Net rack reconstruction method for transforming single radiation line into 2-1 single ring network

Abstract

The invention relates to a net rack reconstruction method for transforming a single-radiation line into a 2-1 single-ring network, and the method comprises the steps: recognizing at least two feeder lines in a single-radiation operation state in a power grid, obtaining the electrical parameters, load rates and wiring modes of the feeder lines, selecting the two single-radiation feeder lines with the load rate difference meeting a preset condition as a target transformation line, and carrying out the reconstruction of the two single-radiation feeder lines. A cable line is newly built from an intermediate node of a first target feeder line to be connected to an intermediate node of a second target feeder line, automatic section switches are additionally arranged on the two target feeder lines respectively to form a main section, and the two original single-radiation feeder lines are connected to form a 2-1 single-ring network structure through the newly built cable line and the additionally arranged automatic section switches. And an optical fiber communication network and a distribution automation terminal are configured to realize fault positioning, isolation and non-fault section power supply recovery functions.

Description

Technical Field

[0001] This application relates to the field of information technology, specifically to a network reconstruction method for transforming a single radial line into a 2-1 single ring network. Background Technology

[0002] With urban development and the continuous growth of electricity load, the reliability and flexibility of power distribution networks are facing higher requirements. In the existing power distribution network structure, the single radial connection mode has been widely used in the past due to its simple structure and low investment. However, this mode has obvious inherent defects:

[0003] The lack of communication between lines prevents them from transferring power to each other. When a fault occurs in any section of the line, the entire line will be shut down, resulting in poor power supply reliability.

[0004] Furthermore, when the line load is unbalanced, load transfer cannot be carried out through network reconfiguration, which limits the operational flexibility of the power grid;

[0005] Currently, although there are renovation schemes that completely change the grid structure by building new lines or substations, such schemes usually involve huge civil engineering, equipment investment and long construction period, which are not economical. Moreover, in the early stage of construction, there may be problems such as extremely low load rate and poor investment benefits. Especially in areas such as Class B power supply areas, where there are dual requirements for investment efficiency and power supply reliability, how to optimize and reconstruct the grid with minimal investment based on existing equipment has become a key technical problem.

[0006] There is a lack of an efficient, economical, and targeted solution in the current technology to quickly transform such single-radial lines into a highly reliable ring network structure. Summary of the Invention

[0007] In order to solve the problems existing in the prior art, the purpose of this application is to provide a network reconstruction method for transforming a single radial line into a 2-1 single ring network.

[0008] The network reconstruction method for converting a single radial line into a 2-1 single ring network as described in this application includes:

[0009] S101. Identify at least two feeders in the power grid that are operating in a single radial mode, and obtain their electrical parameters, load rate and wiring mode;

[0010] S102. Select two single-radial feeders with load rate differences that meet the preset conditions as the target lines for modification.

[0011] S103. A new cable line is built from the intermediate node of the first target feeder to the intermediate node of the second target feeder;

[0012] S104. Automatic segmentation switches are added to the two target feeders to form a main trunk segmentation.

[0013] S105. By connecting the two original single radial feeders to form a 2-1 single ring network structure through the newly built cable line and the added automated sectionalizing switch;

[0014] S106. Configure fiber optic communication network and power distribution automation terminal for fault location, isolation and power restoration of non-faulty sections.

[0015] Preferably, in S101, the electrical parameters and wiring data of the single-radial operating feeder identified by the power grid monitoring system are obtained, and the operating status is determined by the support vector machine algorithm.

[0016] For feeders with excessive load rates, the load distribution characteristics are analyzed by combining electrical parameters and wiring methods. Based on the load distribution and line topology information, the K-means clustering algorithm is used to generate feeder operation optimization suggestions.

[0017] Preferably, in S102, based on the real-time scanning of the power grid structure by the power grid monitoring system, feeders in single-radial operation are identified and a set of identifiers is formed;

[0018] Obtain the load rate data of the corresponding feeders, use the Euclidean distance algorithm to calculate the load rate difference between feeders, and include feeder pairs that exceed the set threshold in the candidate renovation list.

[0019] After obtaining the wiring method through topology analysis, the decision tree algorithm is used to determine the priority of each feeder's transformation, and the target feeder list is determined based on the priority ranking and the principle of maximum load difference.

