Power distribution network intelligent planning system and method for tourist attractions

Abstract

The invention relates to the technical field of power distribution network intelligent control, in particular to a power distribution network intelligent planning system and method for scenic spots, and the method comprises the steps: precisely estimating the power demand fluctuation through the time series data analysis of the activity schedule and historical power use mode of the scenic spots, and generating a load prediction table; a power grid topological structure and current and voltage fluctuation are monitored in real time through an isolated forest algorithm, potential fault points are accurately diagnosed, the risk of fault occurrence is effectively reduced, fault prediction has higher sensitivity, potential problems can be recognized in advance, the influence of power grid faults on power supply stability is reduced, and the power supply reliability is improved. The operation cost and saving are analyzed through the random forest, and the cost-benefit ratio of different optimization schemes is quantitatively evaluated, so that more accurate decision support is provided for power grid operation optimization, the flexibility and stability of a power system are improved, and efficient utilization of power resources and maximization of economic benefits are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of distribution networks, and in particular to an intelligent planning system and method for distribution networks used in tourist attractions. Background Art

[0002] The field of intelligent control technology for distribution networks aims to achieve efficient allocation and utilization of power resources through advanced control technologies and algorithms, and to improve power supply reliability, reduce energy consumption and save operating costs by collecting and analyzing distribution network operation data in combination with prediction models and optimization algorithms.

[0003] The purpose of the intelligent planning system for distribution networks used in tourist attractions is to rationally plan and dispatch power resources based on the characteristics of power demand at tourist attractions, ensure the stability and efficiency of power supply, meet the power demand of tourist attractions during peak hours, reduce operational risks caused by insufficient power supply or power grid failures, and improve the overall user experience and service quality.

[0004] Existing technologies have low accuracy in predicting fluctuations in electricity demand at scenic spots and often rely on single historical data analysis. They cannot fully consider the particularity and time-period changes of scenic spot activities, resulting in irrational allocation of electricity resources, insufficient or surplus electricity during peak hours, and affecting user experience. The design and scheduling of distribution networks lack flexibility, and conventional topological structures are less efficient when dealing with high-demand periods, resulting in electricity waste or unstable supply. Fault prediction and emergency response mechanisms cannot effectively identify potential faults in advance, lack pertinence and flexibility, and affect the normal operation of scenic spots and increase operational risks. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a distribution network intelligent planning system and method for tourist attractions.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: A distribution network intelligent planning system for tourist attractions comprises:

[0007] Grid demand forecasting module: Based on the activity schedule and historical power usage patterns of tourist attractions, through time series data analysis, it estimates the power demand fluctuations within a specific time period, calculates the expected power demand, and generates a load forecast table for tourist attractions;

[0008] Power supply scheme design module: according to the load forecast table of the scenic spot, analyze the location and capacity of the distribution station and transformer, integrate the adaptation path of the power grid line, optimize the line layout, and obtain the power supply system design diagram;

[0009] Fault prediction and diagnosis module: Based on the power supply system design diagram, the isolation forest algorithm is used to extract the topological data of the power grid, analyze the key nodes and connections of the power grid structure, detect abnormal fluctuations in current and voltage, identify potential fault sources and risk areas, and obtain fault point diagnosis results;

[0010] Emergency response plan module: based on the fault point diagnosis results, analyze the impact range and emergency level of the fault point, formulate quick operation steps for fault point isolation and power reload, and create emergency operation guidelines;

[0011] Cost and benefit analysis module: conduct economic model analysis on the power supply system design and emergency operation guide, use random forest to calculate the total cost and expected savings of system operation, compare the cost-benefit ratio of different solutions, and output a cost-benefit report;

[0012] System optimization and update module: Based on the cost-benefit report, through the power grid monitoring data, continuously track the system performance indicators, adjust the power grid configuration and management strategy according to the real-time feedback, and generate the configuration update plan.

[0013] As a further solution of the present invention, the power grid demand forecasting module includes a time series analysis submodule, a power demand estimation submodule and a load table generation submodule, wherein:

[0014] Time series analysis submodule: Based on the activity schedule and historical power usage patterns of tourist attractions, the time series segmentation and statistical analysis of the activity schedule are carried out to extract the power usage fluctuation pattern within the time interval, extract the high and low frequency band division and periodic characteristics of power usage, and generate a power time series feature table;

[0015] Power demand estimation submodule: Based on the power time series characteristic table, the power demand in each time period is estimated, and the peak demand and valley demand are dynamically calculated in combination with the specific power demand change characteristics of activity schedules and holidays to generate power demand fluctuation data;

[0016] Load table generation submodule: Based on the power demand fluctuation data, the load data in different scenic spots are divided, the load change values ​​within the time period are summarized and partitioned and classified, and the load demand timetable of the scenic spots is divided to generate a load forecast table for the scenic spots.

[0017] As a further solution of the present invention, the power supply scheme design module includes a load analysis submodule, a line optimization submodule and a system design submodule, wherein:

[0018] Load analysis submodule: Based on the scenic spot load forecast table, extract the data of load characteristics and power distribution equipment demand in the scenic spot area, calculate the number and location range of distribution stations in combination with load demand, allocate transformer capacity and plan installation location, and generate a power distribution equipment configuration list;

[0019] Line optimization submodule: Based on the distribution equipment configuration list, the grid adaptation path is screened and the line layout plan is optimized, the matching characteristics of the line and the equipment are integrated to calculate the path, the layout is adjusted in combination with the line transmission characteristics, and a line optimization path diagram is generated;

[0020] System design submodule: Based on the line optimization path map, detailed planning of line layout in the scenic area is carried out, the specific location and capacity of the distribution station and transformer are integrated, the connection and adaptation of various areas of the power supply system are carried out, the overall layout diagram of the power supply lines and equipment is drawn, and the power supply system design diagram is generated.

[0021] As a further solution of the present invention, the fault prediction and diagnosis module includes a topology analysis submodule, an anomaly detection submodule and a fault diagnosis submodule, wherein:

[0022] Topology analysis submodule: Based on the power supply system design diagram, the spatial position and connection relationship of each node in the power grid are extracted, the path information between nodes is counted and the data characteristics of the connection path are analyzed, the characteristics of key nodes in the power grid are analyzed by summarizing and classifying the connection data, key connection areas are extracted and high-importance nodes are identified, and power grid topology data is generated;

[0023] Anomaly detection submodule: Based on the grid topology data, the isolation forest algorithm is used to extract the time series fluctuation information of current and voltage, perform continuous sampling and classification of the fluctuation range, and statistically analyze the data distribution of abnormal points in combination with the change trend of the fluctuation amplitude. The abnormal fluctuation points of voltage or current are located by point-by-point comparison, the fluctuation range is defined, and the potential abnormal area is confirmed, and the abnormal fluctuation area data is generated;

[0024] Fault diagnosis submodule: Based on the abnormal fluctuation area data, the node and connection relationship data of the abnormal area are extracted, and the structural characteristics of the abnormal nodes are confirmed by node-by-node analysis. The key fault points are gradually located in combination with the topological distribution and fluctuation characteristics in the area. The risk nodes and potential problem areas are identified by decomposing the connection areas, and the fault point diagnosis results are generated.

