Intelligent power distribution system and power distribution method

Abstract

The invention relates to the technical field of electric energy distribution, in particular to an intelligent power distribution system and a power distribution method.According to the intelligent power distribution system and method, power flow paths and load density distribution are defined by collecting load distribution values of nodes of a power distribution network and calculating power flow, line transmission loss and load value density of load transmission paths among the nodes; according to the method, a data foundation is laid for subsequent dynamic adjustment, partition node power requirements and transmission path characteristics are integrated by adopting an alternating direction multiplier method, a dynamic partition model is established, a global problem is decomposed into a local optimization sub-problem in a partition optimization process, partition boundary adjustment and power balanced distribution are realized through iterative calculation, and a partition node is optimized. A greedy algorithm is adopted to screen load priorities and demanded quantities, the electric energy guarantee capability of high-priority loads is improved, through adjustment of a voltage offset range and dynamic adjustment of reactive power compensation equipment, the influence of voltage fluctuation on power balance between partitions is effectively reduced, the voltage stability is improved, and the energy consumption and the power loss are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of electric energy distribution, and in particular to an intelligent power distribution system and a power distribution method. Background Art

[0002] The field of electric energy distribution technology focuses on achieving efficient utilization and stable supply of electric energy resources, improving the safety, reliability and flexibility of the distribution system through intelligent means, and meeting users' needs for power quality and economic efficiency.

[0003] The purpose of intelligent power distribution system and distribution method is to optimize the operation of power distribution network, improve the efficiency of power distribution, reduce energy consumption and loss, and enhance the adaptability of power distribution system to changes in external environment, aiming to realize efficient dispatch and allocation of power resources, and at the same time enhance the real-time monitoring of the operation status of power distribution network and intelligent fault handling capabilities, so as to achieve efficient, safe and economical operation of power distribution network.

[0004] Existing technologies have deficiencies in real-time monitoring and optimal scheduling of the operating status of distribution networks. It is difficult to accurately extract high-load areas, resulting in deviations in the positioning of high-load nodes and the calibration of coverage ranges, causing uneven load distribution. Relying on a single linear prediction model, it is unable to effectively capture complex fluctuation trends, which can easily lead to inaccurate predictions of the load peak range, causing problems such as excess or insufficient allocation of resources. Partition structures are mostly static designs, and it is difficult to adjust boundaries according to real-time needs, resulting in inefficient power flow and frequent local overloads. In addition, the voltage regulation means are relatively lagging, and it is impossible to dynamically optimize voltage fluctuations and power balance, resulting in local path overloads and overall power distribution imbalances, affecting the operating safety and economy of the distribution network. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an intelligent power distribution system and a power distribution method.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: an intelligent power distribution system comprises:

[0007] Load status positioning module: obtains the load distribution value of the distribution network node through the data collected by the electric meter, calculates the power flow, line transmission loss and load value density of the load transmission path between the nodes according to the distribution value, compares the load value density between the nodes, and extracts the area according to the density difference, calibrates the load node position and coverage range at the same time, and generates the load density distribution result;

[0008] Regional load prediction module: Based on the load density distribution results, the time series of the load nodes are calculated by time series decomposition method, and the change amplitude and change direction of the load fluctuation are extracted according to the trend change, the peak range of the future high-load area is determined, and the high-load area prediction result is generated;

[0009] Dynamic partition construction module: Based on the prediction results of the high-load area, the alternating direction multiplier method is used to extract the node power flow path and screen the path by comparing the transmission power between partition nodes with the partition demand. At the same time, the partition boundary is adjusted in combination with the power demand distribution to establish a dynamic partition structure model;

[0010] Priority power control module: Based on the dynamic partition structure model, a greedy algorithm is used to screen the load priority and demand in the partition, determine whether the demand of the high-priority partition meets the load distribution standard, adjust the power distribution according to the priority partition, and calculate the power flow direction and flow across the partition to obtain the partition priority power distribution plan;

[0011] Voltage state adjustment module: based on the partition priority power allocation scheme, compares the voltage offset range in the partition and the operating output of the reactive compensation equipment, adjusts the reactive power compensation value in the partition, calculates the impact of the power balance between partitions on the voltage fluctuation through "power flow calculation", outputs the voltage adjustment range, and generates a voltage dynamic adjustment value;

[0012] Global power coordination module: Based on the voltage dynamic adjustment value, compare the cross-partition power flow and regional transmission capacity, screen the paths that exceed the transmission capacity and adjust the power flow, optimize the cross-partition load power flow distribution, and adjust the power flow balance value of the entire network to establish a global load power coordination value.

[0013] As a further solution of the present invention, the load state positioning module includes a load flow calculation submodule, a load density difference extraction submodule and a high load calibration submodule, wherein:

[0014] Load flow calculation submodule: obtain the load distribution value of the distribution network node through the data collected by the electric meter, extract the load transmission path between the nodes according to the distribution value, calculate the path power flow by analyzing the power value between the path nodes, and extract the power transmission value in the line, gradually accumulate to form a complete transmission flow data, calculate the line power loss in combination with the transmission distance between the nodes, and generate the power flow and transmission loss results;

[0015] Load density difference extraction submodule: based on the power flow and transmission loss results, extract the unit density value of each node load, identify the area with large density difference by comparing the density value difference between nodes one by one, extract the nodes and paths in the density mutation area, analyze the load distribution characteristics in the area, and generate the load density difference regional distribution;

[0016] High load calibration submodule: based on the regional distribution of load density differences, locate the core nodes in the density mutation area, extract the load coverage around the core nodes, analyze the load distribution level within the coverage area, establish a regional load distribution model, calibrate the node positions and coverage areas with high load density, and generate load density distribution results;

[0017] The load density distribution result includes the high-load node location, the high-load area coverage and the line transmission power flow table.

[0018] As a further solution of the present invention, the regional load prediction module includes a load time trend calculation submodule, a load fluctuation amplitude analysis submodule and a high load area prediction submodule, wherein:

[0019] Load time trend calculation submodule: based on the load density distribution result, extract the time series data of high-load nodes, analyze the fluctuation characteristics in the time series through the time series decomposition method, decompose the directional characteristics of load changes, extract the fluctuation range and directional trend within the time period, build a dynamic change model of time trend, and generate time series change trend results;

[0020] Load fluctuation amplitude analysis submodule: based on the time series change trend results, extract the change value of the time series fluctuation amplitude, compare the fluctuation value range in the historical load data, extract the abnormal fluctuation points and time periods, combine the load fluctuation values ​​of the time nodes, gradually calculate the change range of the time series fluctuation amplitude, and generate the load fluctuation amplitude analysis results;

[0021] High-load area prediction submodule: Based on the load fluctuation amplitude analysis results, the fluctuation amplitude range of the high-load area is extracted, and the high-load fluctuation trend in the future time period is analyzed by combining the load distribution model in the fluctuation range, and the peak range and fluctuation characteristics of the high-load area are gradually predicted to generate the high-load area prediction results;

[0022] The high-load area prediction result includes load fluctuation amplitude, load change trend direction and future load peak range.