[0020] Preferably, in S103, the cable line length is calculated using the shortest path algorithm based on the intermediate node positions of the first and second target feeders obtained by scanning the power grid monitoring system.

[0021] If the length is less than the preset threshold, the power grid structure is analyzed through topology analysis, and the connection path is optimized using the simulated annealing algorithm to determine the load-balanced cable connection scheme.

[0022] Voltage stability is verified through current distribution analysis. If the stability requirements are met, the final cable connection scheme and implementation parameters are generated, and the load balancing effect is verified through power grid simulation tools.

[0023] Preferably, in S104, the power grid structure is analyzed using topology analysis based on the target feeder node location obtained from the power grid monitoring system.

[0024] If the segmentation conditions are met, the k-means clustering algorithm is used to determine the location of the segmentation points, and the load at each segmentation point is calculated through current distribution analysis.

[0025] If the load value meets the balancing threshold, the configuration scheme of the automated sectionalizing switch is optimized by the simulated annealing algorithm, and the operating status is verified by power grid simulation.

[0026] If the voltage fluctuation meets the stability requirements, generate a deployment plan and parameters for the automated sectionalizing switch.

[0027] Preferably, in step S105, topology analysis is used to evaluate the node connection status based on the interconnection path data obtained from the power grid monitoring system.

[0028] If the connection reliability meets the standard, the network redundancy is calculated by graph theory algorithm, and combined with the cable load capacity analysis results, the current distribution is calculated and the voltage stability is evaluated under the premise of meeting the capacity threshold.

[0029] Based on the stability verification results, cable line deployment parameters and automated switch layout schemes are generated. After operation simulation and fault recovery capability assessment, if all indicators meet the requirements, a complete deployment scheme for a single-ring network structure containing dynamic response verification results is generated.

[0030] Preferably, in S106, based on the real-time signal strength data of the optical fiber communication path, if the signal quality meets the threshold requirements, a path optimization algorithm is used to calculate the optimal communication path and analyze its traffic distribution balance.

[0031] Under the premise of meeting the load threshold, evaluate the communication response time of the distribution automation terminal, locate abnormal nodes and determine the boundary of the fault area;

[0032] If the isolation command execution time meets the requirements, simulate the power restoration process of the non-faulty section, analyze the power distribution balance, and generate dynamic adjustment parameters for the terminal.

[0033] Verify the dynamic transmission performance of fiber optic communication. If the performance meets the standards, generate optimized configuration parameters and verify the overall operation status of the power distribution network through power grid simulation.

[0034] The method for reconstructing a single-radial line into a 2-1 single-ring network as described in this application has the advantage that by connecting two independent single-radial lines to form a 2-1 single-ring network, mutual power supply and backup between the lines can be achieved. When any line fails, power can be transferred through the other line to quickly restore power to the non-faulty section and minimize the impact of the fault.

[0035] This invention constructs a flexible load transfer channel by selectively connecting lines with large differences in load rates. When the load increases in the future, the load can be flexibly allocated between the two lines, which not only relieves the pressure on heavily loaded lines, but also makes full use of the capacity of lightly loaded lines, revitalizes existing assets, and improves the overall utilization rate of the power grid.

[0036] This invention simultaneously constructs a fiber optic communication network and distribution automation terminals (DTU / FTU), adopting a "master station-tiered collaborative" feeder automation mode. This enables the system to achieve millisecond-level precise fault location, second-level automatic isolation, and minute-level automatic power restoration, greatly shortening power outage time and transforming fault handling from traditional manual troubleshooting to fully automatic intelligent processing, significantly reducing the number of households affected by power outages. Attached Figure Description

[0037] Figure 1 This application describes a method for reconfiguring a single-radial line into a 2-1 single-ring network. Figure 1 ;

[0038] Figure 2 This application describes a method for reconfiguring a single-radial line into a 2-1 single-ring network. Figure 2 . Detailed Implementation

[0039] like Figures 1-2 As shown, the network reconstruction method for converting a single radial line into a 2-1 single ring network according to this application includes the following steps:

[0040] like Figures 1-2 As shown, S101 identifies at least two feeders in the power grid that are in single-radial operation and obtains their electrical parameters, load rate and wiring mode.

