Substation self-healing optimization control method and system based on flexible interconnection

Abstract

The present invention relates to a self-healing optimization control method and system for a substation based on flexible interconnection, comprising the following steps: S1: constructing an operation monitoring network based on the equipment ledger in the substation; S2: deploying flexible interconnection equipment at key nodes in the substation to achieve fault isolation and power outage transfer, dynamic load transfer and voltage regulation, distributed power supply, and flexible coordinated control of energy storage equipment; S3: introducing an equipment status evaluation model based on the substation operation data obtained by the operation monitoring network to obtain the status evaluation and risk warning of each device in the substation; S4: constructing a dynamic topology management mechanism of flexible interconnection between substations, and dynamically adjusting the interconnection topology between substations according to the load transmission demand and power supply status; S5: based on the status evaluation and risk warning of each device, according to the dynamic topology management mechanism of flexible interconnection, load regulation and fault recovery are performed. The present invention realizes efficient and intelligent substation operation and improves power supply reliability.

Description

Technical Field

[0001] The present invention relates to the field of intelligent power grid, and in particular to a self-healing optimization control method and system for a substation based on flexible interconnection. Background Art

[0002] With the rapid development of distribution networks towards intelligence and distribution, traditional distribution networks have gradually exposed their limitations such as low operating efficiency, long fault recovery time, and insufficient coordination capabilities when facing problems such as load fluctuations, new energy access, and equipment complexity. The large-scale access of distributed power sources (such as photovoltaics, wind power, etc.) and energy storage equipment has further exacerbated the complexity of distribution network operation, especially in terms of uneven load, voltage fluctuations, and rapid fault isolation and power supply restoration. The traditional rigid grid structure is difficult to meet the needs of new energy development. In addition, problems such as sudden failures or aging of equipment in the substation area may cause large-scale power outages or reduced power quality, affecting power supply reliability.

[0003] The existing substation control solutions have gradually exposed their limitations such as low operating efficiency, long fault recovery time, and insufficient coordination capabilities when facing problems such as load fluctuations, new energy access, and equipment complexity. In particular, the rigid grid structure is difficult to meet the needs of modern grids in terms of uneven load, voltage fluctuations, and rapid fault isolation and power supply restoration. Summary of the invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide a self-healing optimization control method and system for a substation based on flexible interconnection, so as to realize efficient and intelligent substation operation and improve power supply reliability.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A self-healing optimization control method for a substation based on flexible interconnection includes the following steps:

[0007] S1: Build an operation monitoring network based on the equipment records in the substation area;

[0008] S2: Deploy flexible interconnection equipment at key nodes in the substation area to achieve fault isolation and power outage transfer, dynamic load transfer and voltage regulation, and flexible coordinated control of distributed power sources and energy storage equipment;

[0009] S3: Based on the substation operation data obtained from the operation monitoring network, the equipment status evaluation model is introduced to obtain the status evaluation and risk warning of each equipment in the substation;

[0010] S4: Build a dynamic topology management mechanism for flexible interconnection between substations, and dynamically adjust the interconnection topology between substations according to load transmission requirements and power supply conditions;

[0011] S5: Based on the status evaluation and risk warning of each device, load regulation and fault recovery are carried out according to the dynamic topology management mechanism of flexible interconnection.

[0012] Furthermore, S1 is specifically:

[0013] When building an operation monitoring network, first digitally enter the equipment ledger content as an initial reference for the network architecture;

[0014] Smart meters, smart switches, and sensor equipment are deployed at key nodes in the substation area to collect equipment operation data and fault data in real time;

[0015] Within the substation area, NB-IoT is used to collect real-time equipment operation data and fault data and aggregate them to the edge of the substation area.

[0016] Furthermore, the flexible interconnection device includes a flexible intelligent switch, a power electronic interface and an energy storage system, as follows:

[0017] The flexible intelligent switch can realize fault isolation, power outage transfer and load transfer, and is used to quickly cut off the fault line or reconstruct the power supply path. When a line fault is detected, the fault load is isolated, and at the same time, combined with the power electronic interface, the load is transferred to the adjacent feeder or substation to realize power outage transfer.