[0023] As a further solution of the present invention, the dynamic partition construction module includes a transmission power screening submodule, a partition boundary adjustment submodule and a dynamic partition model establishment submodule, wherein:

[0024] Transmission power screening submodule: based on the prediction results of the high-load area, extract the transmission power values ​​between the partition nodes, analyze the power distribution characteristics on the transmission path according to the transmission power values, screen the power flow paths one by one and count the power values, compare the differences between the power values ​​and the partition demand, select the paths with higher power than the demand, analyze the node correlation, and generate a high-power path data set;

[0025] Partition boundary adjustment submodule: based on the high-power path data set, extract the power demand distribution in the high-power path coverage area, analyze the relationship between the node power demand and the high-power path distribution, extract regional nodes with prominent power demand, adjust the partition boundary, calculate the partition adjustment range in combination with the power balance principle, redefine and adjust the partition boundary, and generate the power optimized partition boundary;

[0026] Dynamic partition model establishment submodule: Based on the power optimization partition boundary, the alternating direction multiplier method is used to integrate the power requirements and transmission path characteristics of the partition nodes, extract the power matching relationship between the nodes according to the power requirements and the transmission path characteristics, analyze the power distribution characteristics between the nodes and paths in the dynamic partition, build a dynamic partition matching model of the nodes and paths, and extract the dynamic adjustment rules in the model to generate a dynamic partition structure model;

[0027] The dynamic partition structure model includes partition boundary positions, partition power flow paths and partition power balance parameter sets.

[0028] As a further solution of the present invention, the difference between the power value and the partition demand is compared, and the difference between the power value and the partition demand is quantitatively analyzed by analyzing the relationship between the total power transmitted by the path and the total demand of the regional node one by one. According to the transmission direction of the path power, the partition power demand data covered by the path is gradually counted, and the relationship between the path power and the demand is compared to screen out the path that meets the partition demand or the power exceeds the demand;

[0029] The alternating direction multiplier method is used to decompose the overall partition optimization problem into multiple independently computable sub-problems, gradually adjust the power distribution and path load characteristics of the partition nodes, and adjust the partition structure at the same time;

[0030] The dynamic adjustment rules in the extraction model are extracted by continuously analyzing the relationship between the changes in the power transmission path and the node demand distribution. When the power load in the dynamic partition is unbalanced, the adjustment rules automatically trigger real-time adjustment operations of the path and partition range.

[0031] As a further solution of the present invention, the priority power control module includes a load priority screening submodule, a power distribution adjustment submodule and a partition power flow calculation submodule, wherein:

[0032] Load priority screening submodule: based on the dynamic partition structure model, extract partition nodes and their corresponding load demands, gradually screen the priority attributes of the nodes and mark high-priority nodes, analyze the power demand distribution of the high-priority node group, and generate high-priority load distribution data;

[0033] Power allocation adjustment submodule: Based on the high-priority load distribution data, a greedy algorithm is used to extract the load demand value according to the partition priority, and the power allocation ratio is gradually adjusted in combination with the power transmission relationship between partitions. The demand for cross-regional power flow between partitions is counted, and the power allocation value of the high-priority partition is adjusted and matched with the power demand, so as to generate a high-priority power allocation plan;

[0034] Partition power flow calculation submodule: Based on the high-priority power allocation scheme, extract the power transmission path between nodes across partitions, gradually analyze the power transmission flow and statistically analyze the power distribution, analyze the node coverage and flow direction of cross-partition power flow according to the power distribution, and analyze the power flow relationship in combination with the power allocation scheme to generate a partition priority power allocation scheme;

[0035] The partition priority power allocation scheme includes a high priority partition power input amount, power flow distribution between partitions and a power allocation parameter set.

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

[0037]

[0038] in: is the updated power allocation value of partition i, is the current power allocation value of partition i, λ is the power adjustment coefficient, D i is the high priority load demand value of partition i, T i is the current power transfer amount of partition i, ψ i is the power matching priority coefficient of partition i, f j is the load transmission path weight of partition j, ω j is the power consumption efficiency coefficient of partition j, Normalized value of load deviation of all high priority partitions under multi-weight correction.

[0039] As a further solution of the present invention, the voltage state regulation module includes a voltage offset analysis submodule, a reactive power adjustment submodule and a voltage dynamic calculation submodule, wherein:

[0040] Voltage offset analysis submodule: based on the partition priority power allocation scheme, extract the node voltage operation data in the partition, calculate the voltage offset value of the node one by one, mark the nodes whose offset value exceeds the range by comparing the node voltage offset with the standard offset range in the partition, extract the operating output value of the reactive compensation device associated with the offset node, analyze the matching situation of the current output power of the device and the node offset, and generate voltage offset range data;

[0041] Reactive power adjustment submodule: based on the voltage offset range data, extract the reactive power demand of the offset node, calculate the available power output of the compensation device node by node, match the node offset value and adjust the reactive power compensation amount, redistribute the reactive power output in the partition, optimize the compensation distribution of the offset node, and generate reactive power compensation value data;

[0042] Voltage dynamic calculation submodule: Based on the reactive power compensation value data, extract the power balance distribution value between partitions, use "power flow calculation" to calculate the impact of power balance on node voltage fluctuations by partition, adjust the voltage value output after the voltage offset of each partition, calculate the overall voltage adjustment range of the partition, and extract the dynamic voltage distribution value within the adjustment range to generate a voltage dynamic adjustment value;

[0043] The voltage dynamic adjustment value includes a voltage offset range, a reactive power compensation adjustment parameter and a voltage balance value between partitions.