[0041] Further, in step S101, the power grid structure is scanned by the power grid monitoring system to identify at least two feeders in single-radial operation state, thereby obtaining a set of feeder identifiers;

[0042] Based on the set of feeder identifiers, the electrical parameters of each feeder are obtained from the power grid database to obtain an electrical parameter dataset;

[0043] The support vector machine algorithm is used to classify the electrical parameter dataset, determine the operating status of each feeder, and determine the list of feeders in single-radial operating status.

[0044] For a list of feeders operating in a single-radial configuration, calculate the load rate for each feeder to obtain a load rate dataset;

[0045] The feeder list for single-radial operation is analyzed using topology analysis tools to obtain the wiring method of each feeder, resulting in a wiring method dataset.

[0046] If the load rate in the load rate dataset exceeds the preset threshold, the load distribution is analyzed by combining the electrical parameter dataset and the wiring method dataset to obtain the load distribution characteristics.

[0047] Based on load distribution characteristics and line topology, the feeders are grouped using the K-means clustering algorithm to obtain feeder operation optimization suggestions.

[0048] Specifically, in step S101, in order to identify at least two feeders in the power grid that are in single-radial operation and to obtain their electrical parameters, load rate and wiring mode, the following information technology means can be used;

[0049] First, real-time operation data is collected using the distribution network SCADA system. Assuming that a certain area power grid contains 10 feeders, the feeders with single-end power supply are selected by querying the topology connection status of the feeders.

[0050] Data returned by the SCADA system indicates that there are no other power connections at the end nodes of feeders L1 and L2, confirming that they are in single-radial operation.

[0051] Next, the electrical parameters of L1 and L2 are extracted, including rated voltage of 10kV, rated current of 200A, line length of 15km and 12km respectively, and impedance of 0.15Ω / km and 0.13Ω / km respectively.

[0052] Real-time load data is obtained by querying the database. The actual current of L1 is 120A and that of L2 is 150A. The calculated load rates are 120 / 200 = 60% and 150 / 200 = 75%, respectively.

[0053] Load analysis shows that L1 is operating relatively stably, while L2 is close to a high load state, and the risk of overload needs to be monitored.

[0054] Subsequently, the wiring pattern was analyzed. Through topology analysis algorithm (depth-first search based on graph theory), it was confirmed that L1 is a radial wiring with single-ended power supply, and L2 is a radial wiring but includes branch nodes, with branch load accounting for 20%.

[0055] To ensure logical rigor, based on historical data analysis, the load rate fluctuation ranges of L1 and L2 in the past 24 hours were 55%-65% and 70%-80%, respectively, indicating that the operating status is stable but L2 needs to optimize scheduling.

[0056] Finally, the electrical parameters, load rate, and wiring mode of L1 and L2 are output and stored in the database for subsequent scheduling.

[0057] like Figures 1-2 As shown in Figure S102, select two single-radial feeders with load rate differences that meet the preset conditions as the target lines for modification.

[0058] Furthermore, in step S102, the power grid structure is scanned in real time by the power grid monitoring system to identify feeders in a single-radial state and obtain a set of feeder identifiers;

[0059] Retrieve load rate data corresponding to the feeder identifier set from the power grid database to generate a load rate dataset;

[0060] The load rate difference between each pair of feeders in the load rate dataset is calculated using the Euclidean distance algorithm, resulting in a load rate difference set.

[0061] If the difference value in the load rate difference set is greater than the preset difference threshold, the corresponding feeder pair is determined and a candidate feeder modification list is generated.

[0062] The feeders in the candidate feeder modification list are analyzed using topology analysis tools to obtain the wiring method of each feeder, resulting in a wiring method dataset.

[0063] Based on the wiring method dataset, the decision tree algorithm is used to classify the candidate feeder modification list, determine the modification priority of each feeder, and obtain a priority ranking list.

[0064] Based on the priority sorting list, select the two feeders with the largest difference in load rate and determine them as the target feeders for modification, thus generating the final list of modified feeders.

[0065] Specifically, in step S102, the real-time data of a power grid in a certain area containing 8 feeders is collected using the distribution network SCADA system. The feeders operating in a single radial manner are screened through a topology analysis algorithm (breadth-first search based on graph theory), and feeders L3 and L4 are confirmed to be single-ended power supply.

