historical load data, analyze the extracted features using data mining algorithms, and use a trained load forecasting model to predict the load of each area of ​​the distribution network for future time periods to obtain the forecast results. The access point positioning module is used to search and determine multiple candidate access points in the distribution network automation model based on the distribution network topology, equipment parameters and prediction results. It uses a genetic algorithm to select the optimal access point from the multiple candidate access points, marks it in the distribution network automation model, and sends the optimal access point information to the medium-voltage generator vehicle access control module.

[0014] Furthermore, it also includes an optimization module, which is used to monitor the operating status of the distribution network and the operating indicators of the medium-voltage generator after the medium-voltage generator is connected in real time, and to use optimization algorithms to dynamically adjust the operating indicators of the medium-voltage generator and optimize the operating status of the distribution network.

[0015] Furthermore, it also includes a metering module for real-time collection of the power generation from the medium-voltage generator and transmission of the power generation to the power consumption collection system and the synchronous line loss system, and correction of line loss and transformer area power loss based on the power generation.

[0016] Furthermore, the automated map model management module includes a data processing unit. The data processing unit performs data cleaning on the distribution network automated map model data, identifies and processes missing values, outliers, and duplicate data, converts data from different sources into a unified format and coordinate system, standardizes equipment IDs, attribute fields, and power grid topology, uses geographic coordinates, equipment names, and asset numbers as key fields to associate geographic data with the power grid topology, stores the power grid topology using a relational database, and establishes equipment status tables, topology connection tables, and geographic information tables to obtain the distribution network automated map model database.

[0017] Compared with the prior art, the present invention has the following advantages and beneficial effects: This invention provides a method and system for power supply protection using a medium-voltage generator vehicle based on distribution automation diagrams to locate the access point. By acquiring distribution network automation diagram data and establishing a database of this data, an accurate and comprehensive data foundation is provided for subsequent load analysis, access point evaluation, and location. Real-time analysis and short-term forecasting of distribution network load through load prediction allow for advance understanding of load change trends, leading to more rational selection of access points and adjustment of medium-voltage generator vehicle operating parameters. This effectively improves the targeting and effectiveness of power supply protection provided by the medium-voltage generator vehicle, better meeting the power demand of the distribution network at different times. It accurately selects the optimal access point from numerous candidate access points, avoiding the subjectivity and blindness of manual selection, ensuring that the impact of the medium-voltage generator vehicle on the distribution network is minimized, while maximizing its power supply protection function and improving the safety and stability of the distribution network. This achieves rapid, accurate, and efficient access of medium-voltage generator vehicles to the distribution network for power supply protection, improving the reliability and stability of power supply. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 A flowchart of a method for ensuring power supply to a medium-voltage generator vehicle based on a distribution automation diagram for locating access points, provided in the first embodiment of the present invention; Figure 2 The present invention provides a structural block diagram of a medium-voltage generator vehicle based on a power distribution automation map model positioning access point power supply system, which is provided as another embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0020] like Figure 1 As shown, the first embodiment of the present invention provides a method for ensuring power supply to a medium-voltage generator vehicle based on the location of the access point in a distribution automation map, which is applicable to the power supply system for a medium-voltage generator vehicle based on the location of the access point in a distribution automation map provided in this embodiment. The method includes the following steps: The distribution network automation model data is acquired and processed to obtain processed data. The processed data is then used to establish a distribution network automation model database. The distribution network automation model data includes the distribution network topology, equipment parameters, geographic information, real-time operation data, and historical load data. Real-time operation data and historical load data are obtained from the distribution network automation diagram database. Features are extracted from the real-time operation data and historical load data, and data mining algorithms are used to analyze the extracted features. The trained load forecasting model is used to predict the load of each area of ​​the distribution network in the future time period, and the prediction results are obtained. Based on the distribution network topology, equipment parameters, and prediction results, multiple candidate access points are searched and determined in the distribution network automation model. A genetic algorithm is used to select the optimal access point from the multiple candidate access points, and the optimal access point is marked in the distribution network automation model. The optimal access point information is then sent to the medium-voltage generator vehicle access control module.

[0021] Specifically, a medium-voltage generator vehicle establishes data interfaces with the power supply guarantee system, distribution network automation system, power supply service system, and map-based operational system based on the distribution automation map-based positioning access point system. It uses the IEC 61970 / 61968 CIM standard protocol to acquire data such as power grid topology, equipment parameters, and real-time operating status. It integrates with the GIS system through OGC standard services to acquire geospatial data, including line coordinates, equipment location, and terrain features. For systems without standard interfaces, dedicated data adapters are developed to extract data through direct database connections and file imports.