[0044] As a further solution of the present invention, the global power coordination module includes a cross-partition capacity comparison submodule, a power flow adjustment submodule and a full-network power balancing submodule, wherein:

[0045] Cross-district capacity comparison submodule: based on the voltage dynamic adjustment value, extract the power flow value of the cross-district transmission path, analyze the relationship between the flow value and the regional transmission capacity path by path, screen the path where the power flow exceeds the transmission capacity, analyze the power distribution characteristics of the overloaded path node by node, count the transmission load status of the nodes and paths, and generate cross-district overload path data;

[0046] Power flow adjustment submodule: based on the cross-partition overload path data, extract the power flow characteristics in the overload path, adjust the path load ratio by redistributing the cross-partition power flow, redistribute the high-flow path load in combination with the node power demand and transmission capacity, adjust the power flow of the overload path and calculate the flow distribution value, and generate the cross-partition power flow distribution;

[0047] The whole network power balancing submodule: based on the cross-partition power flow distribution, extract the power flow data of the whole network partition, optimize the power balance by adjusting the load distribution node by node, gradually optimize the power load distribution between nodes in combination with the cross-partition power flow data, rebalance the power flow of the whole network, and generate a global load power coordination value;

[0048] The global load power coordination value includes cross-partition power flow distribution, network-wide transmission capacity adjustment value and network-wide power balancing parameter.

[0049] An intelligent power distribution method is implemented based on the above intelligent power distribution system, comprising the following steps:

[0050] Step 1: Obtain the load distribution value of the distribution network node through the data collected by the electric meter, calculate the power flow, line transmission loss and load value density of the load transmission path between the nodes according to the distribution value, and associate the load data of the density area with the node position and coverage range by comparing the density of the load values ​​between the nodes, and calibrate the high-load nodes and coverage range at the same time to generate the load density distribution result;

[0051] Step 2: Based on the load density distribution result, extract the load time series of the high-load node, perform correlation calculation through the fluctuation amplitude and fluctuation direction of the load time series data, compare the fluctuation range in the historical load data, determine the load peak range of the future load area, analyze the load trend change in the area and generate the high-load area prediction result in combination with the peak range;

[0052] Step 3: Based on the prediction results of the high-load area, the alternating direction multiplier method is used to extract the high-power path by comparing the transmission power value between the partition nodes and the partition demand, and the partition boundary parameters are dynamically adjusted according to the power demand. The partition boundary position is correlated with the node power flow path, the power balance distribution within the partition and the partition boundary range are optimized, and a dynamic partition structure model is generated;

[0053] Step 4: Based on the dynamic partition structure model, a greedy algorithm is used to screen the priority value of the load within the partition, and the load demand within the partition is compared and adjusted with the load distribution standard. The partition priority power distribution plan is obtained by calculating the power flow direction and flow across the partitions;

[0054] Step 5: Based on the partition priority power allocation scheme, compare the voltage offset range between partitions and the operating output value of the reactive power compensation equipment, calculate the impact range of the power balance between partitions on the voltage fluctuation through "power flow calculation", optimize the voltage state by adjusting the reactive power compensation value, combine the distribution parameters of cross-partition power flow and transmission capacity, screen out the paths that exceed the capacity and adjust the power flow direction, and generate a global load power coordination value.

[0055] Compared with the prior art, the advantages and positive effects of the present invention are:

[0056] 1. In the present invention, by collecting the load distribution values ​​of the distribution network nodes and calculating the power flow, line transmission loss and load value density of the load transmission path between the nodes, the power flow path and load density distribution are clarified, laying a data foundation for subsequent dynamic adjustment;

[0057] 2. In the present invention, the alternating direction multiplier method is used to integrate the power requirements of partition nodes and the characteristics of the transmission path, and a dynamic partition model is established. In the partition optimization process, the global problem is decomposed into local optimization sub-problems, and the partition boundary adjustment and power balanced distribution are achieved through iterative calculation, which overcomes the limitation of the traditional static partition that responds slowly to load changes and improves the real-time and dynamic adaptability of power distribution;

[0058] 3. In the present invention, a greedy algorithm is used to screen load priorities and demands, thereby improving the power guarantee capability of high-priority loads. By adjusting the voltage offset range and dynamically adjusting the reactive power compensation equipment, the impact of voltage fluctuations on power balance between partitions is effectively reduced, voltage stability is improved, and energy consumption and power loss are reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is a system flow chart of the present invention;

[0060] Figure 2 It is a schematic diagram of the system framework of the present invention;

[0061] Figure 3 It is a schematic diagram of the method steps of the present invention. DETAILED DESCRIPTION

[0062] 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.

[0063] See also Figure 1 The present invention provides a technical solution: an intelligent power distribution system comprising:

[0064] Load status positioning module: obtains the load distribution value of the distribution network node through the data collected by the electric meter, calculates the power flow, line transmission loss and load value density of the load transmission path between the nodes according to the distribution value, compares the load value density between the nodes, and extracts the area according to the density difference, calibrates the load node position and coverage range at the same time, and generates the load density distribution result;

[0065] Regional load prediction module: Based on the load density distribution results, the time series decomposition method is used to calculate the trend change of the time series of the load nodes. The change amplitude and direction of the load fluctuation are extracted according to the trend change, the peak range of the future high-load area is determined, and the high-load area prediction results are generated;

[0066] Dynamic partition construction module: Based on the prediction results of high-load areas, the alternating direction multiplier method is used to extract and screen the node power flow paths by comparing the transmission power between partition nodes with the partition demand. At the same time, the partition boundaries are adjusted in combination with the power demand distribution to establish a dynamic partition structure model.

[0067] Priority power control module: Based on the dynamic partition structure model, the greedy algorithm is used to screen the load priority and demand in the partition, determine whether the demand of the high-priority partition meets the load distribution standard, adjust the power distribution according to the priority partition, and calculate the power flow direction and flow across partitions to obtain the partition priority power distribution plan;

[0068] Voltage state adjustment module: Based on the partition priority power allocation scheme, compare the voltage offset range within the partition and the operating output of the reactive compensation equipment, adjust the reactive power compensation value within the partition, calculate the impact of the power balance between partitions on voltage fluctuations through "power flow calculation", output the voltage adjustment range, and generate the voltage dynamic adjustment value;

[0069] Global power coordination module: Based on the dynamic voltage adjustment value, compare the cross-partition power flow and regional transmission capacity, screen the paths that exceed the transmission capacity and adjust the power flow, optimize the cross-partition load power flow distribution, and adjust the power flow balance value of the entire network to establish a global load power coordination value.