[0066] The SCADA system extracts the electrical parameters of L3 and L4: L3 rated voltage 10kV, rated current 250A, line length 10km, impedance 0.12Ω / km; L4 rated voltage 10kV, rated current 300A, line length 8km, impedance 0.14Ω / km.

[0067] Real-time load data query shows that the actual current of L3 is 100A and the actual current of L4 is 210A. The calculated load rates are 100 / 250 = 40% and 210 / 300 = 70% respectively.

[0068] The preset condition is that the load rate difference is greater than 25%, and the difference between the two is 70%-40%=30%, which meets the condition;

[0069] The wiring pattern analysis used a topology tracing algorithm to confirm that L3 is a pure radial wiring with no branches; L4 is a radial wiring with one branch node, and the branch load accounts for 15%.

[0070] To assess the priority of the upgrade, the historical database was accessed to analyze the load rate fluctuations over the past 48 hours: L3 fluctuated between 35% and 45%, and L4 fluctuated between 65% and 75%, indicating that L4 had a higher load and greater fluctuations, and should be upgraded first.

[0071] Based on the difference in load rate, the renovation benefit index is calculated: L3 benefit index is 40×0.5=20, L4 is 70×0.8=56 (weights are based on load rate and fluctuation range);

[0072] Ultimately, L3 and L4 are selected as the target lines for modification, and the data is stored in the scheduling database for use in optimization planning.

[0073] like Figures 1-2 As shown, S103, a new cable line is built from the intermediate node of the first target feeder to the intermediate node of the second target feeder.

[0074] Further, in step S103, the power grid structure is scanned by the power grid monitoring system to obtain the intermediate node positions of the first target feeder and the second target feeder, thereby obtaining a node position dataset;

[0075] Based on the node location dataset, the shortest path algorithm is used to calculate the cable length between the intermediate node of the first target feeder and the intermediate node of the second target feeder, thus obtaining the line length data;

[0076] If the line length data is less than the preset length threshold, the topological relationships in the power grid structure are analyzed using topology analysis tools to obtain a topological relationship dataset.

[0077] Based on the topology dataset, the simulated annealing algorithm is used to optimize the connection path of the cable lines, determine the load-balanced cable connection scheme, and obtain the optimized connection path.

[0078] By using current distribution analysis tools, current distribution data in the optimized connection path is obtained, and it is determined whether the voltage stability meets the preset stability threshold to obtain the stability verification results.

[0079] If the stability verification results meet the preset threshold, the final cable connection scheme is generated based on the optimized connection path, and the implementation parameters of the newly built cable line are determined.

[0080] Based on the final cable connection scheme, the load balancing effect is verified using power grid simulation tools to obtain the verified cable connection scheme.

[0081] Specifically, in step S103, the power distribution network SCADA system acquires the electrical parameters of the target feeders L1 and L2 through real-time data acquisition:

[0082] L1 has a rated voltage of 10kV, a rated current of 200A, a line length of 12km, and an impedance of 0.15Ω / km.

[0083] L2 has a rated voltage of 10kV, a rated current of 280A, a line length of 9km, and an impedance of 0.13Ω / km.

[0084] The system uses a depth-first search algorithm to analyze the power grid topology and determine the midpoint node positions of L1 and L2. The midpoint of L1 is located at 6km, node N1, and the midpoint of L2 is located at 4.5km, node N2.

[0085] The SCADA system queries the geographical distance between nodes and calculates the straight-line distance from N1 to N2 to be 2.5km. Considering the actual laying path, the cable length is corrected to 3km.

[0086] The cable selection algorithm selects a cross-sectional area of ​​240mm² based on load prediction. 2 The copper core cable has an impedance of 0.08Ω / km and a rated capacity of 400A.

[0087] After the system simulates the connection of the newly built cable, it calculates the load distribution of L1 and L2: the original load current of L1 is 80A, the original load current of L2 is 220A. After the connection, according to Kirchhoff's current law, it is estimated that the current shunted from L1 to L2 is 50A, the load of L1 is reduced to 30A, and the load of L2 is increased to 270A.

[0088] The new load factors are 30 / 200 = 15% and 270 / 280 = 96.43%, respectively.

[0089] To assess connection stability, historical databases were accessed to extract voltage data for nodes N1 and N2 over the past 72 hours. The voltage fluctuation range at the midpoint of L1 was 9.8-10.2kV, and the voltage fluctuation range at the midpoint of L2 was 9.6-10.1kV. The voltage difference was less than 0.5kV, which met the parallel connection requirements.