[0018] The power electronic interface realizes voltage regulation, harmonic control, power quality optimization and dynamic reactive power regulation, and supports the grid-connected coordinated control of distributed power sources; when a local load increases sharply and causes a voltage drop, the power electronic interface adjusts the voltage by dynamically injecting reactive power, and at the same time realizes filtering compensation for harmonics caused by nonlinear loads, thereby improving power quality;

[0019] The energy storage system provides emergency power for the substation after a power outage. When the power supply load changes beyond a threshold, the energy storage system serves as a power supplement to improve the substation's ability to adapt to sudden load changes.

[0020] Furthermore, flexible interconnection equipment is deployed at key nodes in the substation area, as follows:

[0021] Install flexible intelligent switches in the feeder sections of the substation area to distinguish different load sections and reduce the impact of power outages and faults; install flexible intelligent switches at the contact points of adjacent substations to quickly switch the power supply path when a fault occurs and realize inter-regional diversion and transfer; and deploy flexible intelligent switches in different branches of low-voltage lines to only cut off the fault area when a small-scale fault occurs on the user side to ensure the continuity of power supply;

[0022] Power electronic interfaces are arranged at the grid connection points of distributed photovoltaic or wind power to smooth the fluctuating power generation and dynamically adjust reactive power; power electronic interfaces are arranged at the low-voltage busbar of the distribution transformer to achieve overall voltage regulation in the substation, compensate for reactive power and filter harmonics; power electronic interfaces are arranged at the load end position in the heavy load area to suppress the voltage drop and harmonic excess problems at the end during grid operation, and enhance the voltage stability of the substation;

[0023] Deploy energy storage systems in areas where load concentration is greater than the preset value to reduce the pressure on the main grid during peak hours and optimize local power supply loads. Install energy storage systems at points where distributed photovoltaic and wind power are connected to the grid to store excess photovoltaic power generation during the day and release it during peak load periods to maximize the absorption of new energy. Install energy storage systems at the junctions between substations to provide short-term power supply to critical loads in the event of island operation or emergency power supply restoration in substations, and to support dynamic load regulation in adjacent substations.

[0024] Furthermore, the equipment status evaluation model includes a health scoring model and a fault prediction model; the health scoring model calculates the equipment health score through multiple factors; the fault prediction model predicts the potential failure probability based on historical electrical fault data combined with the LSTM model.

[0025] Furthermore, the health score model calculates the device health score through multiple factors, as follows:

[0026] The equipment health score is calculated through multi-factor fusion, integrating operation data, environmental data and historical data to quantify the health status of the equipment.

[0027] The multiple factors include dynamic indicators, historical indicators and environmental indicators; the dynamic indicators include load rate L f 、Temperature rise T r and runtime T u ; Historical indicators include the number of failures N f and the number of repairs N m ; Environmental indicators include ambient temperature T e and humidity H;

[0028] Use weighted linearity to combine the indicators into a health score:

[0029] ;

[0030] Among them, HI is the health score; w1, w2, ..., w7 are weight coefficients;

[0031] The health score is modified according to the set health level classification rules through fuzzy logic method;

[0032] And introduce the degradation model to simulate the decline of health score:

[0033]

[0034] Where HI(t) is the health score of the equipment at time t; HI0 is the initial health score; λ is the degradation coefficient; and t is the operating time.