[0022] The acquired data undergoes data cleaning to identify and process missing, outlier, and duplicate values. Data from different sources is converted to a unified format and coordinate system. Equipment IDs, attribute fields, and power grid topology are standardized. Geographic coordinates, equipment names, and asset numbers are used as key fields to associate geographic data with the power grid topology. A relational database is used to store the power grid topology, and equipment status tables, topology connection tables, and geographic information tables are established to obtain a distribution network automation model database. A message queue is deployed in the distribution network automation model database to monitor data source change events, achieving near real-time data updates. An incremental update mechanism is established to synchronize only changed equipment or topology data, reducing network transmission and system load. By acquiring and managing detailed distribution network model data, an accurate and comprehensive data foundation is provided for subsequent load analysis, access point assessment and location, and operation monitoring, improving the scientific nature and accuracy of power supply protection decisions.

[0023] Real-time operation data and historical load data are obtained from the distribution network automation graph database. Fourier transform is used to convert the load sequence to the frequency domain, identifying the daily and weekly cycles of the load. Autocorrelation function analysis is used to analyze the autocorrelation of the load sequence and determine the order of the time series model. A sliding window method is used to statistically analyze the mean, variance, maximum, and minimum load values ​​within the window, capturing local load variation characteristics. Principal component analysis is used to reduce the dimensionality of high-dimensional features, extracting key information and reducing redundancy. A graph convolutional network (GCN) combined with the distribution network topology is used to map the load data onto graph nodes, mining the electrical distance and connection characteristics between nodes. K-means clustering algorithm is used to cluster historical load curves into typical patterns, such as weekday patterns, weekend patterns, and holiday patterns. Apriori association rule mining algorithm is used to discover the correlation between load and external factors, such as an 80% increase in air conditioning load when the temperature is >35℃. By uncovering the correlation rules between loads across multiple regions, such as a load peak in region A followed by a load increase in region B within one hour, LSTM (Long Short-Term Memory) networks are used to capture long-term dependencies in load sequences and identify load change patterns across days and weeks, such as the load growth trend during weekday morning peak hours. Data mining algorithms are then used to analyze the extracted features, comprehensively uncovering the patterns and regularities of load data changes, providing a solid foundation for subsequent accurate forecasting.

[0024] In this embodiment, the load forecasting module introduces the Attention mechanism into the CNN (Convolutional Neural Network) and Seq2Seq models for short-term load forecasting of the distribution network area. CNN extracts spatiotemporal features, Seq2Seq captures long-term temporal dependencies, and Attention focuses on key information, ultimately improving multi-step forecasting accuracy. Seq2Seq (Sequence to Sequence) is a method that can generate another sequence from a given sequence using a specific generation method, and these two sequences can be of unequal length. This structure is also called an Encoder-Decoder model. The load forecasting model in this embodiment includes a CNN module, a Seq2Seq module, and an Attention module. The CNN module performs convolution operations on the input spatiotemporal load data to extract spatial and temporal features of the load between regions. The encoder of the Seq2Seq module encodes the spatial and temporal features extracted by the CNN module into a context vector, capturing long-term time-series dependencies, such as intraday peaks and troughs and weekly cycles. The decoder generates load forecast values ​​for future time steps based on the context vector and the weights assigned by the Attention module. The Attention module calculates the correlation between each time step of the decoder and all time steps of the encoder and dynamically assigns weights, allowing the load forecasting model to focus on the historical information most important for the current forecast, such as extreme weather periods or peak load periods. By performing real-time analysis and short-term forecasting of the distribution network load, it is possible to grasp the load change trend in advance, making the selection of access points and the adjustment of the operating parameters of medium-voltage generators more reasonable, effectively improving the pertinence and effectiveness of medium-voltage generator power supply, and better meeting the power demand of the distribution network at different times.