[0090] The system simulated 1000 load scenarios using the Monte Carlo algorithm and confirmed that the probability of line overload after the new cable was installed was less than 2%.

[0091] Finally, the cable connection scheme parameters are stored in the power distribution network planning database for use in dispatch optimization.

[0092] like Figures 1-2 As shown in Figure S104, automatic segmentation switches are added to the two target feeders to form the main trunk segment.

[0093] Further, in step S104, the power grid structure is scanned by the power grid monitoring system to obtain the node positions of the first target feeder and the second target feeder, thus obtaining a node position dataset;

[0094] Based on the node location dataset, topology analysis tools are used to parse the topological relationships in the power grid structure to obtain a topology relationship dataset.

[0095] If the topological relationship dataset meets the preset segmentation conditions, the k-means clustering algorithm is used to classify the node position dataset, determine the segmentation point positions of the first target feeder and the second target feeder, and obtain the segmentation point dataset.

[0096] Based on the segment point dataset, the current load data of each segment point is calculated using a current distribution analysis tool to obtain the current load dataset.

[0097] If the load values ​​in the current load dataset meet the preset load balancing threshold, then based on the segmentation point dataset and the current load dataset, the configuration scheme of the automated segmented switch is optimized using the simulated annealing algorithm to obtain the switch configuration dataset.

[0098] Based on the switch configuration dataset, the operating status of the segmented switches is simulated using a power grid simulation tool to obtain the operating status dataset;

[0099] If the voltage fluctuation value in the operating status dataset meets the preset stability threshold, the final automated segmented switch deployment scheme is generated based on the switch configuration dataset, and the deployment parameter dataset is determined.

[0100] Specifically, in step S104, the power distribution network SCADA system acquires the electrical parameters of the target feeders L3 and L4 through real-time data acquisition:

[0101] L3 has a rated voltage of 10kV, a rated current of 250A, a line length of 15km, and an impedance of 0.14Ω / km.

[0102] L4 has a rated voltage of 10kV, a rated current of 300A, a line length of 10km, and an impedance of 0.12Ω / km.

[0103] The system uses a breadth-first search algorithm to analyze the power grid topology and determine suitable segmentation points for L3 and L4. The segmentation point for L3 is located at 7.5km, node M1, and the segmentation point for L4 is located at 5km, node M2.

[0104] The SCADA system queries historical load data. The current load current of L3 is 100A and the load current of L4 is 200A. Based on the geographical information, the straight-line distance between M1 and M2 is calculated to be 2km. The actual laying path is corrected to 2.5km.

[0105] Based on load prediction, the system calls the automated switch selection algorithm to select a vacuum circuit breaker with a rated current of 400A, a response time of less than 50ms, and a switching impedance of 0.01Ω.

[0106] After simulating the addition of a switch, the system estimates the load current of the first segment after L3 segmentation to be 60A and the last segment to be 40A through node voltage analysis.

[0107] After L4 segmentation, the first segment is 120A and the last segment is 80A, with load rates of 60 / 250 = 24% and 120 / 300 = 40% respectively;

[0108] To verify the segment stability, the system extracted the voltage data of M1 and M2 over the past 48 hours. The voltage fluctuation range of L3 was 9.7-10.3kV, and the fluctuation range of L4 was 9.5-10.2kV. The voltage difference was less than 0.6kV, which met the requirements for segment operation.

[0109] The system uses a particle swarm optimization algorithm to simulate 500 fault scenarios, confirming that the fault isolation success rate after segmentation is 98% and the overload probability is less than 1.5%.

[0110] Finally, the configuration parameters of the sectionalizing switches are stored in the distribution network automation database for use in fault recovery and dispatch optimization.

[0111] like Figures 1-2 As shown in S105, the two original single radial feeders are connected to form a "2-1" single ring network structure through the newly built cable line and the added automatic sectionalizing switch.

[0112] Further, in step S105, the interconnection path dataset is obtained through the power grid monitoring system, and the node connection status of the interconnection path is analyzed using a topology analysis tool to obtain the node connection dataset;

[0113] If the connection status in the node connection dataset meets the preset reliability threshold, then the network redundancy of the connection path is calculated using graph theory algorithm based on the node connection dataset to obtain the redundancy dataset.