[0035] Furthermore, the fault prediction model predicts the probability of potential faults based on historical electrical fault data combined with the LSTM model, as follows:

[0036] Input feature vector X t =[L f ,T r ,I,V,T e ,H, fault label], where I and V are current and voltage respectively;

[0037] The state of the LSTM model unit is updated:

[0038] Forget Gate: ;

[0039] Input Gate: ;

[0040] Candidate values: ;

[0041] Cell status update: ;

[0042] Output Gate: ;

[0043] Hide status updates: ;

[0044] in, is the input feature vector at time step t; is the hidden state at time step t; For the Gate of Oblivion; is the input gate; is the candidate value, i.e., the candidate memory cell at the current moment; is the cell state, indicating the memory cell state at the current moment; is the output gate; is the hidden state at the current moment; , , , They are the weight matrices of the forget gate, input gate, candidate value, and output gate respectively; , , , are the bias vectors of the forget gate, input gate, candidate value, and output gate, respectively. is the activation function;

[0045] Train the LSTM model through the training data set and optimize the model parameters so that the model can learn the conditional probability relationship of the fault occurrence at the current time point. Use the cross entropy loss function to measure the error between the actual fault label and the predicted result.

[0046] After the LSTM model training is completed, the failure probability P of the device at the future moment is predicted. fault :

[0047] ;

[0048] Among them, ht is the hidden state of the last layer of the LSTM model; W out , b are the weight and bias of the output layer respectively; sigmoid is the activation function, which maps the output to the range of (0, 1) as the failure probability;

[0049] The potential failure probability P of the output device fault After that, the risk levels can be divided according to the following rules:

[0050] Low risk: P fault <0.2, medium risk: 0.2≤P fault <0.5, high risk: P fault ≥0.5.

[0051] Furthermore, S4 is as follows:

[0052] A multi-agent structure is built between substations, including local agents and global coordination agents. Local agents refer to the allocation of local agents to each substation area, which are responsible for monitoring the operation status of the substation area and preliminarily judging the load pressure; global coordination agents refer to the construction of agents at the station control center level, which summarize all substation information and perform global optimization;

[0053] The local agent aggregates the status data at fixed time intervals; the local agents in the substation area propagate the parameters of adjacent areas to form regional status visualization;

[0054] Design MAS-based topology optimization using an iterative coordination algorithm:

[0055] State exchange: Each Agent exchanges the operating data of its own area; the adjacent areas form a short-term prediction matrix Dij, which is used to calculate the power that needs to be exchanged;

[0056] Each local agent uses the data of the adjacent substation as a constraint to optimize the adjustment amount of the local flexible intelligent switch:

[0057] ;

[0058] in, is the load deviation of the local area; is the voltage deviation of the local area; α1 is the load deviation weight; α2 is the voltage deviation weight;

[0059] The global agent summarizes all local optimization results f local , perform global solution to obtain the topological open-close combination , output switching instructions for flexible interconnected devices, dynamic load transfer plans, and charging and discharging plans for distributed power sources and energy storage.

[0060] Further, S5 is as follows:

[0061] Based on the operation monitoring network and combined with the equipment status evaluation model, the equipment health score and failure probability are obtained;

[0062] Based on flexible interconnected devices, dynamically adjust the load flow or grid topology between stations to meet the needs of different operation scenarios;

[0063] Optimize load transfer paths and distribution based on equipment status, load distribution, and grid limitations, quickly identify and isolate faults through equipment status and real-time topology status, and reconstruct power supply paths using a flexible interconnection mechanism.

[0064] A substation self-healing optimization control system based on flexible interconnection includes a processor, a memory, and a computer program stored in the memory. When the processor executes the computer program, it specifically executes the steps in the substation self-healing optimization control method based on flexible interconnection as described above.

[0065] The present invention has the following beneficial effects:

[0066] 1. The present invention not only solves the problems that traditional power grids are difficult to cope with load fluctuations, new energy consumption and slow fault recovery, but also significantly improves the operating efficiency and intelligence level of the power system;

[0067] 2. The present invention uses flexible interconnection equipment and dynamic topology management mechanism to achieve dynamic transfer and precise distribution of loads in the case of load fluctuation or overload in the substation, avoid overload and power supply interruption, optimize load balance, reduce the operating pressure of overloaded equipment, and extend the service life of equipment. It uses flexible intelligent switches to quickly isolate the fault area and reduce the scope of power outage caused by the spread of faults; through multi-path dynamic power supply topology, it can achieve rapid power supply recovery in healthy substations and shorten fault recovery time;