[0025] As a further solution of the present invention, the isolation forest algorithm is according to the formula:

[0026]

[0027] Where: S′(x, t, w, σ) is the improved anomaly score, h(x) is the average path length of sample x, c(n) is the normalized path length, Δt is the time interval between fluctuation points, σ ​​is the standard deviation of the fluctuation amplitude, μ is the mean of the fluctuation sequence, and w 1 、w 2 and w 3 is the weight coefficient, and x is the current sample point.

[0028] As a further solution of the present invention, the emergency response plan module includes an impact analysis submodule, an isolation operation submodule and an emergency guidance submodule, wherein:

[0029] Impact analysis submodule: Based on the fault point diagnosis results, the connection nodes and regional boundaries of the fault point are extracted, and the transmission characteristics between nodes related to the fault point are analyzed. The impact range is defined in combination with the connection characteristics and the connection structure of the affected area is analyzed. The main nodes that may be affected are identified through layer-by-layer transmission analysis, and the fault impact range data is generated;

[0030] Isolation operation submodule: based on the fault impact range data, extract key nodes and path data in the connection area, perform isolation operation path analysis on the fault area, adjust the power supply load by gradually isolating the connection path, re-plan the power distribution of the remaining area in combination with the load data, and generate isolation and reload operation steps;

[0031] Emergency guidance submodule: Based on the isolation and overload operation steps, the implementation area data is extracted, each step is disassembled and analyzed, and an implementation document is formed by organizing the key points of operation and execution conditions, combined with the distribution of access and load, to generate an emergency operation guide.

[0032] As a further solution of the present invention, the cost and benefit analysis module includes a cost calculation submodule, a benefit estimation submodule and a benefit comparison submodule, wherein:

[0033] Cost calculation submodule: Based on the power supply system design drawing and emergency operation guide, extract the data of equipment installation location and line laying length, summarize the various costs of operating power consumption and scheduling by counting the maintenance cycle and consumables consumption, integrate and calculate all cost data during operation, and generate the total operation cost of the power grid;

[0034] Benefit estimation submodule: Based on the total operating cost of the power grid, random forest is used to extract the power saving data brought by power supply optimization and energy-saving measures, and the potential loss data reduced by statistical scheduling optimization and accident emergency treatment are accumulated to accumulate the economic values ​​of various savings, extract the total actual benefits of energy saving and management optimization, and generate power grid saving benefits;

[0035] Benefit comparison submodule: Based on the power grid saving benefit, the cost data and benefit data of each optimization scheme are extracted, and a step-by-step comparative calculation is performed between the two. By analyzing the proportion of different schemes in savings and costs, the cost-benefit ratio data of each scheme is extracted and multi-dimensional grouping and summarization are performed to generate a cost-benefit data table.

[0036] As a further solution of the present invention, the random forest is according to the formula:

[0037]

[0038] Where: E′ is the energy saving benefit of the improved power grid, S j is the power saving benefit brought by the j-th power supply optimization, M j is the saving benefit generated by the jth energy-saving measure, L j is the potential loss reduced by the jth dispatch optimization and accident emergency treatment, P j is the rated power of the jth device, T j is the operating time of the jth device, E j is the energy efficiency ratio of the jth device, W 1 , W 2 , W 3 and W 4 is the weight coefficient, and m is the number of measures or the number of data items analyzed.

[0039] As a further solution of the present invention, the system optimization and update module includes a performance monitoring submodule, a real-time adjustment submodule and an update generation submodule, wherein:

[0040] Performance monitoring submodule: Based on the cost-effectiveness data table, extract system performance indicators and related operation data, dynamically collect the load distribution of the power grid lines and the power consumption of the equipment operation, conduct performance comparison analysis in different time periods, extract data on equipment aging or performance degradation to identify optimization areas, and generate performance monitoring data;

[0041] Real-time adjustment submodule: Based on the performance monitoring data, the specific distribution locations of high-load lines and aging equipment are extracted to redistribute line loads, and the equipment performance is optimized by gradually adjusting the operating parameters of related equipment. Local adjustment and optimization operations are performed in combination with load and equipment data to generate adjustment optimization results;

[0042] Update generation submodule: Based on the adjustment and optimization results, extract the optimized equipment configuration and line allocation data, dynamically update the overall configuration file of the power grid, and integrate and generate implementable system configuration and power grid optimization files by summarizing the new parameters of equipment operation and load distribution data, and generate a configuration update plan.

[0043] A method for intelligent planning of a distribution network for a tourist attraction, wherein the method is executed based on the above-mentioned intelligent planning system for a distribution network for a tourist attraction, and comprises the following steps:

[0044] Step 1: According to the activity schedule and historical power usage pattern of tourist attractions, extract the power usage records of each time period, calculate the load variation range within the time period, count the high-frequency and low-frequency fluctuation characteristics in the time series, obtain the load demand data of each time period, and generate the load demand characteristics of the tourist attractions;

[0045] Step 2: Based on the load demand characteristics of the scenic spot, extract the power supply device data and load distribution characteristics in the area, calculate the power supply load distribution value of each area, optimize the path layout of the power grid line by combining the load distribution value and the line transmission characteristics, integrate the installation location and capacity information of the power supply device, and generate the power supply line layout data;

[0046] Step 3: Based on the power supply line layout data, extract the node locations and connection characteristics of the power grid, count the topological distribution between the lines and divide different path areas, analyze the key nodes and load transmission paths of the power grid, combine the isolation forest method to detect the time series fluctuation data of current and voltage, mark the nodes and areas where abnormal fluctuations may occur, and combine these nodes to make association judgments on potential fault points, and generate fault-related area distribution;

[0047] Step 4: Based on the distribution of the fault-related areas, extract the node and line information of the high-risk areas, calculate the load priority and isolation order of each node, formulate the isolation steps and the order of enabling the backup load according to the priority, integrate the isolation steps and the load distribution operation to generate an operation guide, and generate an emergency isolation and load distribution operation plan;

[0048] Step 5: Based on the emergency isolation and load distribution operation plan, extract the regional load change data and operation parameters after each operation, combine the random forest method to extract the characteristic distribution of the load data and related costs during the operation, calculate the saved operation cost and spare resource consumption during the dispatching process, conduct a comparative analysis of the cost and saving data, and generate cost and saving comparison data;

[0049] Step 6: Based on the cost and savings comparison data, extract the operating performance indicators and parameter distribution under different configurations, count the changes in power load distribution in each area, calculate the configuration optimization value of the high-load area, reallocate the load transmission and line paths of the power grid, adjust the dynamic configuration of the power grid and generate load transmission update documents, and generate distribution configuration update plans.