[0025] In this embodiment, the specific method for searching and determining multiple candidate access points in the distribution network automation model based on the distribution network topology, equipment parameters, and prediction results includes: initially selecting nodes with larger conductor cross-sectional areas or nodes close to load centers; selecting nodes from the initially selected nodes whose rated current of the switching equipment is greater than the maximum output current of the generator and whose bus voltage level matches the output voltage of the generator, resulting in nodes after the first screening; through the first screening, nodes that meet the basic access conditions are initially selected, and obviously unsuitable nodes are eliminated. From the nodes after the first screening, nodes in areas with rapid future load growth trends are selected to obtain nodes after the second screening, in order to prepare for peak loads in advance and improve power supply protection. The second screening further narrows down the scope by combining load prediction results with the nodes after the first screening. The nodes after the second screening are sorted according to the priority of the power supply protection tasks to obtain multiple candidate access points.

[0026] For each candidate access point, an access point evaluation model is used to calculate evaluation parameters such as voltage stability, short-circuit current level, load transfer capacity, distance to important users, and line loss after access. By comprehensively analyzing and comparing these evaluation parameters, a genetic algorithm is used to select the optimal access point from multiple candidate access points, and its precise location is marked on the distribution automation diagram. The optimal access point information is then sent to the medium-voltage generator vehicle access control module.

[0027] The specific method for selecting the optimal access point from multiple candidate access points using a genetic algorithm includes: assuming there are N candidate access points, where N is a natural number. Chromosome encoding is used, with a chromosome length of N. Each gene corresponds to a candidate access point; a gene value of 1 indicates that the medium-voltage generator is connected to that candidate point, and a gene value of 0 indicates that it is not connected to that point.

[0028] Voltage stability, line loss, and economic cost are used as target optimization indicators. Voltage stability is measured by the voltage deviation at the connection point, and the lower the better. Line loss is measured by the active power loss from the connection point to the critical load, and the lower the better. Economic cost is measured by the distance from the generator truck to the connection point, and the lower the better. Each indicator is normalized to make it comparable, resulting in a normalized value V for voltage stability, a normalized value P for line loss, and a normalized value D for economic cost. Different weights are assigned to voltage stability, line loss, and economic cost, W1, W2, and W3, respectively. The fitness function F is calculated using the following formula: ; Where A is the penalty term used to handle constraints, W1 is the weight for voltage stability, W2 is the weight for line loss, and W3 is the weight for economic cost. Constraints include voltage and capacity constraints. If an access scheme violates these constraints, such as exceeding voltage limits or line overload, a large negative penalty value is applied, significantly reducing its fitness and causing it to be eliminated in the evolutionary process. The smaller the original value (1 minus the normalized value), the greater its contribution to fitness.

[0029] M individuals are randomly generated, where M is the population size. Each individual represents a randomly selected candidate entry point, forming the initial population. Individuals in the current population are eliminated, crossovered, and mutated, and the resulting individuals become the current population. For example, with a population size of M=5 and candidate entry points N=10, a possible initial population is [3, 7, 1, 9, 4]. Each individual in the population requires evaluation: its corresponding objective function value is calculated, all constraints are checked, and its fitness value is calculated based on the fitness function. A tournament selection method is used to select superior individuals from the current population as parents to produce offspring; individuals with higher fitness have a higher probability of being selected. Simulating biological gene recombination, the genes of two parent individuals are exchanged to produce new offspring, increasing population diversity. One or more gene positions in an individual's gene string are randomly modified to prevent the population from prematurely converging to a local optimum. Through selection, crossover, and mutation operations, M new individuals are generated to form a new generation population. The process of evaluating individuals and generating a new generation population is repeated until the preset maximum number of generations is reached. The individual with the highest fitness value in the last generation population is output, and the entry point it represents is the optimal entry point found by the genetic algorithm.

[0030] Another embodiment of the present invention provides a method for ensuring power supply for a medium-voltage generator vehicle based on the location of its access point in a distribution automation diagram. Unlike the first embodiment, this method further includes: real-time monitoring of the operating status of the distribution network after the medium-voltage generator vehicle's access and the operating indicators of the medium-voltage generator vehicle. The operating status includes changes in parameters such as voltage, current, and power, while the operating indicators include output power, operating temperature, and fuel consumption. An optimization algorithm is used to dynamically adjust the operating indicators of the medium-voltage generator vehicle and optimize the operating status of the distribution network. This includes adjusting the opening and closing states of line sectionalizing switches and optimizing load distribution to reduce line losses, improve the operating efficiency of the distribution network, and ensure that the distribution network remains in a safe, stable, and efficient operating state during the power supply guarantee period. When abnormal changes occur in the operating status of the distribution network, a timely warning message is issued, along with corresponding fault handling suggestions. When abnormal conditions such as voltage exceeding limits, current overload, or low power factor are detected in the distribution network, a timely warning signal is issued, and corresponding fault handling suggestions are provided according to a preset fault handling strategy to guide maintenance personnel in troubleshooting and repair.