[0068] 3. The present invention uses energy storage devices and power electronic interfaces to achieve dynamic reactive power compensation and power adjustment, significantly improve voltage stability, and avoid node overvoltage / undervoltage problems. For scenarios where distributed power sources are connected, it can reasonably adjust voltage fluctuations and improve the stability of new energy grid connection. Dynamic topology optimization can reduce long-distance power transmission during peak loads, optimize current paths and line impedances, and reduce power transmission losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0070] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0071] refer to Figure 1 In this embodiment, a self-healing optimization control method based on flexible interconnection is provided, comprising the following steps:

[0072] S1: Build an operation monitoring network based on the equipment records in the substation area;

[0073] S2: Deploy flexible interconnection equipment (such as flexible intelligent switches, power electronic interfaces, energy storage systems, etc.) at key nodes in the substation area to achieve fault isolation and power outage transfer, dynamic load transfer and voltage regulation, and flexible coordinated control of distributed power sources and energy storage equipment;

[0074] S3: Based on the substation operation data obtained from the operation monitoring network, the equipment status evaluation model is introduced to obtain the status evaluation and risk warning of each equipment in the substation;

[0075] S4: Build a dynamic topology management mechanism for flexible interconnection between substations, and dynamically adjust the interconnection topology between substations according to load transmission requirements and power supply conditions;

[0076] S5: Based on the status evaluation and risk warning of each device, load regulation and fault recovery are carried out according to the dynamic topology management mechanism of flexible interconnection.

[0077] In this embodiment, S1 is specifically:

[0078] When building an operation monitoring network, first digitally enter the equipment ledger content as an initial reference for the network architecture; the equipment ledger content includes:

[0079] Basic device description information: device ID, device type (such as transformer, switch, smart meter, sensor, etc.), installation location and manufacturer;

[0080] Operating parameter design range: rated capacity, voltage range and operating temperature threshold;

[0081] Connection relationship and topology structure: the physical connection relationship and logical data association relationship of the equipment in the substation power grid;

[0082] Maintenance record history information: including overload time, cumulative operating time and historical fault records;

[0083] Smart meters, smart switches, and sensor equipment are deployed at key nodes in the substation to collect equipment operation data (voltage, current, load, temperature, etc.) and fault data (overload, open circuit, short circuit, low voltage, etc.) in real time;

[0084] Within the substation area, NB-IoT is used to collect real-time equipment operation data and fault data and aggregate them to the edge of the substation area.

[0085] In this embodiment, the flexible interconnection device includes a flexible intelligent switch, a power electronic interface and an energy storage system, as follows:

[0086] The flexible intelligent switch can realize fault isolation, power outage transfer and load transfer, and is used to quickly cut off the fault line or reconstruct the power supply path. When a line fault is detected, the fault load is isolated, and at the same time, combined with the power electronic interface, the load is transferred to the adjacent feeder or substation to realize power outage transfer.

[0087] The power electronic interface realizes voltage regulation, harmonic control, power quality optimization and dynamic reactive power regulation, and supports the grid-connected coordinated control of distributed power sources; when a local load increases sharply and causes a voltage drop, the power electronic interface adjusts the voltage by dynamically injecting reactive power, and at the same time realizes filtering compensation for harmonics caused by nonlinear loads, thereby improving power quality;

[0088] The energy storage system provides emergency power for the substation after a power outage. When the power supply load changes beyond a threshold, the energy storage system serves as a power supplement to improve the substation's ability to adapt to sudden load changes.