[0050] Compared with the prior art, the advantages and positive effects of the present invention are:

[0051] 1. In the present invention, by analyzing the time series data of tourist attraction activity schedules and historical power usage patterns, the power demand fluctuation is accurately estimated and a load forecast table is generated, so that the energy saving potential in the historical power data can be more efficiently extracted, and the accurate prediction of power demand and the quantification of the saving effect can be achieved;

[0052] 2. In the present invention, the isolation forest algorithm is used to monitor the grid topology and current and voltage fluctuations in real time, accurately diagnose potential fault points, effectively reduce the risk of faults, make fault prediction more sensitive, identify potential problems in advance, and reduce the impact of grid faults on power supply stability;

[0053] 3. In the present invention, the operating costs and savings are analyzed by random forest, and the cost-effectiveness ratio of different optimization schemes is quantitatively evaluated, which provides more accurate decision support for the optimization of power grid operation, continuously tracks the system performance and adjusts the power grid configuration according to real-time feedback, improves the flexibility and stability of the power system, and realizes the efficient utilization of power resources and maximizes the economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a system flow chart of the present invention;

[0055] Figure 2 It is a schematic diagram of the system framework of the present invention;

[0056] Figure 3 It is a schematic diagram of the method steps of the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0058] See also Figure 1 The present invention provides a technical solution: a distribution network intelligent planning system for tourist attractions comprises:

[0059] Grid demand forecasting module: Based on the activity schedule and historical power usage patterns of tourist attractions, through time series data analysis, it estimates the power demand fluctuations within a specific time period, calculates the expected power demand, and generates a load forecast table for tourist attractions;

[0060] Power supply scheme design module: According to the scenic spot load forecast table, analyze the location and capacity of the distribution station and transformer, integrate the adaptation path of the power grid line, optimize the line layout, and obtain the power supply system design diagram;

[0061] Fault prediction and diagnosis module: Based on the power supply system design diagram, the isolation forest algorithm is used to extract the topological data of the power grid, analyze the key nodes and connections of the power grid structure, detect abnormal fluctuations in current and voltage, identify potential fault sources and risk areas, and obtain fault point diagnosis results;

[0062] Emergency response plan module: Based on the fault point diagnosis results, analyze the impact range and emergency level of the fault point, formulate quick operation steps for fault point isolation and power reload, and create emergency operation guidelines;

[0063] Cost and benefit analysis module: conduct economic model analysis on the power supply system design drawings and emergency operation guidelines, use random forest to calculate the total cost and expected savings of system operation, compare the cost-benefit ratio of different solutions, and output a cost-benefit report;

[0064] System optimization and update module: Based on cost-benefit reports and grid monitoring data, it continuously tracks system performance indicators, adjusts grid configuration and management strategies based on real-time feedback, and generates configuration update plans.

[0065] See also Figure 2 ,The power grid demand forecasting module includes a time series analysis submodule, a power demand estimation submodule, and a load table generation submodule, where:

[0066] Time series analysis submodule: Based on the activity schedule and historical power usage patterns of tourist attractions, the time series segmentation and statistical analysis of the activity schedule are carried out to extract the power usage fluctuation pattern within the time interval, extract the high and low frequency band division and periodic characteristics of power usage, and generate a power time series feature table;

[0067] Power demand estimation submodule: Based on the power time series feature table, it estimates the power demand in each time period, combines the specific power demand change characteristics of activity schedules and holidays, dynamically calculates the peak demand and valley demand, and generates power demand fluctuation data;

[0068] Load table generation submodule: Based on the power demand fluctuation data, the load data in different scenic spots are divided, the load change values ​​within the time period are summarized and partitioned and classified, and the load demand timetable of the scenic spots is divided to generate the load forecast table of the scenic spots;

[0069] Time series analysis submodule: Based on the activity schedule and historical electricity usage pattern of tourist attractions, the dynamic time warping algorithm is used to segment the activity schedule into time series. The execution of the dynamic time warping algorithm includes inputting the activity schedule sequence and the historical electricity usage pattern sequence, setting the time series length matching parameters and the dynamic adjustment step size of the time point, and using the dynamic programming method to measure and calculate the sequence similarity when performing sequence matching and output the optimal path. At the same time, the statistical analysis algorithm is applied to the segmented time series to extract the fluctuation law. The statistical analysis algorithm specifically includes calculating the mean, variance and kurtosis of the electricity usage changes within the time period, and applying the short-time Fourier transform algorithm to divide the electricity usage data into high and low frequency bands. During the high and low frequency band division process, the frequency demarcation point needs to be set and the spectrum analysis operation needs to be performed. The window function is used to segment the time period signal and calculate the frequency power density spectrum. The autocorrelation analysis algorithm is used to extract the periodic characteristics, and the time lag value range is set to calculate the time domain correlation of the power data and determine the period length to generate the power time series feature table;

[0070] Power demand estimation submodule: Based on the power time series feature table, the support vector regression algorithm is used to estimate the power demand in each time period. When executing the support vector regression algorithm, the kernel function type needs to be set to radial basis function, the penalty coefficient is adjusted to 10, and the kernel function parameter is 0.1. The power demand fitting calculation is performed on the data of each time period in the power time series feature table. Combined with the activity schedule and the characteristics of the change in power demand specific to holidays, the time series decomposition algorithm is used to dynamically calculate the peak and valley values ​​of holiday demand. The execution of the time series decomposition algorithm includes inputting historical power demand data of holidays, setting the decomposition window size to 24 hours, separating the trend component and the periodic component, extracting the trend component through the smoothing filtering method, and using the discrete wavelet transform method to decompose the periodic component to generate power demand fluctuation data;

[0071] Load table generation submodule: Based on the power demand fluctuation data, the K-means clustering algorithm is used to divide the load data in different scenic spots. When performing clustering, the number of cluster centers needs to be set to 5. The initial cluster center is determined by random generation. The number of iterations is set to 100 times. The load change values ​​within the time period are summarized for the clustered load data. The partition integration algorithm is used to integrate and classify the load demand data in different regions. The partition integration algorithm includes grouping the load demand data according to the geographical area identification, classifying them by time zone based on the load characteristics of the time period, and using a smoothing method based on linear interpolation to process the time series continuity of the integrated load data. The load demand timetable is divided according to the scenic spots to generate a load forecast table for the scenic spots.

[0072] See also Figure 2 The power supply scheme design module includes a load analysis submodule, a line optimization submodule and a system design submodule, among which:

[0073] Load analysis submodule: Based on the scenic spot load forecast table, extract the data of load characteristics and power distribution equipment demand in the scenic spot area, calculate the number and location range of distribution stations based on load demand, allocate transformer capacity and plan installation location, and generate a power distribution equipment configuration list;

[0074] Line optimization submodule: Based on the distribution equipment configuration list, it screens the grid adaptation path and optimizes the line layout plan, integrates the matching characteristics of the line and the equipment to calculate the path, adjusts the layout based on the line transmission characteristics, and generates a line optimization path diagram;

[0075] System design submodule: Based on the line optimization path map, detailed planning of line layout in the scenic area is carried out, the specific location and capacity of the distribution station and transformer are integrated, the connection and adaptation of various areas of the power supply system are carried out, the overall layout diagram of the power supply line and equipment is drawn, and the power supply system design diagram is generated;

[0076] Load analysis submodule: Based on the scenic spot load prediction table, a linear regression algorithm is used to extract the load characteristics and distribution equipment demand data in the scenic spot area. When the linear regression algorithm is executed, the load data in the scenic spot load prediction table needs to be input, the target variable is set as the load demand value, and the characteristic variable is set as the time period and the scenic spot area identifier. The least squares method is used to estimate the parameters and output the load demand characteristic value. When the number and location range of distribution stations are calculated in combination with the load demand, the clustering analysis method is used to divide the distribution station demand. The clustering analysis method includes inputting the load data set, setting the number of clusters to 3, and using the K-means clustering algorithm to calculate the cluster center and group the data. When allocating transformer capacity and planning the installation location, the partition optimization method is used, the regional load characteristic data is input, and the regional capacity threshold is set to 2000kVA. The linear programming method is used to allocate and optimize the transformer capacity and generate a distribution equipment configuration list.