[0031] Another embodiment of the present invention provides a method for ensuring power supply of a medium-voltage generator vehicle based on the location of the access point of the distribution automation map, which differs from the first embodiment in that the method further includes: real-time collection of the power generation of the medium-voltage generator vehicle, and transmission of the power generation to the power consumption collection system and the synchronous line loss system, and correction of line loss and transformer area power line loss based on the power generation, which is beneficial for calculating power generation cost and efficiency.

[0032] This invention provides a method for locating access points for medium-voltage generator vehicles based on distribution automation diagrams to ensure power supply. By acquiring distribution network automation diagram data and establishing a database of such data, an accurate and comprehensive data foundation is provided for subsequent load analysis, access point evaluation, and location. Real-time analysis and short-term forecasting of distribution network load through load prediction allow for advance understanding of load change trends, leading to more rational selection of access points and adjustment of medium-voltage generator vehicle operating parameters. This effectively improves the targeting and effectiveness of medium-voltage generator vehicle power supply, better meeting the power demand of the distribution network at different times. It can accurately select the optimal access point from numerous candidate access points, avoiding the subjectivity and blindness of manual access point selection, ensuring that the impact of medium-voltage generator vehicle access on the distribution network is minimized, while maximizing its power supply protection function and improving the safety and stability of the distribution network. This method enables medium-voltage generator vehicles to quickly, accurately, and efficiently access the distribution network for power supply protection, improving the reliability and stability of power supply.

[0033] like Figure 2 As shown, another embodiment of the present invention provides a medium-voltage generator vehicle-based power supply system for locating access points using a distribution automation diagram, used to implement the method described in the above embodiments, including: The automated diagram management module is used to acquire automated diagram data of the distribution network, process the automated diagram data of the distribution network to obtain processed data, and establish a distribution network automated diagram database based on the processed data. The automated diagram data of the distribution network includes the topology of the distribution network, equipment parameters, geographical information and real-time operation data. The load forecasting module is used to obtain real-time operation data and historical load data from the distribution network automation model database, extract features from the real-time operation data and historical load data, analyze the extracted features using data mining algorithms, and use a trained load forecasting model to predict the load of each area of ​​the distribution network for future time periods to obtain the forecast results. The access point positioning module searches for and determines multiple candidate access points in the distribution network automation model based on the distribution network topology, equipment parameters, and prediction results. It then uses a genetic algorithm to select the optimal access point from the multiple candidate access points, marks it in the distribution network automation model, and sends the optimal access point information to the medium-voltage generator vehicle access control module.

[0034] The automated graph management module, serving as the core data support module, establishes direct connections with the load forecasting module, access point location module, and optimization module, providing foundational data for these three modules. The output of the load forecasting module is connected to the input of the access point location module, transmitting the load analysis and forecasting results to it. The output of the access point location module is connected to the input of the medium-voltage generator vehicle access control module, transmitting optimal access point information to it.

[0035] In this embodiment, the system also includes an optimization module, which is used to monitor the operating status of the distribution network and the operating indicators of the medium-voltage generator after the medium-voltage generator is connected in real time, and to use optimization algorithms to dynamically adjust the operating indicators of the medium-voltage generator and optimize the operating status of the distribution network.

[0036] In this embodiment, it also includes a metering module for real-time collection of the power generation of the medium-voltage generator and transmission of the power generation to the power consumption collection system and the synchronous line loss system, and correction of line loss and transformer area power line loss based on the power generation.

[0037] In this embodiment, the automated map model management module includes a data processing unit. The data processing unit performs data cleaning on the distribution network automated map model data, identifies and processes missing values, outliers, and duplicate data, converts data from different sources into a unified format and coordinate system, standardizes equipment IDs, attribute fields, and power grid topology, uses geographic coordinates, equipment names, and asset numbers as key fields to associate geographic data with the power grid topology, stores the power grid topology using a relational database, and establishes an equipment status table, a topology connection table, and a geographic information table to obtain the distribution network automated map model database.

[0038] The execution process of each module can be carried out according to the steps of the method for locating access points for a medium-voltage generator vehicle based on a distribution automation diagram in the first embodiment. These steps will not be described in detail in this embodiment.