[0114] Based on the redundancy dataset, the cable load capacity of the connection path is analyzed using cable selection tools to obtain the load capacity dataset;

[0115] If the load value in the load capacity dataset meets the preset capacity threshold, then the current distribution of the connection path is calculated using simulation analysis tools based on the load capacity dataset and the connection path dataset to obtain the current distribution dataset.

[0116] Based on the current distribution dataset, the voltage stability of the single-ring network structure is evaluated using voltage analysis tools to obtain a voltage stability dataset.

[0117] If the fluctuation values ​​in the voltage stability dataset meet the preset stability threshold, then the deployment parameters of the cable line are generated based on the voltage stability dataset and the connection path dataset, and the line deployment dataset is obtained.

[0118] Based on the line deployment dataset, the installation locations of automated sectionalizing switches are determined using an automated switch configuration tool, resulting in a switch location dataset.

[0119] If the position distribution in the switch position dataset meets the preset equilibrium threshold, then the operating state of the single-ring network structure is simulated using simulation tools based on the switch position dataset to obtain the operating state dataset.

[0120] Based on the operational status dataset, the fault recovery capability of the single-ring network structure is evaluated using fault analysis tools to obtain a fault recovery dataset.

[0121] If the recovery time in the fault recovery dataset meets the preset recovery threshold, then the configuration parameters of the automated sectionalizing switch are generated based on the fault recovery dataset and the switch position dataset, and the switch configuration dataset is obtained.

[0122] Based on the switch configuration dataset, the dynamic response characteristics of the single-ring network structure are verified using power grid simulation tools to obtain the dynamic response dataset.

[0123] If the response metrics in the dynamic response dataset meet the preset performance threshold, then the final deployment scheme of the single-ring network structure is generated based on the dynamic response dataset and the switch configuration dataset, and the final deployment dataset is obtained.

[0124] Specifically, in step S105, the power distribution network SCADA system acquires the electrical parameters of the newly built cable lines L5 and L6 in real time:

[0125] L5 has a rated voltage of 10kV, a rated current of 200A, a line length of 12km, and an impedance of 0.15Ω / km.

[0126] L6 has a rated voltage of 10kV, a rated current of 280A, a line length of 8km, and an impedance of 0.13Ω / km.

[0127] The system uses a depth-first search algorithm to analyze the power grid topology and determines that the L5 segment point is located at 6km, node N1, and the L6 segment point is located at 4km, node N2.

[0128] Using a geographic information system, the straight-line distance between N1 and N2 is calculated to be 1.8 km, which is corrected to 2.2 km considering terrain and laying path.

[0129] The SCADA system queries the load data of the past 72 hours. The load current of L5 is 80A and the load current of L6 is 160A. Based on time series analysis, the peak loads for the next 24 hours are predicted to be 90A and 180A, respectively.

[0130] The system invokes the switch selection algorithm to select an SF6 ROAD circuit breaker with a rated current of 350A, a response time of less than 40ms, and a switch impedance of 0.008Ω.

[0131] After the simulated interconnection forms a "2-1" single-ring network structure, the system estimates, through power flow calculation and analysis, that the load current of the first segment after L5 segmentation is 50A and the last segment is 30A; after L6 segmentation, the first segment is 100A and the last segment is 60A, with load rates of 50 / 200 = 25% and 100 / 280 = 35.7%, respectively.

[0132] To ensure the stability of the ring network, the system extracted the voltage data of N1 and N2 over the past 72 hours. The voltage fluctuation range of L5 was 9.8-10.4kV, and the fluctuation range of L6 was 9.6-10.3kV. The voltage difference was less than 0.5kV, which met the requirements for ring network operation.

[0133] The system uses a genetic algorithm to simulate 1000 fault scenarios, confirming that the fault isolation success rate under the ring network structure reaches 97.5%, and the short-circuit current over-limit probability is less than 1.2%.

[0134] Finally, the ring network configuration parameters are stored in the distribution network automation database for real-time scheduling and fault diagnosis.

[0135] like Figures 1-2 As shown, S106 configures an optical fiber communication network and a power distribution automation terminal for fault location, isolation, and power restoration of non-faulty sections.