[0077] Line optimization submodule: Based on the distribution equipment configuration list, the shortest path algorithm is used to screen the grid adaptation path and optimize the line layout plan. The execution of the shortest path algorithm includes inputting the geographic location data of the distribution equipment, setting the initial path weight as the line length, setting the weight correction parameter as the terrain complexity, using the Dijkstra algorithm to calculate the path point by point and generate the optimal path set, integrating the line and equipment matching characteristics for path calculation, using the mixed integer linear programming method, inputting the equipment load characteristics and line capacity data, setting the constraint conditions as the balance between line transmission capacity and equipment load, and adjusting the layout in combination with the line transmission characteristics. The power system flow calculation method is used to analyze the voltage loss, input the line impedance parameters and load data, set the node voltage range to 0.95 to 1.05 times the rated value, calculate the line voltage loss and optimize the line layout, and generate a line optimization path diagram;

[0078] System design submodule: Based on the line optimization path map, the geographic information system modeling method is used to carry out detailed planning of the line layout in the scenic area. When the geographic information system modeling method is executed, the line optimization path map data and the scenic area map data need to be input, and the line path is remapped using the vector data format. The line buffer range is set to 30 meters to avoid overlap. When integrating the specific location and capacity of the distribution station and the transformer, the location allocation algorithm is used to input the geographical location and capacity data of the distribution station and the transformer, and the capacity matching range is set to 95% to 105% of the rated value. The capacity adaptation model is used to integrate the data. When connecting and adapting the various areas of the power supply system, the network topology optimization algorithm is used to input the line connection data and equipment location data, and the connection node weight is set to the line impedance value. The minimum spanning tree is generated using the Prim algorithm, and the overall layout diagram of the power supply line and equipment is drawn to generate the power supply system design diagram.

[0079] See also Figure 2 ,The fault prediction and diagnosis module includes a topology analysis submodule, an anomaly detection submodule and a fault diagnosis submodule, among which:

[0080] Topology analysis submodule: Based on the power supply system design diagram, the spatial position and connection relationship of each node in the power grid are extracted, the path information between nodes is counted and the data characteristics of the connection path are analyzed. The characteristics of key nodes in the power grid are analyzed by summarizing and classifying the connection data, key connection areas are extracted, and high-importance nodes are identified to generate power grid topology data.

[0081] Anomaly detection submodule: Based on the grid topology data, the isolation forest algorithm is used to extract the time series fluctuation information of current and voltage, perform continuous sampling and classification of the fluctuation range, and statistically analyze the data distribution of abnormal points based on the change trend of the fluctuation amplitude. The abnormal fluctuation points of voltage or current are located by point-by-point comparison, the fluctuation range is defined, and the potential abnormal area is confirmed to generate abnormal fluctuation area data;

[0082] Fault diagnosis submodule: Based on the abnormal fluctuation area data, the node and connection relationship data of the abnormal area are extracted, and the structural characteristics of the abnormal nodes are confirmed by node-by-node analysis. The key fault points are gradually located by combining the topological distribution and fluctuation characteristics in the area. The risk nodes and potential problem areas are identified by decomposing the connection area, and the fault point diagnosis results are generated;

[0083] Topology analysis submodule: Based on the power supply system design diagram, a depth-first search algorithm is used to extract the spatial position and connection relationship of each node in the power grid. The execution of the depth-first search algorithm includes inputting the node connection matrix in the power supply system design diagram, setting the starting node as the first connection point in the design diagram, and recursively traversing the connection path of the node. At the same time, the spatial position of the node is recorded and mapped. When performing statistics on the path information between nodes, the shortest path algorithm is used to calculate the path length between nodes, input the node connection relationship and path weight, set the path weight as the line length value, and use the Dijkstra algorithm to calculate the shortest path between all node pairs one by one and record the path length and node order. By summarizing and classifying the connection data, a clustering algorithm is used to analyze the characteristics of key nodes in the power grid. The node connectivity and node weight data are input, and the number of clusters is set to 3. The K-means clustering algorithm is used to extract the center of each type of node and classify and record them, extract key connection areas and identify high-importance nodes, and generate power grid topology data;

[0084] Anomaly detection submodule: Based on the power grid topology data, the isolation forest algorithm is used to extract the current and voltage time series fluctuation information. When executing the isolation forest algorithm, the current and voltage time series data must be input. The number of forest trees is set to 100, the sample sampling rate is 256, and the tree structure is constructed by random partitioning. The data points in each sequence are split and the anomaly score is calculated. When continuous sampling and classification of the fluctuation interval are performed, the sliding window method is used to extract the fluctuation interval data. The time series is input, the window size is set to 10 minutes, and the step length is 1 minute. The data distribution of the abnormal points is statistically analyzed in combination with the fluctuation amplitude change trend. The density estimation algorithm is used to perform statistical analysis on the distribution of the abnormal points. The kernel function is set to Gaussian kernel and the bandwidth parameter is 0.5. The local density of each abnormal point is calculated and the density peak point is marked. The voltage or current fluctuation abnormal point is located by point-by-point comparison, the fluctuation range is delineated, and the potential abnormal area is confirmed, and the abnormal fluctuation area data is generated;

[0085] Fault diagnosis submodule: Based on the abnormal fluctuation area data, the node importance algorithm is used to extract the nodes and connection relationship data of the abnormal area. When the node importance algorithm is executed, the node connection matrix of the abnormal fluctuation area needs to be input, and the importance calculation standard is set as connectivity and betweenness. The betweenness centrality method is used to calculate the number of times each node appears in all shortest paths. When performing node-by-node analysis to confirm the structural characteristics of the abnormal nodes, the node decomposition method is used to perform a hierarchical analysis of the topological connection structure of the nodes. The node connection matrix and node weights are input, and the high-weight nodes are gradually decomposed and their associated nodes are recorded. When the key fault points are gradually located in combination with the topological distribution and fluctuation characteristics in the region, a multi-level diagnosis algorithm is used for layer-by-layer screening and confirmation. The topological distribution map and fluctuation characteristic data are input, and the diagnosis level is set to 3 layers. High-risk nodes are marked layer by layer and their connection relationships are analyzed. Risk nodes and potential problem areas are identified by decomposing the connection areas to generate fault point diagnosis results.