[0089] In this embodiment, flexible interconnection equipment is deployed at key nodes in the substation area, as follows:

[0090] Install flexible intelligent switches in the feeder sections of the substation area to distinguish different load sections and reduce the impact of power outages and faults; install flexible intelligent switches at the contact points of adjacent substations to quickly switch the power supply path when a fault occurs and realize inter-regional diversion and transfer; and deploy flexible intelligent switches in different branches of low-voltage lines to only cut off the fault area when a small-scale fault occurs on the user side to ensure the continuity of power supply;

[0091] Power electronic interfaces are arranged at the grid connection points of distributed photovoltaic or wind power to smooth the fluctuating power generation and dynamically adjust reactive power; power electronic interfaces are arranged at the low-voltage busbar of the distribution transformer to achieve overall voltage regulation in the substation, compensate for reactive power and filter harmonics; power electronic interfaces are arranged at the load end position in the heavy load area to suppress the voltage drop and harmonic excess problems at the end during grid operation, and enhance the voltage stability of the substation;

[0092] Deploy energy storage systems in areas where load concentration is greater than the preset value to reduce the pressure on the main grid during peak hours and optimize local power supply loads. Install energy storage systems at points where distributed photovoltaic and wind power are connected to the grid to store excess photovoltaic power generation during the day and release it during peak load periods to maximize the absorption of new energy. Install energy storage systems at the junctions between substations to provide short-term power supply to critical loads in the event of island operation or emergency power supply restoration in substations, and to support dynamic load regulation in adjacent substations.

[0093] In this embodiment, the equipment status evaluation model includes a health scoring model and a fault prediction model; the health scoring model calculates the equipment health score through multiple factors; the fault prediction model predicts the potential fault probability based on historical electrical fault data combined with the LSTM model.

[0094] In this embodiment, the health score model calculates the device health score through multiple factors, as follows:

[0095] The equipment health score is calculated through multi-factor fusion, integrating operation data, environmental data and historical data to quantify the health status of the equipment.

[0096] The multiple factors include dynamic indicators, historical indicators and environmental indicators; the dynamic indicators include load rate L f 、Temperature rise T r and runtime T u ; Historical indicators include the number of failures N f and the number of repairs N m ; Environmental indicators include ambient temperature T e and humidity H;

[0097] Use weighted linearity to combine the indicators into a health score:

[0098] ;

[0099] Among them, HI is the health score; w1, w2, ..., w7 are weight coefficients;

[0100] The health score is modified according to the set health level classification rules through fuzzy logic method:

[0101] In this embodiment, according to HI, the health status can be divided into 5 levels: excellent (90-100), good (70-89), normal (50-69), poor (30-49), and very poor (<30).

[0102] And introduce the degradation model to simulate the decline of health score:

[0103]

[0104] Where HI(t) is the health score of the equipment at time t; HI0 is the initial health score; λ is the degradation coefficient; and t is the operating time.

[0105] In this embodiment, the fault prediction model predicts the potential fault probability based on historical electrical fault data combined with the LSTM model, as follows:

[0106] Input feature vector X t =[L f ,T r ,I,V,T e ,H, fault label], where I and V are current and voltage respectively;

[0107] The cell state of the LSTM model is updated:

[0108] Forget Gate: ;

[0109] Input Gate: ;

[0110] Candidate values: ;

[0111] Cell status update: ;

[0112] Output Gate: ;

[0113] Hide status updates: ;

[0114] in, is the input feature vector at time step t; is the hidden state at time step t; For the Gate of Oblivion; is the input gate; is the candidate value, i.e., the candidate memory cell at the current moment; is the cell state, indicating the memory cell state at the current moment; is the output gate; is the hidden state at the current moment; , , , They are the weight matrices of the forget gate, input gate, candidate value, and output gate respectively; , , , are the bias vectors of the forget gate, input gate, candidate value, and output gate, respectively. is the activation function;

[0115] Train the LSTM model through the training data set and optimize the model parameters so that the model can learn the conditional probability relationship of the fault occurrence at the current time point. Use the cross entropy loss function to measure the error between the actual fault label and the predicted result.