[0039] The first embodiment of the present invention will be described below using two specific examples: The medium-voltage generator truck of the first embodiment of this invention, based on the distribution automation model positioning access point power supply protection system, was applied to a 110kV substation under power outage conditions. The automation model management module quickly acquired and updated model data, including the substation and surrounding distribution network topology, equipment parameters, and real-time operating data. The load forecasting module analyzed and predicted the load in the out-of-voltage area and its surroundings, finding that the load in this area would remain relatively stable at a certain level for the next 3 hours. Based on relevant data, the access point positioning module identified three candidate access points in the distribution automation model. Through evaluation model calculations, the optimal access point was ultimately selected as the one closest to the out-of-voltage substation, with good voltage stability and low line loss. The medium-voltage generator truck access control module generated an access operation plan, remotely controlling the medium-voltage generator truck to travel to the access point and complete the access according to the plan. During the access process, all real-time monitored electrical parameters remained normal. The optimization module continuously monitored after access and fine-tuned the output power of the medium-voltage generator truck based on the monitoring data to ensure stable distribution network voltage, restoring stable power supply to the area and reducing losses caused by the power outage. The metering module is used to collect the power generation of the medium-voltage generator in real time and transmit the power generation to the power consumption acquisition system and the synchronous line loss system, and correct the line loss and the power consumption line loss of the transformer area based on the power generation.

[0040] This invention's medium-voltage generator truck, based on a distribution automation map model positioning access point power supply system, was applied to the power supply guarantee work for the National College Entrance Examination (NCEE). The automated map model management module acquires detailed map model data of the examination site and surrounding distribution network, including equipment parameters and geographical information of the lines where the examination site is located, and updates it in real time. The load forecasting module, combining historical load data from the NCEE period, predicts the load change trend of the examination site and its surrounding area during the examination period, anticipating a certain increase in load after the start of the examination. The access point positioning module, based on the forecast results and distribution network data, selects the access point closest to the examination site with strong load transfer capability as the optimal access point. The medium-voltage generator truck access control module controls the medium-voltage generator truck to arrive at the access point in advance and complete the access process safely and efficiently. The optimization module monitors in real time during the examination period and dynamically adjusts the operating parameters of the medium-voltage generator truck and the operating mode of the distribution network to ensure a stable and reliable power supply to the examination site, guaranteeing the smooth conduct of the NCEE.

[0041] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for power protection of a medium-voltage power generation vehicle based on positioning of an access point according to a power distribution automation map, characterized in that, The method comprises the following steps: obtaining power distribution network automation diagram data, processing the power distribution network automation diagram data to obtain processed data, and establishing a power distribution network automation diagram database according to the processed data, wherein the power distribution network automation diagram data comprises the topological structure, equipment parameters, geographical information, real-time operation data and historical load data of the power distribution network; obtaining real-time operation data and historical load data from the power distribution network automation diagram database, extracting features from the real-time operation data and historical load data, analyzing the extracted features by using a data mining algorithm, predicting the load of each region of the power distribution network in a future time period by using a trained load prediction model, and obtaining a prediction result; the load prediction model comprises a CNN module, a Seq2Seq module and an Attention module, the CNN module performs convolution operation on input spatio-temporal load data to extract spatial features and temporal features of the load between regions, the encoder of the Seq2Seq module encodes the spatial features and temporal features extracted by the CNN module into context vectors to capture long time sequence dependence, the decoder generates a load prediction value of a future time step based on the context vectors and the weight assigned by the Attention module, and the Attention module calculates the correlation between each time step of the decoder and all time steps of the encoder and dynamically assigns the weight; a plurality of candidate access points are searched for and determined in the power distribution network automation diagram according to the topological structure, equipment parameters and prediction result of the power distribution network, a genetic algorithm is used to screen an optimal access point from the plurality of candidate access points, the optimal access point is marked in the power distribution network automation diagram, and the information of the optimal access point is sent to a medium-voltage power generation vehicle access control module.

2. The method of claim 1, wherein, The method further comprises the following steps: monitoring the running state of the power distribution network after the medium-voltage power generation vehicle is accessed and the running index of the medium-voltage power generation vehicle in real time, and dynamically adjusting the running index of the medium-voltage power generation vehicle and optimizing the running state of the power distribution network by using an optimization algorithm.