[0086] Isolation forest algorithm, according to the formula:

[0087]

[0088] Where: S′(x, t, w, σ) is the improved anomaly score, h(x) is the average path length of sample x, c(n) is the normalized path length, Δt is the time interval between fluctuation points, σ ​​is the standard deviation of the fluctuation amplitude, μ is the mean of the fluctuation sequence, and w 1 、w 2 and w 3 is the weight coefficient, x is the current sample point;

[0089] Execution process: First, for the voltage or current fluctuation point x in the distribution network, the isolation forest algorithm is used to calculate its average path length h(x), and the normalized path length c(n) is calculated based on the data scale n to generate the isolation forest term Quantify the degree of isolation of the fluctuation points, and then calculate the time interval between adjacent time points based on the time series data of the fluctuation points, Δt=t i -t i-1 , the time interval is normalized by the standard deviation σ of the fluctuation amplitude, and the normalized time interval term is obtained Again, by statistically analyzing the mean μ of the distribution network time series fluctuation and the deviation of the fluctuation point |x-μ|, the mean deviation term is obtained by normalization: Finally, the weights w 1 ,w 2 ,w 3 Perform weighted summation to generate anomaly score S′(x, t, w, σ). The weight value is determined by optimizing the training data, and the grid search method is used to tune w in real fluctuation data. 1 ,w 2 ,w 3The value of is taken as the goal to minimize the anomaly detection error, and finally a comprehensive score is generated to determine whether the fluctuation point is abnormal, providing accurate fluctuation anomaly detection and regional demarcation basis for the intelligent planning of distribution networks in tourist attractions.

[0090] See also Figure 2 The emergency response plan module includes an impact analysis submodule, an isolation operation submodule, and an emergency guidance submodule, among which:

[0091] Impact analysis submodule: Based on the fault point diagnosis results, the connection nodes and regional boundaries of the fault point are extracted, and the transmission characteristics between nodes related to the fault point are analyzed. The impact range is defined in combination with the connection characteristics and the connection structure of the affected area is analyzed. The main nodes that may be affected are identified through layer-by-layer transmission analysis, and the fault impact range data is generated;

[0092] Isolation operation submodule: Based on the fault impact range data, extract the key nodes and path data in the connection area, analyze the isolation operation path of the fault area, adjust the power supply load by gradually isolating the connection path, re-plan the power distribution of the remaining area based on the load data, and generate isolation and reload operation steps;

[0093] Emergency Guidance Submodule: Based on the isolation and overload operation steps, extract the implementation area data, disassemble and analyze each step, organize the key points of operation and execution conditions, combine the distribution of access and load to form an implementation document, and generate an emergency operation guide;

[0094] Impact analysis submodule: Based on the fault point diagnosis results, the shortest path algorithm is used to extract the connection nodes and regional boundaries of the fault point. When the shortest path algorithm is executed, the fault point connection relationship data and node weights need to be input, and the path weight is set as the line impedance value. The Dijkstra algorithm is used to gradually calculate the shortest path from the fault point to each connection node and record the path sequence. When analyzing the transmission characteristics between nodes related to the fault point, the power flow calculation method is used, the node connection relationship matrix and load current data are input, and the voltage range is set to 0.95 to 1.05 times the rated value. The node injection power balance equation is used to gradually calculate the transmission characteristics between nodes. Power and current distribution, combined with connection characteristics to define the impact range and analyze the connection structure of the affected area, a clustering algorithm is used to classify and analyze the node impact range, the node transmission characteristic data and node identification are input, the number of clusters is set to 3, the K-means clustering algorithm is used to extract the boundaries of various impact areas and record the boundary nodes, and a layer-by-layer transmission analysis is performed to identify the main nodes that may be affected. The depth-first search algorithm is used to recursively analyze the affected nodes, the node connection matrix and the initial fault point identification are input, the recursive depth is set to the total number of nodes, all visited nodes are recorded and the key nodes are marked, and the fault impact range data is generated;

[0095] Isolation operation submodule: Based on the fault impact range data, a topological cutting algorithm is used to extract key nodes and path data in the connection area. When executing the topological cutting algorithm, the connection relationship matrix of the fault area needs to be input, and the cutting weight is set as the line transmission power. The minimum cut algorithm is used to gradually separate the connection paths and mark the key nodes. When performing isolation operation path analysis in the fault area, the network flow model analysis method is used, and the key node and line path data are input. The flow threshold is set to 80% of the rated capacity. The maximum flow minimum cut principle is used to calculate the set of paths that need to be isolated. When combining the load data to re-plan the power distribution of the remaining area, a load balancing optimization algorithm is used, and the load data of the remaining area and the equipment capacity data are input. The load balancing target is set to a difference of no more than 10% between the maximum load and the minimum load. The linear programming method is used to redistribute the load, and the isolation and reload operation steps are generated;

[0096] Emergency guidance submodule: Based on the isolation and overloading operation steps, the task decomposition algorithm is used to extract the implementation area data. When the task decomposition algorithm is executed, the operation steps and the implementation area range need to be input. The decomposition granularity is set to a single operation task. The depth-first traversal method is used to gradually decompose the operation steps into independent tasks and record the task sequence. When decomposing and analyzing each step, the rule matching algorithm is used to parse the execution conditions of the steps. The operation steps and condition rule sets are input. The string matching method is used to retrieve the condition matching results one by one and record the parsing results. When sorting out the operation points and execution conditions, the logical sorting algorithm is used to classify and integrate the parsing results. The parsing results and task sequence are input, and the tasks are reordered in order of priority and the final operation points are recorded. The implementation document is formed by combining the distribution of pathways and loads to generate an emergency operation guide.

[0097] See also Figure 2 The cost and benefit analysis module includes a cost calculation submodule, a benefit estimation submodule and a benefit comparison submodule, among which:

[0098] Cost calculation submodule: Based on the power supply system design and emergency operation guide, extract the data of equipment installation location and line laying length, summarize the various costs of operating power consumption and scheduling by counting the maintenance cycle and consumables consumption, integrate and calculate all cost data during the operation period, and generate the total operation cost of the power grid;

[0099] Benefit estimation submodule: Based on the total operating cost of the power grid, random forest is used to extract the power saving data brought by power supply optimization and energy-saving measures. By statistically analyzing the potential loss data reduced by dispatch optimization and accident emergency treatment, the economic values ​​of various savings are accumulated, the actual total benefits of energy saving and management optimization are extracted, and the power grid saving benefits are generated;

[0100] Benefit comparison submodule: Based on the power grid saving benefits, extract the cost data and benefit data of each optimization scheme, and perform step-by-step comparative calculations between the two. By analyzing the proportion of different schemes in savings and costs, extract the cost-benefit ratio data of each scheme and perform multi-dimensional grouping and induction to generate a cost-benefit data table;

[0101] Cost calculation submodule: Based on the power supply system design drawing and emergency operation guide, a linear regression algorithm is used to extract the equipment installation location and line laying length data. When the linear regression algorithm is executed, the coordinate data of the equipment installation location and the coordinate data of the path points of the line laying need to be input. The target variable is set as the total length of the line, and the regression coefficient is set as 1. The distance between the coordinate points is calculated using the least squares method and accumulated as the total length. When the maintenance cycle and consumables consumption are counted, a weighted average algorithm is used. The historical record data of the maintenance cycle and consumables consumption is input, and the weight is set as the maintenance frequency. The weighted calculation formula is used to count and summarize the consumption of different consumables. When summarizing the various costs of operating power consumption and scheduling, the item-by-item classification algorithm is used to organize the cost data. The power consumption and scheduling cost data during the operation period are input, and the classification standards are set as time periods and cost categories. The cost data is classified, summarized and summed using the grouping statistical method. All cost data during the operation period are integrated and calculated to generate the total operating cost of the power grid;

[0102] Benefit estimation submodule: Based on the total operating cost of the power grid, the random forest algorithm is used to extract the power saving data brought by power supply optimization and energy-saving measures. When the random forest algorithm is executed, the power supply optimization plan and historical power data need to be input. The number of forest trees is set to 100, and the sample sampling rate is set to 80%. The decision tree set is generated by random partitioning and feature selection. The power saving of the input data is predicted and counted. When the potential loss data reduced by statistical scheduling optimization and accident emergency treatment is used, the Monte Carlo simulation method is used to estimate the loss. The scheduling optimization parameters and accident emergency treatment data are input. The number of simulations is set to 1000 times. The random number generator is used to simulate and calculate the potential losses under different parameter combinations for multiple times and output the average value. When accumulating the economic values ​​of various savings, the summing algorithm is used to input the economic values ​​of energy-saving measures, power saving and loss reduction, and the total economic saving value is calculated by item-by-item addition. The actual total amount of benefits from energy saving and management optimization is extracted to generate the power grid saving benefits.