[0116] After the LSTM model training is completed, the failure probability P of the device at the future moment is predicted. fault :

[0117] ;

[0118] Among them, ht is the hidden state of the last layer of LSTM; W out , b are the weight and bias of the output layer respectively; sigmoid is the activation function, which maps the output to the range of (0, 1) as the failure probability;

[0119] The potential failure probability P of the output device fault After that, the risk levels can be divided according to the following rules:

[0120] Low risk: P fault <0.2, medium risk: 0.2≤P fault <0.5, high risk: P fault ≥0.5.

[0121] In this embodiment, S4 is specifically as follows:

[0122] A multi-agent structure is built between substations, including local agents and global coordination agents. Local agents refer to the allocation of local agents to each substation area, which are responsible for monitoring the operation status of the substation area and preliminarily judging the load pressure; global coordination agents refer to the construction of agents at the station control center level, which summarize all substation information and perform global optimization;

[0123] The local agent aggregates the status data at fixed time intervals; the local agents in the substation area propagate the parameters of adjacent areas to form regional status visualization;

[0124] Design MAS-based topology optimization using an iterative coordination algorithm:

[0125] State exchange: Each Agent exchanges the operating data of its own area; the adjacent areas form a short-term prediction matrix Dij, which is used to calculate the power that needs to be exchanged;

[0126] Each local agent uses the data of the adjacent substation as a constraint to optimize the adjustment amount of the local flexible intelligent switch:

[0127] ;

[0128] in, is the load deviation of the local area; is the voltage deviation of the local area; α1 is the load deviation weight; α2 is the voltage deviation weight;

[0129] The global agent summarizes all local optimization results f local , perform global solution to obtain the topological open-close combination , output switching instructions (open and closed status) of flexible interconnected devices, dynamic load transfer plans, and charging and discharging plans of distributed power sources and energy storage.

[0130] In this embodiment, S5 is specifically as follows:

[0131] Based on the operation monitoring network and combined with the equipment status evaluation model, the equipment health score and failure probability are obtained;

[0132] Based on flexible interconnection devices (flexible switches, energy storage, power electronic interfaces), dynamically adjust the load flow or grid topology between stations to meet the needs of different operation scenarios;

[0133] Optimize load transfer paths and distribution based on equipment status, load distribution, and grid limitations, quickly identify and isolate faults through equipment status and real-time topology status, and reconstruct power supply paths using a flexible interconnection mechanism.

[0134] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0135] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0136] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0137] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0138] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.

Claims

1. A self-healing optimization control method for a substation based on flexible interconnection, characterized in that: The following steps are involved: S1: Build an operation monitoring network based on the equipment records in the substation area; S2: Deploy flexible interconnection equipment at key nodes in the substation area to achieve fault isolation and power outage transfer, dynamic load transfer and voltage regulation, and flexible coordinated control of distributed power sources and energy storage equipment; S3: Based on the substation operation data obtained from the operation monitoring network, the equipment status evaluation model is introduced to obtain the status evaluation and risk warning of each equipment in the substation; S4: Build a dynamic topology management mechanism for flexible interconnection between substations, and dynamically adjust the interconnection topology between substations according to load transmission requirements and power supply conditions; S5: Based on the status evaluation and risk warning of each device, load regulation and fault recovery are carried out according to the dynamic topology management mechanism of flexible interconnection; The equipment status evaluation model includes a health scoring model and a fault prediction model; the health scoring model calculates the equipment health score through multiple factors; the fault prediction model predicts the potential fault probability based on historical electrical fault data combined with the LSTM model; The health score model calculates the device health score through multiple factors, as follows: The equipment health score is calculated through multi-factor fusion, integrating operation data, environmental data and historical data to quantify the health status of the equipment. The multiple factors include dynamic indicators, historical indicators and environmental indicators; the dynamic indicators include load rate L f 、Temperature rise T r and runtime T u ; Historical indicators include the number of failures N f and the number of repairs N m ; Environmental indicators include ambient temperature T e and humidity H; Use weighted linearity to combine the indicators into a health score: ； Among them, HI is the health score; w1, w2, ..., w7 are weight coefficients; The health score is modified according to the set health level classification rules through fuzzy logic method; And introduce the degradation model to simulate the decline of health score: ； Where HI(t) is the health score of the equipment at time t; HI0 is the initial health score; λ is the degradation coefficient; t is the operating time; The S4 is specifically as follows: A multi-agent structure is built between substations, including local agents and global coordination agents. Local agents refer to the allocation of local agents to each substation area, which are responsible for monitoring the operation status of the substation area and preliminarily judging the load pressure; global coordination agents refer to the construction of agents at the station control center level, which summarize all substation information and perform global optimization; The local agent aggregates the status data at fixed time intervals; the local agents in the substation area propagate the parameters of adjacent areas to form regional status visualization; Design MAS-based topology optimization using an iterative coordination algorithm: State exchange: Each Agent exchanges the operating data of its own area; the adjacent areas form a short-term prediction matrix Dij, which is used to calculate the power that needs to be exchanged; Each local agent uses the data of the adjacent substation as a constraint to optimize the adjustment amount of the local flexible intelligent switch: ； in, is the load deviation of the local area; is the voltage deviation of the local area; α1 is the load deviation weight; α2 is the voltage deviation weight; The global agent summarizes all local optimization results f local , perform global solution to obtain the topological open-close combination , output switching instructions for flexible interconnected devices, dynamic load transfer plans, and charging and discharging plans for distributed power sources and energy storage.