[0070] See also Figure 2 The load state positioning module includes a load flow calculation submodule, a load density difference extraction submodule and a high load calibration submodule, wherein:

[0071] Load flow calculation submodule: obtain the load distribution value of the distribution network node through the data collected by the electric meter, extract the load transmission path between the nodes according to the distribution value, calculate the path power flow by analyzing the power value between the path nodes, and extract the power transmission value in the line, gradually accumulate to form a complete transmission flow data, calculate the line power loss in combination with the transmission distance between the nodes, and generate the power flow and transmission loss results;

[0072] Load density difference extraction submodule: Based on the power flow and transmission loss results, the unit density value of each node load is extracted. By comparing the density value differences between nodes one by one, the areas with large density differences are identified, the nodes and paths in the density mutation area are extracted, the load distribution characteristics in the area are analyzed, and the regional distribution of load density differences is generated;

[0073] High-load calibration submodule: Based on the regional distribution of load density differences, locate the core nodes in the density mutation area, extract the load coverage around the core nodes, analyze the load distribution level within the coverage area, establish a regional load distribution model, calibrate the location and coverage of nodes with high load density, and generate load density distribution results;

[0074] Load flow calculation submodule: Based on the load distribution value of the distribution network node collected by the meter, the path power flow analysis method is used to analyze the load transmission path between the nodes. By analyzing the power value between the path nodes, the path power flow between the nodes is calculated, and the power transmission value in the line is gradually accumulated by cumulative addition operation. The accumulation operation is performed in sequence according to the power value of the path node, and the power transmission value of each line in the path is accumulated one by one to form complete transmission flow data. The line power loss is calculated by combining the transmission distance between the nodes using the transmission loss calculation method, and the power flow value and its loss value are output to generate the power flow and transmission loss results;

[0075] Load density difference extraction submodule: Based on the power flow and transmission loss results, the load density value of the node is extracted using the unit load density calculation method, and the load density value of the node is calculated one by one using the density calculation method. The unit density value of each node is extracted, and the density value difference of the node is compared pair by pair by pair by pair by pair by pair. In the comparison method, the density difference is calculated by node by node comparison, and the area with large density difference is identified. The nodes and paths in the density mutation area are extracted, and the load distribution characteristics in the area are analyzed by the path topology analysis method. The connection relationship of the path nodes is analyzed one by one using the path by path iterative analysis method. The analysis content includes the connection strength and power flow between nodes, and the regional distribution of load density difference is generated;

[0076] High-load calibration submodule: Based on the regional distribution of load density differences, the regional core node positioning algorithm is used to locate the core nodes in the density mutation area. The load coverage around the core nodes is extracted by calculating the connection path density and coverage of the core nodes. The load distribution level within the coverage range is analyzed using the distribution hierarchy algorithm. The hierarchy is achieved by arranging the load values ​​of the nodes in the region in descending order and grouping them. The grouping is based on the preset density difference threshold between the load values. When the difference in the node load values ​​exceeds the threshold, a new group is formed. The hierarchical relationship is determined by iterative grouping, and a regional load distribution model is established. The node position and coverage of high load density are calibrated by analyzing the distribution model to generate the load density distribution result.

[0077] Among them, the load density distribution results include the location of high-load nodes, the coverage of high-load areas and the line transmission power flow table.

[0078] See also Figure 2 The regional load prediction module includes a load time trend calculation submodule, a load fluctuation amplitude analysis submodule and a high load area prediction submodule, among which:

[0079] Load time trend calculation submodule: Based on the load density distribution results, extract the time series data of high-load nodes, analyze the fluctuation characteristics in the time series through the time series decomposition method, decompose the directional characteristics of load changes, extract the fluctuation range and directional trend within the time period, build a dynamic change model of the time trend, and generate the time series change trend results;

[0080] Load fluctuation amplitude analysis submodule: Based on the time series change trend results, extract the change value of the time series fluctuation amplitude, compare the fluctuation value range in the historical load data, extract abnormal fluctuation points and time periods, combine the load fluctuation values ​​of the time nodes, gradually calculate the change range of the time series fluctuation amplitude, and generate the load fluctuation amplitude analysis results;

[0081] High-load area prediction submodule: Based on the load fluctuation range analysis results, the fluctuation range of the high-load area is extracted, and combined with the load distribution model in the fluctuation range, the high-load fluctuation trend in the future time period is analyzed, and the peak range and fluctuation characteristics of the high-load area are gradually predicted to generate the high-load area prediction results;

[0082] Load time trend calculation submodule: Based on the load density distribution results, extract the time series data of high-load nodes, use the time series decomposition algorithm to analyze the fluctuation characteristics in the time series, decompose the time series data, extract the periodic components, trend components and random components in the load change, use the trend item in the decomposition model to extract the directional characteristics, use the directional calculation formula to determine the directional characteristics of the load change, and match and adjust the extracted directional characteristics. Combined with the fluctuation range of the time series, gradually analyze the change trend within the time period, use the dynamic model fitting method to build a dynamic change model of the time trend, use the fitting calculation to analyze the dynamic change characteristics of the data points one by one, and generate the time series change trend results;

[0083] Load fluctuation amplitude analysis submodule: Based on the time series change trend results, the fluctuation amplitude of the time series is analyzed using the fluctuation amplitude extraction method. The maximum change point and extreme value point in the time series are extracted, and the fluctuation amplitude calculation formula is used to calculate the change range. The fluctuation amplitude is compared point by point in combination with the fluctuation value range in the historical load data, and the abnormal fluctuation points and the corresponding time periods are extracted. The distribution law of the abnormal fluctuation points is verified by the historical data backtracking method, and the change range of the time series fluctuation amplitude is calculated step by step in combination with the time nodes to generate the load fluctuation amplitude analysis results;

[0084] High-load area prediction submodule: Based on the load fluctuation amplitude analysis results, the regional prediction analysis algorithm is used to extract the fluctuation amplitude range of the high-load area. The peak change trend of the load value is determined by analyzing the load distribution model in the fluctuation range. The high-load fluctuation trend in the future time period is gradually analyzed by combining the time series prediction method. The linear regression method is used to fit and verify the future fluctuation trend. The peak range of the high-load area is gradually estimated by the iterative prediction method. The distribution model is corrected in combination with the fluctuation amplitude characteristics. The node power change law within the fluctuation range is gradually analyzed to generate the high-load area prediction results.

[0085] Among them, the high-load area prediction results include load fluctuation amplitude, load change trend direction and future load peak range.