[0136] Further, in step S106, the real-time signal strength dataset of the optical fiber communication path is obtained through the power grid monitoring system, and the signal strength is evaluated using signal analysis tools to obtain a signal quality dataset;

[0137] If the quality values ​​in the signal quality dataset meet the preset signal threshold, then the optimal communication path is calculated using a path optimization algorithm based on the signal quality dataset and the optical fiber communication path, and the optimal path dataset is obtained.

[0138] Based on the optimal path dataset, the traffic allocation of the optical fiber communication path is analyzed using communication scheduling tools to obtain the traffic allocation dataset;

[0139] If the distribution balance in the traffic distribution dataset meets the preset load threshold, then the communication response time of the distribution automation terminal is evaluated using simulation analysis tools based on the traffic distribution dataset and the optimal path dataset, and the response time dataset is obtained.

[0140] Based on the response time dataset, abnormal nodes in the power distribution network are detected using fault location tools to obtain an abnormal node dataset.

[0141] If the number of nodes in the abnormal node dataset is lower than the preset fault threshold, then a clustering algorithm is used to determine the boundary of the fault area based on the abnormal node dataset and the terminal location, and the fault area dataset is obtained.

[0142] Based on the fault area dataset, fault isolation instructions are generated using isolation control tools to obtain an isolation instruction dataset;

[0143] If the execution time of the instructions in the isolation instruction dataset meets the preset isolation threshold, then the power supply restoration process of the non-faulty section is simulated using simulation tools based on the isolation instruction dataset and terminal configuration parameters to obtain the power supply restoration dataset.

[0144] Based on the power restoration dataset, the power distribution in non-faulty sections is analyzed using load balancing tools to obtain the power distribution dataset;

[0145] If the distribution balance in the power distribution dataset meets the preset balance threshold, then the dynamic adjustment parameters of the distribution automation terminal are generated based on the power distribution dataset and the terminal configuration parameters, and the dynamic adjustment dataset is obtained.

[0146] Based on the dynamically adjusted dataset, the dynamic transmission performance of the optical fiber communication network is evaluated using communication verification tools to obtain the transmission performance dataset;

[0147] If the performance indicators in the transmission performance dataset meet the preset performance threshold, then the optimized configuration parameters of the optical fiber communication network are generated based on the transmission performance dataset and the dynamic adjustment dataset, and the optimized configuration dataset is obtained.

[0148] Based on the optimized configuration dataset, the overall operating status of the distribution network is verified using power grid simulation tools, resulting in an operating status dataset.

[0149] Specifically, in step S106, the distribution network automation system achieves fault location, isolation, and power restoration of non-faulty sections by deploying fiber optic communication networks and distribution automation terminals;

[0150] First, the system is configured with an optical fiber communication network, using single-mode optical fiber with a transmission rate of 10Gbps, covering a power distribution area of ​​10km×10km, deploying 48 optical fiber nodes, each node supporting 8 terminal accesses, and communication latency of less than 5ms.

[0151] The system collects power distribution terminal data in real time through a fiber optic network, including voltage, current and switch status, with a data sampling frequency of 10 times per second;

[0152] The distribution automation terminal is installed at key nodes of the feeder. The terminal model is DTU-200, which can process 1,000 data frames per second and has a built-in fault detection algorithm.

[0153] The system uses wavelet transform algorithm to analyze the collected current waveform data and sets a fault threshold: when the sudden change in current exceeds 150% of the rated current (the threshold is 300A for a line with a rated current of 200A), it is judged as a fault.

[0154] Fault location uses the time domain reflectometry (TDR) method, which sends pulse signals through optical fiber, measures the reflection time, and calculates the distance to the fault point with an accuracy of ±50m.

[0155] For example, a line is 8km long, the pulse propagation speed is 2×10^8m / s, the reflection time is 20μs, and the fault point is calculated to be located at 4km.

[0156] Fault isolation is achieved by controlling a vacuum circuit breaker via an automated terminal. The rated current is 400A, and the disconnection time is less than 30ms. The system determines the fault section based on a breadth-first search algorithm and automatically generates an isolation command with an isolation time of less than 100ms.