[0103] Benefit comparison submodule: Based on the power grid saving benefit, a hierarchical regression algorithm is used to extract the cost data and benefit data of each optimization scheme. When the hierarchical regression algorithm is executed, the cost data and benefit data of the optimization scheme need to be input, and the stratification variable is set as the optimization measure type. The hierarchical regression model is used to perform group regression analysis on the data of different optimization schemes and extract the regression coefficient. When performing step-by-step comparison calculations between the two, a multi-objective optimization algorithm is used to input cost and benefit data, and the optimization goal is set to minimize the difference between cost and benefit. The cost and benefit comparison value of each scheme is calculated using the linear programming method. When analyzing the relationship between the proportion of different schemes in savings and costs, the ratio analysis method is used to input the cost and saving benefit comparison data of each scheme, and the percentage calculation method is used to count the proportion of cost and benefit in each scheme. The cost-benefit ratio data of each scheme is extracted and multi-dimensional grouping and induction are performed to generate a cost-benefit data table.

[0104] Random forest, according to the formula:

[0105]

[0106] Where: E′ is the energy saving benefit of the improved power grid, S j is the power saving benefit brought by the j-th power supply optimization, M j is the saving benefit generated by the jth energy-saving measure, L j is the potential loss reduced by the jth dispatch optimization and accident emergency treatment, P j is the rated power of the jth device, T j is the operating time of the jth device, E j is the energy efficiency ratio of the jth device, W 1 , W 2 , W 3 and W 4 is the weight coefficient, m is the number of measures or the number of data items analyzed;

[0107] Implementation process: First, for each power supply optimization scheme in the tourist attractions, calculate its power saving benefit S j By analyzing the energy consumption differences after the power supply scheme is adjusted, the economic savings brought by the optimization are obtained. Then, according to the implementation of energy-saving measures, the direct energy-saving benefit M of each measure is extracted. j , combined with the use of energy-saving equipment and the effect of technical transformation, quantify its economic benefits, and further analyze the dispatch optimization and accident emergency treatment data to extract the potential loss L reduced by each measure j , taking the economic losses reduced by rapid response and scientific scheduling as the benefit index of scheduling optimization, and at the same time, extracting the rated power P of the equipment running in the scenic spot j , running time T j And energy efficiency ratio E j, calculate the energy-saving benefit item of equipment operation, and finally use the weight coefficient W 1 ,W 2 ,W 3 ,W 4 The power supply optimization benefit, energy-saving benefit, dispatch optimization benefit and equipment operation energy-saving benefit are weighted and summed respectively. The weight value is determined by training historical data, using grid search method and cross-validation method. The optimization goal is to minimize the model error. The benefits of all optimization measures and energy-saving equipment are accumulated to generate the comprehensive saving benefit E′ of the distribution network of tourist attractions.

[0108] See also Figure 2 ,The system optimization and update module includes a performance monitoring submodule, a real-time adjustment submodule and an update generation submodule, among which:

[0109] Performance monitoring submodule: Based on the cost-effectiveness data table, extract system performance indicators and related operation data, dynamically collect the load distribution of power grid lines and the power consumption of equipment operation, conduct performance comparison analysis in different time periods, extract data on equipment aging or performance degradation to identify optimization areas, and generate performance monitoring data;

[0110] Real-time adjustment submodule: Based on performance monitoring data, it extracts the specific distribution locations of high-load lines and aging equipment, redistributes line loads, optimizes equipment performance by gradually adjusting the operating parameters of related equipment, and performs local adjustments and optimization operations based on load and equipment data to generate adjustment optimization results;

[0111] Update generation submodule: Based on the adjustment and optimization results, extract the optimized equipment configuration and line allocation data, dynamically update the overall configuration file of the power grid, and integrate and generate implementable system configuration and power grid optimization files by summarizing the new parameters of equipment operation and load distribution data, and generate configuration update plans;

[0112] Performance monitoring submodule: Based on the cost-effectiveness data table, a distributed monitoring algorithm is used to extract system performance indicators and operation data. When the distributed monitoring algorithm is executed, the grid line load data and equipment operation power consumption data need to be input. The monitoring cycle is set to 5 minutes. The node monitor is used to collect the real-time load of each line and record the load change value. By dynamically collecting the load distribution of the grid line and the equipment operation power consumption, the sliding window technology is used to analyze the collected data in real time. The window size is set to 10 minutes and the step length is 1 minute. The data in the window is gradually updated and the load change rate is calculated. When performing performance comparison analysis in different time periods, the time series decomposition algorithm is used to input the performance data within the time period, set the decomposition mode to additive decomposition, separate the trend, cycle and random components, and perform differential calculation on the separated trend components to analyze the performance changes. When extracting the data of equipment aging or performance degradation to identify the optimization area, the multivariate regression analysis method is used, the equipment performance indicators and operation time data are input, and the regression variable is set to the equipment aging rate. The least squares method is used to calculate the aging trend and extract the aging equipment to generate performance monitoring data;

[0113] Real-time adjustment submodule: Based on performance monitoring data, a load balancing algorithm is used to extract the distribution locations of high-load lines and aging equipment. When executing the load balancing algorithm, the load data of the high-load line and the operating status of the equipment need to be input. The load balancing target is set to a difference between the maximum load and the minimum load of less than 5%. The linear optimization method is used to redistribute the load and record the adjustment results. When redistributing the line load, a node voltage adjustment algorithm is used, and the line node voltage and current data are input. The voltage adjustment range is set to 0.95 to 1.05 times the rated value. The Newton-Raphson method is used to calculate the adjusted voltage distribution. When optimizing the equipment performance by gradually adjusting the operating parameters of the relevant equipment, an equipment parameter optimization algorithm is used, and the equipment operating parameters and load data are input. The optimization target is set to a power factor close to 1 for the equipment. The gradient descent method is used to gradually adjust the equipment parameters and output the optimized parameter values. Local adjustment and optimization operations are performed in combination with the load and equipment data to generate adjustment optimization results.