3. The method of claim 2, wherein, The method further comprises the following steps: collecting the power generation amount of the medium-voltage power generation vehicle in real time, and transmitting the power generation amount to a power consumption collection system and a same-period line loss system, and correcting the line loss and the line loss of the power consumption of a transformer area according to the power generation amount.

4. The method of claim 1, wherein, The specific method for processing the power distribution network automation diagram data to obtain processed data and establishing a power distribution network automation diagram database according to the processed data comprises the following steps: performing data cleaning on the power distribution network automation diagram data, and identifying and processing missing values, abnormal values and duplicate data; converting data from different sources into a unified format and coordinate system, standardizing the equipment ID, attribute field and power grid topological structure; associating geographical data with the power grid topological structure by using geographical coordinates, equipment names and asset numbers as key fields; storing the power grid topological structure by using a relational database, establishing a device state table, a topological connection table and a geographical information table, and obtaining the power distribution network automation diagram database.

5. The method of claim 1, wherein, The specific method for screening an optimal access point from a plurality of candidate access points by using a genetic algorithm comprises the following steps: taking voltage stability, line loss and economic cost as target optimization indexes; The indicators are normalized to obtain a voltage stability normalized value V, a line loss normalized value P and an economic cost normalized value D; Different weights are given to the voltage stability, line loss and economic cost, respectively W1, W2 and W3, and the calculation formula of the fitness function F is: ； Wherein, A is a penalty term, used to process the constraint condition, W1 is the weight of voltage stability, W2 is the weight of line loss, and W3 is the weight of economic cost.

6. The method of claim 1, wherein, The specific method for searching and determining a plurality of candidate access points in the power distribution network automation graph model according to the topological structure, equipment parameters and prediction results of the power distribution network comprises: Preliminarily selecting nodes with larger line conductor cross-sectional area or nodes close to the load center, and selecting nodes with switch equipment rated current greater than the maximum output current of the power car and bus voltage level matching the output voltage of the power car from the preliminarily selected nodes to obtain the nodes after the first screening; Selecting regional nodes with fast future load growth trend from the nodes after the first screening to obtain the nodes after the second screening; According to the priority of the power protection task, the nodes after the second screening are sorted to obtain a plurality of candidate access points.

7. A medium voltage power generation vehicle based on power protection system for locating access points based on distribution automation map, characterized in that, For implementing the method of any one of claims 1-6, comprising: An automation graph model management module is configured to acquire power distribution network automation graph model data, process the power distribution network automation graph model data to obtain processed data, and establish a power distribution network automation graph model database according to the processed data, wherein the power distribution network automation graph model data comprises the topological structure, equipment parameters, geographic information, real-time operation data and historical load data of the power distribution network; A load prediction module is configured to acquire real-time operation data and historical load data from the power distribution network automation graph model database, extract features from the real-time operation data and historical load data, analyze the extracted features using a data mining algorithm, and predict the load of each region of the power distribution network in a future time period using a trained load prediction model to obtain prediction results; An access point positioning module is configured to search and determine a plurality of candidate access points in the power distribution network automation graph model according to the topological structure, equipment parameters and prediction results of the power distribution network, select an optimal access point from the plurality of candidate access points using a genetic algorithm, mark the optimal access point in the power distribution network automation graph model, and send the optimal access point information to the medium-voltage power car access control module.

8. The system of claim 7, wherein, Further comprising an optimization module configured to monitor the running state of the power distribution network after the medium-voltage power car is accessed and the running index of the medium-voltage power car in real time, and dynamically adjust the running index of the medium-voltage power car and optimize the running state of the power distribution network using an optimization algorithm.

9. The system of claim 8, wherein, Further comprising a metering module configured to collect the power generation of the medium-voltage power car in real time, and transmit the power generation to a power consumption collection system and a same-period line loss system, and correct the line loss and the line loss of the transformer area according to the power generation.

10. The system of claim 7, wherein, The automatic graph model management module comprises a data processing unit, which performs data cleaning on power distribution network automation graph model data, identifies and processes missing values, abnormal values and repeated data, converts data from different sources into a unified format and coordinate system, standardizes device IDs, attribute fields and power grid topological structures, uses geographic coordinates, device names and asset numbers as key fields, associates geographic data with power grid topological structures, stores the power grid topological structures in a relational database, and establishes a device state table, a topological connection table and a geographic information table to obtain a power distribution network automation graph model database.

Images