[0157] Restoring power to non-faulty sections utilizes the topology analysis function of the distribution automation system, calls the A* algorithm to calculate the optimal power supply path, and combines the geographic information system to confirm the length of the feasible path. For example, the distance between two nodes is calculated to be 2.5km, which is corrected to 3km considering the terrain. Power flow calculation verifies that the load rate of the restored path is less than 70%, ensuring stable power supply.

[0158] The system stores fault location, isolation, and recovery data in the database. The record format includes fault time, location, isolation time, and recovery path for subsequent analysis and optimization.

[0159] For those skilled in the art, various other corresponding changes and modifications can be made based on the technical solutions and concepts described above, and all such changes and modifications should fall within the protection scope of the claims of this application.

Claims

1. A network architecture reconstruction method for transforming a single-radiation line into a 2-1 single-ring network, characterized in that, The method comprises the following steps: S101, identifying at least two feeders in a single radiation operation state in the power grid, obtaining their electrical parameters, load rates and connection modes; S102, selecting two single radiation feeders with a load rate difference meeting a preset condition as target transformation lines; S103, newly building a cable line from a middle node of a first target feeder to a middle node of a second target feeder; S104, adding an automatic sectionalizing switch on each of the two target feeders to form a main section; S105, connecting the two original single radiation feeders to form a 2-1 single ring network structure through the newly built cable line and the added automatic sectionalizing switch; S106, configuring a fiber-optic communication network and a distribution automation terminal for fault positioning, isolation and non-fault section power restoration functions.

2. The network architecture reconstruction method of claim 1, wherein the single-radiation line is reconstructed into a 2-1 single-ring network. In step S101, a support vector machine algorithm is used to judge the operation state of the feeder, and a K-means clustering algorithm is used to analyze the load distribution characteristics to obtain optimization suggestions.

3. The network architecture reconstruction method of claim 1, wherein, Step S102 comprises: Real-time scanning of the power grid structure is performed through a power grid monitoring system to identify feeders in a single radiation operation state to obtain a feeder identification set; Load rate data corresponding to the feeder identification set is obtained from a power grid database to obtain a load rate data set; The load rate difference between each pair of feeders in the load rate data set is calculated using the Euclidean distance algorithm to obtain a load rate difference set; If the difference value in the load rate difference set is greater than a preset difference threshold, a corresponding feeder pair is determined to obtain a candidate transformation feeder list; The connection mode of each feeder in the candidate transformation feeder list is obtained by analyzing the feeder through a topology analysis tool to obtain a connection mode data set; According to the connection mode data set, a decision tree algorithm is used to classify the candidate transformation feeder list to determine the transformation priority of each feeder to obtain a priority ranking list; According to the priority ranking list, the two feeders with the largest load rate difference are selected as the target transformation feeders to obtain a final transformation feeder list.

4. The network reconstruction method for converting a single radial line into a 2-1 single ring network according to claim 1, characterized in that, In step S103, the shortest path algorithm is used to calculate the cable line length, and the simulated annealing algorithm is used to optimize the connection path.

5. The network reconstruction method for converting a single radial line into a 2-1 single ring network according to claim 1, characterized in that, In step S104, the k-means clustering algorithm is used to determine the sectionalizing point position, and the simulated annealing algorithm is used to optimize the sectionalizing switch configuration.

6. The network architecture reconstruction method of claim 1, wherein the single-radiation line is reconstructed into a 2-1 single-ring network. In step S105, the graph theory algorithm is used to calculate the network redundancy, and the dynamic response characteristics of the single ring network structure are verified through simulation.

7. The network architecture reconstruction method of claim 1, wherein the single- radial line is reconstructed into a 2-1 single-ring network. Step S106 comprises: The real-time signal strength data set of the fiber-optic communication path is obtained through the power grid monitoring system, and the signal strength is evaluated using a signal analysis tool to obtain a signal quality data set; If the quality value in the signal quality data set meets a preset signal threshold, the optimal communication path is calculated using a path optimization algorithm based on the signal quality data set and the fiber-optic communication path to obtain an optimal path data set; According to the optimal path data set, the traffic distribution of the fiber-optic communication path is analyzed through a communication scheduling tool to obtain a traffic distribution data set; If the allocation balance in the traffic allocation dataset meets a preset load threshold, a simulation analysis tool is used to evaluate a communication response time of the power distribution automation terminal according to the traffic allocation dataset and an optimal path dataset, and a response time dataset is obtained.

Images