[0114] Update generation submodule: Based on the adjustment and optimization results, a dynamic configuration update algorithm is used to extract the optimized equipment configuration and line allocation data. When the dynamic configuration update algorithm is executed, the equipment operation parameters and line load distribution data need to be input. The update cycle is set to 24 hours. The incremental update method is used to record the change values ​​of new parameters and data. When dynamically updating the overall configuration file of the power grid, the file generation algorithm is used to input the dynamically updated parameters and allocation data. The template filling method is used to embed the new data into the configuration template and generate a new configuration file. When summarizing the new parameters and load distribution data of the equipment operation, the hierarchical aggregation algorithm is used to integrate the data from different regions, the geographical area identification of the equipment and load data is input, and the aggregation rules are set to summarize layer by layer by region to generate new configuration files and power grid optimization files. The feasible system configuration and power grid optimization files are integrated to generate a configuration update plan.

[0115] A method for intelligent planning of a distribution network for a tourist attraction, the method for intelligent planning of a distribution network for a tourist attraction is performed based on the above-mentioned intelligent planning system for a distribution network for a tourist attraction, and comprises the following steps:

[0116] Step 1: According to the activity schedule and historical power usage pattern of tourist attractions, extract the power usage records of each time period, calculate the load variation range within the time period, count the high-frequency and low-frequency fluctuation characteristics in the time series, obtain the load demand data of each time period, and generate the load demand characteristics of the tourist attractions;

[0117] Step 2: Based on the load demand characteristics of the scenic spot, extract the power supply device data and load distribution characteristics in the area, calculate the power supply load distribution value of each area, optimize the path layout of the power grid line by combining the load distribution value and line transmission characteristics, integrate the installation location and capacity information of the power supply device, and generate the power supply line layout data;

[0118] Step 3: Based on the power supply line layout data, extract the node locations and connection characteristics of the power grid, count the topological distribution between the lines and divide different path areas, analyze the key nodes and load transmission paths of the power grid, combine the isolation forest method to detect the time series fluctuation data of current and voltage, mark the nodes and areas where abnormal fluctuations may occur, and combine these nodes to make correlation judgments on potential fault points and generate fault-related area distribution;

[0119] Step 4: Based on the distribution of fault-related areas, extract the node and line information of high-risk areas, calculate the load priority and isolation order of each node, formulate isolation steps and backup load activation order according to the priority, integrate isolation steps and load distribution operations to generate operation guidelines, and generate emergency isolation and load distribution operation plans;

[0120] Step 5: Based on the emergency isolation and load distribution operation plan, extract the regional load change data and operation parameters after each operation, combine the random forest method to extract the characteristic distribution of the load data and related costs during the operation, calculate the saved operation cost and spare resource consumption during the dispatch process, conduct a comparative analysis of the cost and saving data, and generate cost and saving comparison data;

[0121] Step 6: Based on the cost and savings comparison data, extract the operating performance indicators and parameter distribution under different configurations, count the changes in power load distribution in each area, calculate the configuration optimization value of the high-load area, reallocate the load transmission and line paths of the power grid, adjust the dynamic configuration of the power grid and generate load transmission update documents, and generate distribution configuration update plans.

[0122] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A distribution network intelligent planning system for tourist attractions, characterized in that: The system comprises: Grid demand forecasting module: Based on the activity schedule and historical power usage patterns of tourist attractions, through time series data analysis, estimate the power demand fluctuations within a specific time period, calculate the expected power demand, and generate a load forecast table for tourist attractions; Power supply scheme design module: according to the load forecast table of the scenic spot, analyze the location and capacity of the distribution station and transformer, integrate the adaptation path of the power grid line, optimize the line layout, and obtain the power supply system design diagram; Fault prediction and diagnosis module: Based on the power supply system design diagram, the isolation forest algorithm is used to extract the topological data of the power grid, analyze the key nodes and connections of the power grid structure, detect abnormal fluctuations in current and voltage, identify potential fault sources and risk areas, and obtain fault point diagnosis results; Emergency response plan module: based on the fault point diagnosis results, analyze the impact range and emergency level of the fault point, formulate quick operation steps for fault point isolation and power reload, and create emergency operation guidelines; Cost and benefit analysis module: conduct economic model analysis on the power supply system design and emergency operation guide, use random forest to calculate the total cost and expected savings of system operation, compare the cost-benefit ratio of different solutions, and output a cost-benefit report; System optimization and update module: Based on the cost-benefit report, through the power grid monitoring data, continuously track the system performance indicators, adjust the power grid configuration and management strategy according to the real-time feedback, and generate the configuration update plan.

2. The intelligent planning system for distribution network of tourist attractions according to claim 1, characterized in that: The power grid demand forecasting module includes a time series analysis submodule, a power demand estimation submodule and a load table generation submodule, wherein: Time series analysis submodule: Based on the activity schedule and historical power usage patterns of tourist attractions, the time series segmentation and statistical analysis of the activity schedule are carried out to extract the power usage fluctuation pattern within the time interval, extract the high and low frequency band division and periodic characteristics of power usage, and generate a power time series feature table; Power demand estimation submodule: Based on the power time series characteristic table, the power demand in each time period is estimated by time division, and the peak demand and valley demand are dynamically calculated in combination with the specific power demand change characteristics of activity schedules and holidays to generate power demand fluctuation data; Load table generation submodule: Based on the power demand fluctuation data, the load data in different scenic spots are divided, the load change values ​​within the time period are summarized and partitioned and classified, and the load demand timetable of the scenic spots is divided to generate a load forecast table for the scenic spots.

3. The intelligent planning system for distribution network of tourist attractions according to claim 1, characterized in that: The power supply scheme design module includes a load analysis submodule, a line optimization submodule and a system design submodule, wherein: Load analysis submodule: Based on the scenic spot load forecast table, extract the data of load characteristics and power distribution equipment demand in the scenic spot area, calculate the number and location range of distribution stations in combination with load demand, allocate transformer capacity and plan installation location, and generate a power distribution equipment configuration list; Line optimization submodule: Based on the distribution equipment configuration list, the grid adaptation path is screened and the line layout plan is optimized, the matching characteristics of the line and the equipment are integrated to calculate the path, the layout is adjusted in combination with the line transmission characteristics, and a line optimization path diagram is generated; System design submodule: Based on the line optimization path map, detailed planning of line layout in the scenic area is carried out, the specific location and capacity of the distribution station and transformer are integrated, the connection and adaptation of various areas of the power supply system are carried out, the overall layout diagram of the power supply lines and equipment is drawn, and the power supply system design diagram is generated.

4. The intelligent planning system for distribution network of tourist attractions according to claim 1, characterized in that: The fault prediction and diagnosis module includes a topology analysis submodule, an anomaly detection submodule and a fault diagnosis submodule, wherein: Topology analysis submodule: Based on the power supply system design diagram, the spatial position and connection relationship of each node in the power grid are extracted, the path information between nodes is counted and the data characteristics of the connection path are analyzed, the characteristics of key nodes in the power grid are analyzed by summarizing and classifying the connection data, key connection areas are extracted and high-importance nodes are identified, and power grid topology data is generated; Abnormal detection submodule: Based on the grid topology data, the isolation forest algorithm is used to extract the time series fluctuation information of current and voltage, and the fluctuation interval is continuously sampled and classified. The data distribution of abnormal points is statistically analyzed in combination with the change trend of the fluctuation amplitude. The abnormal fluctuation points of voltage or current are located by point-by-point comparison, the fluctuation range is defined, and the potential abnormal area is confirmed, and the abnormal fluctuation area data is generated; Fault diagnosis submodule: Based on the abnormal fluctuation area data, the node and connection relationship data of the abnormal area are extracted, and the structural characteristics of the abnormal nodes are confirmed by node-by-node analysis. The key fault points are gradually located in combination with the topological distribution and fluctuation characteristics in the area. The risk nodes and potential problem areas are identified by decomposing the connection areas, and the fault point diagnosis results are generated.