[0086] See also Figure 2 The dynamic partition building module includes a transmission power screening submodule, a partition boundary adjustment submodule and a dynamic partition model building submodule, wherein:

[0087] Transmission power screening submodule: Based on the prediction results of the high-load area, the transmission power values ​​between the partition nodes are extracted, the power distribution characteristics on the transmission path are analyzed according to the transmission power values, the power flow paths are screened one by one and the power values ​​are counted, the difference between the power values ​​and the partition demand is compared, and the paths with higher power than the demand are selected, and the node correlation is analyzed to generate a high-power path data set;

[0088] Partition boundary adjustment submodule: Based on the high-power path data set, extract the power demand distribution in the high-power path coverage area, analyze the relationship between node power demand and high-power path distribution, extract regional nodes with prominent power demand, adjust the partition boundary, calculate the partition adjustment range based on the power balance principle, redefine and adjust the partition boundary, and generate the power-optimized partition boundary;

[0089] Dynamic partition model building submodule: Based on the power optimization partition boundary, the alternating direction multiplier method is used to integrate the power requirements of the partition nodes and the transmission path characteristics. According to the power requirements and transmission path characteristics, the power matching relationship between nodes is extracted, and the power distribution characteristics between nodes and paths in the dynamic partition are analyzed. The dynamic partition matching model of nodes and paths is constructed, and the dynamic adjustment rules in the model are extracted to generate a dynamic partition structure model.

[0090] Transmission power screening submodule: Based on the prediction results of high-load areas, the transmission power values ​​between partition nodes are extracted, and the power distribution characteristics on the transmission path are analyzed using the power distribution analysis algorithm. The power transmission values ​​between path nodes are analyzed using the path step-by-step iterative calculation method and the power flow of each path is recorded. The power flow paths are screened one by one using the power threshold screening method. During the screening process, the paths are classified and counted according to the preset power upper and lower limits. The difference between the path power value and the partition demand is compared by combining the demand difference analysis method. The path with higher power than the demand is selected according to the priority order of the power value difference. The node correlation analysis algorithm is combined to analyze the correlation characteristics between the path nodes. The connection strength of the nodes is quantitatively analyzed by the degree of correlation between the paths to generate a high-power path data set.

[0091] Partition boundary adjustment submodule: Based on the high-power path data set, the regional coverage extraction algorithm is used to extract the power demand distribution in the high-power path coverage area. The power distribution model is combined to perform a detailed analysis of the demand distribution in the path area. The demand hotspot extraction algorithm is used to extract regional nodes with prominent power demand. The distribution characteristics of high-demand nodes are analyzed and regional boundaries are marked. The partition adjustment range is calculated in combination with the power balance calculation method. The partition range is adjusted by gradually expanding or shrinking the boundary operation. During the adjustment process, the path coverage rate is used as the adjustment basis to correct the boundary nodes and recalibrate the regional range. The partition boundaries are redefined and adjusted to generate power-optimized partition boundaries.

[0092] Dynamic partition model establishment submodule: Based on the power optimization partition boundary, the alternating direction multiplier method is used to integrate the power demand and transmission path characteristics of the partition nodes. The node relationship between partitions is decomposed by combining the power demand matrix and the path characteristic matrix. The power matching relationship between nodes is extracted by iteratively updating the constraint variables. The power distribution characteristics between nodes and paths in the dynamic partition are gradually analyzed in combination with the path connection characteristics. The path allocation rules are used to quantify the power transmission weights in the node connection. The node distribution model is corrected by adjusting the path transmission weights. The dynamic partition matching model is constructed in combination with the demand changes and path characteristics of the nodes in the dynamic partition. The dynamic adjustment rules between nodes are gradually extracted using the model parameters, and the dynamic partition structure model is generated in combination with the rules.

[0093] The dynamic partition structure model includes the partition boundary location, partition power flow path and partition power balance parameter set.

[0094] Compare the difference between the power value and the partition demand. The difference between the power value and the partition demand is quantitatively analyzed by analyzing the relationship between the total power transmitted by the path and the total demand of the regional nodes one by one. According to the transmission direction of the path power, the partition power demand data covered by the path is gradually counted, and the relationship between the path power and the demand is compared to screen out the path that meets the partition demand or exceeds the power demand;

[0095] By using the alternating direction multiplier method, the overall partition optimization problem is decomposed into multiple independently computable sub-problems, and the power distribution and path load characteristics of the partition nodes are gradually adjusted, and the partition structure is adjusted at the same time;

[0096] The dynamic adjustment rules within the extraction model are continuously analyzed and extracted from the relationship between the changes in the power transmission path and the node demand distribution. When the power load in the dynamic partition is unbalanced, the adjustment rules automatically trigger real-time adjustment operations on the path and partition range.

[0097] See also Figure 2 The priority power control module includes a load priority screening submodule, a power distribution adjustment submodule and a partition power flow calculation submodule, among which:

[0098] Load priority screening submodule: Based on the dynamic partition structure model, it extracts partition nodes and corresponding load demands, gradually screens the priority attributes of nodes and marks high-priority nodes, analyzes the power demand distribution of high-priority node groups, and generates high-priority load distribution data;

[0099] Power allocation adjustment submodule: Based on high-priority load distribution data, a greedy algorithm is used to extract load demand values ​​according to partition priorities. At the same time, the power allocation ratio is gradually adjusted in combination with the power transmission relationship between partitions. The demand for cross-regional power flow between partitions is counted, and the power allocation value of high-priority partitions is adjusted to match the power demand, generating a high-priority power allocation plan.

[0100] Partition power flow calculation submodule: Based on the high-priority power allocation plan, extract the power transmission path between nodes across partitions, gradually analyze the power transmission flow and statistically analyze the power distribution. According to the power distribution, analyze the node coverage and flow direction of the cross-partition power flow. At the same time, analyze the power flow relationship in combination with the power allocation plan to generate a partition priority power allocation plan.

[0101] Load priority screening submodule: Based on the dynamic partition structure model, the partition nodes and the corresponding load demands are extracted, and the load demand values ​​of the partition nodes are screened for priority attributes using the node priority classification algorithm. The screening process classifies the nodes by defining the priority threshold range, and gradually screens whether the load demand of the nodes exceeds the high priority threshold. The nodes that meet the conditions are marked as high priority nodes. The power demand distribution of the high priority node group is analyzed using the demand distribution analysis method. The total demand value and local distribution characteristics of the high priority node group are gradually calculated in combination with the spatial position and demand value of the nodes to generate high priority load distribution data.

[0102] Power allocation adjustment submodule: Based on high-priority load distribution data, a greedy algorithm is used to gradually extract and analyze the load demand values ​​of high-priority partitions. The greedy algorithm arranges the demands of high-priority nodes in descending order to give priority to the nodes with the largest demand, and gradually allocates power to each node until the current power demand is met or the power resources are exhausted. The power allocation ratio is gradually adjusted in combination with the power transmission relationship between partitions. The power allocation matrix is ​​used to record the power allocation values ​​between partitions and gradually optimize the cross-regional power flow parameters in the matrix. The total demand for cross-regional power transmission is counted. The power allocation value of the high-priority partition is adjusted to match the remaining power value in the demand matrix, and the power allocation state of the high-priority node is gradually corrected to generate a high-priority power allocation plan.

[0103] Partition power flow calculation submodule: Based on the high-priority power allocation scheme, extract the power transmission path between nodes across partitions, use the power flow analysis method to gradually analyze the power transmission flow on the path, accumulate the power flow of each path in node order to calculate the total flow, combine the distribution of path power flow to gradually count the power distribution between nodes, use the power flow analysis method to analyze the power flow direction node by node, and determine the node coverage of power flow by directional classification of path weights in the flow distribution results. Combined with the power allocation scheme, analyze the cross-partition power flow relationship, gradually adjust the coverage and weight value of the power flow path, and generate a partition priority power allocation scheme;

[0104] The partition priority power allocation scheme includes the power input amount of the high priority partition, the power flow distribution between partitions and the power allocation parameter set.

[0105] Greedy algorithm, according to the formula:

[0106]

[0107] in: is the updated power allocation value of partition i, is the current power allocation value of partition i, λ is the power adjustment coefficient, D iis the high priority load demand value of partition i, T i is the current power transfer amount of partition i, ψ i is the power matching priority coefficient of partition i, f j is the load transmission path weight of partition j, ω j is the power consumption efficiency coefficient of partition j, Normalized value of load deviation of all high priority partitions under multi-weight correction;

[0108] Execution process: First, initialize the current power allocation value P of the partition i old and power adjustment coefficient λ, the load demand value D of partition i is calculated according to the high priority load distribution data of the intelligent power distribution system i and the current power transfer amount T i , then introduce the power matching priority coefficient ψ i , by analyzing the importance and urgency of the partition load demand, dynamically generating and adjusting the priority of different partitions in power allocation, and calculating the load transmission path weight f j , which is determined by counting the transmission frequency and power flow of the paths between partitions, reflecting the impact of the power transmission path on the overall distribution, and then determining the power consumption efficiency coefficient ω j , calculated based on the power utilization efficiency and power loss rate of the partition equipment, reflecting the effectiveness of the partition in power use, and the power deviation term (D j -T j ) is corrected, and then the normalized value adjustment value is calculated, and the normalized adjustment value is multiplied by the power adjustment coefficient λ and superimposed on the initial power allocation value P i old On the power distribution value P i new , to achieve dynamic power allocation between high-priority partitions, meet load requirements and optimize cross-regional power flows.

[0109] See also Figure 2 The voltage state regulation module includes a voltage offset analysis submodule, a reactive power adjustment submodule and a voltage dynamic calculation submodule, wherein:

[0110] Voltage offset analysis submodule: Based on the partition priority power allocation scheme, extract the node voltage operation data in the partition, calculate the voltage offset value of the node one by one, mark the nodes whose offset value exceeds the range by comparing the node voltage offset with the standard offset range in the partition, extract the operating output value of the reactive compensation equipment associated with the offset node, analyze the matching situation of the current output power of the equipment and the node offset, and generate voltage offset range data;

[0111] Reactive power adjustment submodule: Based on the voltage offset range data, the reactive power demand of the offset node is extracted, the available power output of the compensation device is calculated node by node, the node offset value is matched and the reactive power compensation amount is adjusted, the reactive power output in the partition is reallocated, the compensation distribution of the offset node is optimized, and the reactive power compensation value data is generated;

[0112] Voltage dynamic calculation submodule: Based on the reactive power compensation value data, extract the power balance distribution value between partitions, use "power flow calculation" to calculate the impact of power balance on node voltage fluctuations by partition, adjust the output voltage value after the voltage offset of each partition, calculate the overall voltage adjustment range of the partition, and extract the dynamic voltage distribution value within the adjustment range to generate the voltage dynamic adjustment value;

[0113] Voltage offset analysis submodule: Based on the partition priority power allocation scheme, extract the voltage operation data of the nodes in the partition, use the voltage offset calculation method to calculate the voltage offset value of the node one by one, calculate the difference between the node voltage and the standard voltage in the partition item by item and record the result, use the offset range comparison algorithm to compare the voltage offset value one by one within the partition range, mark the nodes whose offset values ​​exceed the standard range, and gradually extract the associated reactive compensation equipment operation output values ​​based on the marked nodes, analyze the current output power of the equipment using the equipment power output data and the voltage demand matching rule of the offset node, record the matching of the equipment output power and the node offset value, and generate voltage offset range data;

[0114] Reactive power adjustment submodule: Based on the voltage offset range data, the reactive power demand of the offset node is extracted, and the available power output of the compensation device is calculated node by node using the reactive power calculation method. The remaining available power value is determined by subtracting the maximum output capacity of the device from the current operating parameters of the device. The matching algorithm is used to match the remaining power of the device with the voltage offset value of the node node by node. The node offset compensation value and the device output are coordinated by gradually adjusting the matching output. The reactive power output in the partition is reallocated in combination with the node reactive demand data. The working status of the compensation device in the partition is adjusted and the output data of each device is recorded to generate reactive power compensation value data.

[0115] Voltage dynamic calculation submodule: Based on the reactive power compensation value data, the power balance distribution value between partitions is extracted. The power balance on voltage fluctuation analysis algorithm is used to calculate the fluctuation effect of power balance on node voltage partition by partition. The fluctuation impact analysis calculates the voltage change step by step by superimposing the power change value on each node. The output voltage value in the partition is recalculated in combination with the voltage offset of each partition after adjustment. The overall voltage adjustment range is calculated partition by partition using the dynamic voltage adjustment method. The dynamic voltage distribution value within the adjustment range is extracted node by node and the results are recorded to generate the voltage dynamic adjustment value.

[0116] The voltage dynamic adjustment value includes the voltage offset range, reactive power compensation adjustment parameters and the voltage balance value between partitions.