5. The intelligent planning system for distribution network of tourist attractions according to claim 4, characterized in that: The isolation forest algorithm is based on the formula: Where: S′(x,t,w,σ) is the improved anomaly score, h(x) is the average path length of sample x, c(n) is the normalized path length, Δt is the time interval between fluctuation points, σ ​​is the standard deviation of the fluctuation amplitude, μ is the mean of the fluctuation sequence, w1, w2 and w3 are weight coefficients, and x is the current sample point.

6. The intelligent planning system for distribution network of tourist attractions according to claim 1, characterized in that: The emergency response plan module includes an impact analysis submodule, an isolation operation submodule and an emergency guidance submodule, wherein: Impact analysis submodule: Based on the fault point diagnosis results, the connection nodes and regional boundaries of the fault point are extracted, and the transmission characteristics between nodes related to the fault point are analyzed. The impact range is defined in combination with the connection characteristics and the connection structure of the affected area is analyzed. The main nodes that may be affected are identified through layer-by-layer transmission analysis, and the fault impact range data is generated; Isolation operation submodule: based on the fault impact range data, extract key nodes and path data in the connection area, perform isolation operation path analysis on the fault area, adjust the power supply load by gradually isolating the connection path, re-plan the power distribution of the remaining area in combination with the load data, and generate isolation and reload operation steps; Emergency guidance submodule: Based on the isolation and overload operation steps, the implementation area data is extracted, each step is disassembled and analyzed, and an implementation document is formed by organizing the key points of operation and execution conditions, combined with the distribution of access and load, to generate an emergency operation guide.

7. The intelligent planning system for distribution network of tourist attractions according to claim 1, characterized in that: The cost and benefit analysis module includes a cost calculation submodule, a benefit estimation submodule and a benefit comparison submodule, wherein: Cost calculation submodule: Based on the power supply system design drawing and emergency operation guide, extract the data of equipment installation location and line laying length, summarize the various costs of operating power consumption and scheduling by counting the maintenance cycle and consumables consumption, integrate and calculate all cost data during operation, and generate the total operation cost of the power grid; Benefit estimation submodule: Based on the total operating cost of the power grid, random forest is used to extract the power saving data brought by power supply optimization and energy-saving measures, and the potential loss data reduced by statistical scheduling optimization and accident emergency treatment are accumulated to accumulate the economic values ​​of various savings, extract the total actual benefits of energy saving and management optimization, and generate power grid saving benefits; Benefit comparison submodule: Based on the power grid saving benefit, the cost data and benefit data of each optimization scheme are extracted, and a step-by-step comparative calculation is performed between the two. By analyzing the proportion of different schemes in savings and costs, the cost-benefit ratio data of each scheme is extracted and multi-dimensional grouping and summarization are performed to generate a cost-benefit data table.

8. The intelligent planning system for distribution network of tourist attractions according to claim 7, characterized in that: The random forest, according to the formula: Where: E′ is the energy saving benefit of the improved power grid, S j is the power saving benefit brought by the j-th power supply optimization, M j is the saving benefit generated by the jth energy-saving measure, L j is the potential loss reduced by the jth dispatch optimization and accident emergency treatment, P j is the rated power of the jth device, T j is the operating time of the jth device, E j is the energy efficiency ratio of the jth equipment, W1, W2, W3 and W4 are weight coefficients, and m is the number of measures or the number of data items analyzed.

9. The intelligent planning system for distribution network of tourist attractions according to claim 1, characterized in that: The system optimization and update module includes a performance monitoring submodule, a real-time adjustment submodule and an update generation submodule, wherein: Performance monitoring submodule: Based on the cost-effectiveness data table, extract system performance indicators and related operation data, dynamically collect the load distribution of the power grid lines and the power consumption of the equipment operation, conduct performance comparison analysis in different time periods, extract data on equipment aging or performance degradation to identify optimization areas, and generate performance monitoring data; Real-time adjustment submodule: Based on the performance monitoring data, the specific distribution locations of high-load lines and aging equipment are extracted to redistribute line loads, and the equipment performance is optimized by gradually adjusting the operating parameters of related equipment. Local adjustment and optimization operations are performed in combination with load and equipment data to generate adjustment optimization results; Update generation submodule: Based on the adjustment and optimization results, extract the optimized equipment configuration and line allocation data, dynamically update the overall configuration file of the power grid, and integrate and generate implementable system configuration and power grid optimization files by summarizing the new parameters of equipment operation and load distribution data, and generate a configuration update plan.

10. A method for intelligent planning of distribution networks for tourist attractions, characterized in that: The intelligent planning system for distribution network of tourist attractions according to any one of claims 1 to 9 comprises the following steps: Step 1: According to the activity schedule and historical power usage pattern of tourist attractions, extract the power usage records of each time period, calculate the load variation range within the time period, count the high-frequency and low-frequency fluctuation characteristics in the time series, obtain the load demand data of each time period, and generate the load demand characteristics of the tourist attractions; Step 2: Based on the load demand characteristics of the scenic spot, extract the power supply device data and load distribution characteristics in the area, calculate the power supply load distribution value of each area, optimize the path layout of the power grid line by combining the load distribution value and the line transmission characteristics, integrate the installation location and capacity information of the power supply device, and generate the power supply line layout data; Step 3: Based on the power supply line layout data, extract the node locations and connection characteristics of the power grid, count the topological distribution between the lines and divide different path areas, analyze the key nodes and load transmission paths of the power grid, combine the isolation forest method to detect the time series fluctuation data of current and voltage, mark the nodes and areas where abnormal fluctuations may occur, and combine these nodes to make association judgments on potential fault points, and generate fault-related area distribution; Step 4: Based on the distribution of the fault-related areas, extract the node and line information of the high-risk areas, calculate the load priority and isolation order of each node, formulate the isolation steps and the order of enabling the backup load according to the priority, integrate the isolation steps and the load distribution operation to generate an operation guide, and generate an emergency isolation and load distribution operation plan; Step 5: Based on the emergency isolation and load distribution operation plan, extract the regional load change data and operation parameters after each operation, combine the random forest method to extract the characteristic distribution of the load data and related costs during the operation, calculate the saved operation cost and spare resource consumption during the dispatching process, conduct a comparative analysis of the cost and saving data, and generate cost and saving comparison data; Step 6: Based on the cost and savings comparison data, extract the operating performance indicators and parameter distribution under different configurations, count the changes in power load distribution in each area, calculate the configuration optimization value of the high-load area, reallocate the load transmission and line paths of the power grid, adjust the dynamic configuration of the power grid and generate load transmission update documents, and generate distribution configuration update plans.

Images