[0117] See also Figure 2 The global power coordination module includes a cross-partition capacity comparison submodule, a power flow adjustment submodule, and a full-network power balancing submodule, among which:

[0118] Cross-district capacity comparison submodule: Based on the voltage dynamic adjustment value, extract the power flow value of the cross-district transmission path, analyze the relationship between the flow value and the regional transmission capacity path by path, filter out the paths where the power flow exceeds the transmission capacity, analyze the power distribution characteristics of the overloaded path node by node, count the transmission load status of the nodes and paths, and generate cross-district overload path data;

[0119] Power flow adjustment submodule: Based on the cross-partition overload path data, the power flow characteristics in the overload path are extracted, the path load ratio is adjusted by redistributing the cross-partition power flow, the high-flow path load is redistributed in combination with the node power demand and transmission capacity, the power flow of the overload path is adjusted, and the flow distribution value is calculated to generate the cross-partition power flow distribution;

[0120] The whole network power balancing submodule: Based on the cross-partition power flow distribution, extract the power flow data of the whole network partition, optimize the power balance by adjusting the load distribution node by node, and gradually optimize the power load distribution between nodes in combination with the cross-partition power flow data, rebalance the power flow of the whole network, and generate the global load power coordination value;

[0121] Cross-district capacity comparison submodule: Based on the voltage dynamic adjustment value, the power flow value of the cross-district transmission path is extracted, and the relationship between the power flow value and the regional transmission capacity is analyzed path by path using the path flow analysis method. By comparing the path flow data item by item and screening out the paths where the power flow exceeds the transmission capacity, the power distribution characteristics of the overloaded path are analyzed node by node using the node power distribution analysis method. The transmission power value and occupancy ratio of each node on the path are counted by accumulating node by node, and the path transmission load status is gradually calculated in combination with the node load data in the overloaded path to generate cross-district overload path data;

[0122] Power flow adjustment submodule: Based on the cross-partition overload path data, the power flow characteristics in the overload path are extracted, and the power flow distribution algorithm is used to redistribute the overload path load. The algorithm optimizes the power transmission path by analyzing the power flow direction and flow distribution ratio between nodes in the path, and adjusts the path load ratio by redistributing the cross-partition power flow. The load distribution state of the high-flow path is gradually adjusted in combination with the node power demand and transmission capacity. The adjusted path flow value is calculated using the flow iterative distribution method and the distribution results are recorded to generate the cross-partition power flow distribution;

[0123] The whole network power balancing submodule: Based on the cross-partition power flow distribution, the power flow data of the whole network partition is extracted, and the load distribution of each node is adjusted by using the global load balancing algorithm. The algorithm adjusts the power load distribution between nodes by prioritizing the power load of the whole network nodes and gradually allocating the remaining power flow. The power load relationship between nodes is gradually optimized by combining the dynamic update of the cross-partition power flow data. The power load superposition calculation method is used to rebalance the power flow distribution between the whole network partitions to generate a global load power coordination value.

[0124] Among them, the global load power coordination value includes cross-partition power flow distribution, network-wide transmission capacity adjustment value and network-wide power balancing parameters.

[0125] See also Figure 3 , an intelligent power distribution method, the intelligent power distribution method is performed based on the above intelligent power distribution system, comprising the following steps:

[0126] Step 1: Obtain the load distribution value of the distribution network node through the data collected by the electric meter, calculate the power flow, line transmission loss and load value density of the load transmission path between the nodes according to the distribution value, and associate the load data of the density area with the node position and coverage range by comparing the density of the load values ​​between the nodes, and calibrate the high-load nodes and coverage range at the same time to generate the load density distribution result;

[0127] Step 2: Based on the load density distribution results, extract the load time series of high-load nodes, perform correlation calculations through the fluctuation amplitude and fluctuation direction of the load time series data, compare the fluctuation range in the historical load data, determine the load peak range of the future load area, analyze the load trend changes in the area, and generate the high-load area prediction results based on the peak range;

[0128] Step 3: Based on the prediction results of high-load areas, the alternating direction multiplier method is used to extract high-power paths by comparing the transmission power values ​​between partition nodes and the partition demand. The partition boundary parameters are dynamically adjusted according to the power demand. The partition boundary position is correlated with the node power flow path, the power balance distribution within the partition and the partition boundary range are optimized, and a dynamic partition structure model is generated.

[0129] Step 4: Based on the dynamic partition structure model, a greedy algorithm is used to screen the priority value of the load within the partition. At the same time, the load demand within the partition is compared and adjusted with the load distribution standard. The partition priority power distribution plan is obtained by calculating the power flow direction and flow across partitions.

[0130] Step 5: Based on the partition priority power allocation scheme, compare the voltage offset range between partitions and the operating output value of the reactive power compensation equipment, calculate the impact range of the power balance between partitions on the voltage fluctuation through "power flow calculation", optimize the voltage state by adjusting the reactive power compensation value, combine the distribution parameters of cross-partition power flow and transmission capacity, screen out the paths that exceed the capacity and adjust the power flow direction to generate the global load power coordination value.

[0131] 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. An intelligent power distribution system, characterized in that: The system comprises: Load status positioning module: obtains the load distribution value of the distribution network node through the data collected by the electric meter, calculates the power flow, line transmission loss and load value density of the load transmission path between the nodes according to the distribution value, compares the load value density between the nodes, and extracts the area according to the density difference, calibrates the load node position and coverage range at the same time, and generates the load density distribution result; Regional load prediction module: Based on the load density distribution results, the time series of the load nodes are calculated by time series decomposition method, and the change amplitude and change direction of the load fluctuation are extracted according to the trend change, the peak range of the future high-load area is determined, and the high-load area prediction result is generated; Dynamic partition construction module: Based on the prediction results of the high-load area, the alternating direction multiplier method is used to extract the node power flow path and screen the path by comparing the transmission power between partition nodes with the partition demand. At the same time, the partition boundary is adjusted in combination with the power demand distribution to establish a dynamic partition structure model; Priority power control module: Based on the dynamic partition structure model, a greedy algorithm is used to screen the load priority and demand in the partition, determine whether the demand of the high-priority partition meets the load distribution standard, adjust the power distribution according to the priority partition, and calculate the power flow direction and flow across the partition to obtain the partition priority power distribution plan; Voltage state adjustment module: based on the partition priority power allocation scheme, compare the voltage offset range in the partition and the operating output of the reactive compensation equipment, adjust the reactive power compensation value in the partition, calculate the impact of the power balance between partitions on the voltage fluctuation through "power flow calculation", output the voltage adjustment range, and generate the voltage dynamic adjustment value; Global power coordination module: Based on the voltage dynamic adjustment value, compare the cross-partition power flow and regional transmission capacity, screen the paths that exceed the transmission capacity and adjust the power flow, optimize the cross-partition load power flow distribution, and adjust the power flow balance value of the entire network to establish a global load power coordination value.

2. The intelligent power distribution system according to claim 1, characterized in that: The load density distribution result includes high-load node locations, high-load area coverage, and line transmission power flow table.

3. The intelligent power distribution system according to claim 1, characterized in that: The high load area prediction result includes load fluctuation amplitude, load change trend direction and future load peak range.

4. The intelligent power distribution system according to claim 1, characterized in that: The dynamic partition construction module includes a transmission power screening submodule, a partition boundary adjustment submodule and a dynamic partition model establishment submodule, wherein: Transmission power screening submodule: based on the prediction results of the high-load area, extract the transmission power values ​​between the partition nodes, analyze the power distribution characteristics on the transmission path according to the transmission power values, screen the power flow paths one by one and count the power values, compare the differences between the power values ​​and the partition demand, select the paths with higher power than the demand, analyze the node correlation, and generate a high-power path data set; Partition boundary adjustment submodule: based on the high-power path data set, extract the power demand distribution in the high-power path coverage area, analyze the relationship between the node power demand and the high-power path distribution, extract regional nodes with prominent power demand, and calculate the partition adjustment range in combination with the power balance principle, redefine and adjust the partition boundaries, and generate power-optimized partition boundaries; Dynamic partition model establishment submodule: Based on the power optimization partition boundary, the alternating direction multiplier method is used to integrate the power requirements and transmission path characteristics of the partition nodes, extract the power matching relationship between the nodes according to the power requirements and the transmission path characteristics, analyze the power distribution characteristics between the nodes and paths in the dynamic partition, build a dynamic partition matching model of the nodes and paths, and extract the dynamic adjustment rules in the model to generate a dynamic partition structure model; The dynamic partition structure model includes partition boundary positions, partition power flow paths and partition power balance parameter sets.

5. The intelligent power distribution system according to claim 4, characterized in that: The difference between the power value and the partition demand is compared. The difference between the power value and the partition demand is quantitatively analyzed by analyzing the relationship between the total power transmitted by the path and the total demand of the regional nodes one by one. According to the transmission direction of the path power, the partition power demand data covered by the path is gradually counted, and the relationship between the path power and the demand is compared to screen out the path that meets the partition demand or the power exceeds the demand; The alternating direction multiplier method is used to decompose the overall partition optimization problem into multiple independently computable sub-problems, gradually adjust the power distribution and path load characteristics of the partition nodes, and adjust the partition structure at the same time; The dynamic adjustment rules in the extraction model are extracted by continuously analyzing the relationship between the changes in the power transmission path and the node demand distribution. When the power load in the dynamic partition is unbalanced, the adjustment rules automatically trigger real-time adjustment operations of the path and partition range.

6. The intelligent power distribution system according to claim 1, characterized in that: The priority power control module includes a load priority screening submodule, a power distribution adjustment submodule and a partition power flow calculation submodule, wherein: Load priority screening submodule: based on the dynamic partition structure model, extract partition nodes and their corresponding load demands, gradually screen the priority attributes of the nodes and mark high-priority nodes, analyze the power demand distribution of the high-priority node group, and generate high-priority load distribution data; Power allocation adjustment submodule: Based on the high-priority load distribution data, a greedy algorithm is used to extract the load demand value according to the partition priority, and the power allocation ratio is gradually adjusted in combination with the power transmission relationship between partitions. The demand for cross-regional power flow between partitions is counted, and the power allocation value of the high-priority partition is adjusted and matched with the power demand, so as to generate a high-priority power allocation plan; Partition power flow calculation submodule: Based on the high-priority power allocation scheme, extract the power transmission path between nodes across partitions, gradually analyze the power transmission flow and statistically analyze the power distribution, analyze the node coverage and flow direction of cross-partition power flow according to the power distribution, and analyze the power flow relationship in combination with the power allocation scheme to generate a partition priority power allocation scheme; The partition priority power allocation scheme includes a high priority partition power input amount, power flow distribution between partitions and a power allocation parameter set.

7. The intelligent power distribution system according to claim 1, characterized in that: The greedy algorithm, according to the formula: in: is the updated power allocation value of partition i, is the current power allocation value of partition i, λ is the power adjustment coefficient, D i is the high priority load demand value of partition i, T i is the current power transfer amount of partition i, ψ i is the power matching priority coefficient of partition i, f j is the load transmission path weight of partition j, ω j is the power consumption efficiency coefficient of partition j, Normalized value of load deviation of all high priority partitions under multi-weight correction.

8. The intelligent power distribution system according to claim 1, characterized in that: The voltage dynamic adjustment value includes a voltage offset range, a reactive power compensation adjustment parameter and an inter-zone voltage balance value.

9. The intelligent power distribution system according to claim 1, characterized in that: The global load power coordination value includes cross-partition power flow distribution, full network transmission capacity adjustment value and full network power balancing parameter.

10. An intelligent power distribution method, characterized in that: The intelligent power distribution system according to any one of claims 1 to 9 comprises the following steps: Step 1: Obtain the load distribution value of the distribution network node through the data collected by the electric meter, calculate the power flow, line transmission loss and load value density of the load transmission path between the nodes according to the distribution value, and associate the load data of the density area with the node position and coverage range by comparing the density of the load values ​​between the nodes, and calibrate the high-load nodes and coverage range at the same time to generate the load density distribution result; Step 2: Based on the load density distribution result, extract the load time series of the high-load node, perform correlation calculation through the fluctuation amplitude and fluctuation direction of the load time series data, compare the fluctuation range in the historical load data, determine the load peak range of the future load area, analyze the load trend change in the area and generate the high-load area prediction result in combination with the peak range; Step 3: Based on the prediction results of the high-load area, the alternating direction multiplier method is used to extract the high-power path by comparing the transmission power value between the partition nodes and the partition demand, and the partition boundary parameters are dynamically adjusted according to the power demand. The partition boundary position is correlated with the node power flow path, the power balance distribution within the partition and the partition boundary range are optimized, and a dynamic partition structure model is generated; Step 4: Based on the dynamic partition structure model, a greedy algorithm is used to screen the priority value of the load within the partition, and the load demand within the partition is compared and adjusted with the load distribution standard. The partition priority power distribution plan is obtained by calculating the power flow direction and flow across the partitions; Step 5: Based on the partition priority power allocation scheme, compare the voltage offset range between partitions and the operating output value of the reactive power compensation equipment, calculate the impact range of the power balance between partitions on the voltage fluctuation through "power flow calculation", optimize the voltage state by adjusting the reactive power compensation value, combine the distribution parameters of cross-partition power flow and transmission capacity, screen out-of-capacity paths and adjust the power flow direction to generate a global load power coordination value.

Images