2. The flexible interconnection-based area self-healing optimization control method according to claim 1 is characterized in that: The S1 is specifically: When building an operation monitoring network, first digitally enter the equipment ledger content as an initial reference for the network architecture; Smart meters, smart switches, and sensor equipment are deployed at key nodes in the substation area to collect equipment operation data and fault data in real time; Within the substation area, NB-IoT is used to collect real-time equipment operation data and fault data and aggregate them to the edge of the substation area.

3. The flexible interconnection-based area self-healing optimization control method according to claim 1 is characterized in that: The flexible interconnection device includes a flexible intelligent switch, a power electronic interface and an energy storage system, as follows: The flexible intelligent switch can realize fault isolation, power outage transfer and load transfer, and is used to quickly cut off the fault line or reconstruct the power supply path. When a line fault is detected, the fault load is isolated, and at the same time, combined with the power electronic interface, the load is transferred to the adjacent feeder or substation to realize power outage transfer. The power electronic interface realizes voltage regulation, harmonic control, power quality optimization and dynamic reactive power regulation, and supports grid-connected coordinated control of distributed power sources; When a local load surge causes a voltage drop, the power electronics interface dynamically injects reactive power to adjust the voltage, while filtering and compensating for harmonics caused by nonlinear loads to improve power quality. The energy storage system provides emergency power for the substation after a power outage. When the power supply load changes beyond a threshold, the energy storage system serves as a power supplement to improve the substation's ability to adapt to sudden load changes.

4. The flexible interconnection-based area self-healing optimization control method according to claim 3 is characterized in that: The deployment of flexible interconnection equipment at key nodes in the substation area is as follows: Install flexible intelligent switches in the feeder sections of the substation area to distinguish different load sections and reduce the impact of power outages and faults; install flexible intelligent switches at the contact points of adjacent substations to quickly switch the power supply path when a fault occurs, realizing inter-regional diversion and transfer; Flexible intelligent switches are installed on different branches of low-voltage lines to cut off only the faulty area when a small-scale fault occurs on the user side, ensuring the continuity of power supply. Power electronic interfaces are arranged at the grid connection points of distributed photovoltaic or wind power to smooth the fluctuating power generation and dynamically adjust reactive power; power electronic interfaces are arranged at the low-voltage busbar of the distribution transformer to achieve overall voltage regulation in the substation, compensate for reactive power and filter harmonics; power electronic interfaces are arranged at the load end position in the heavy load area to suppress the voltage drop and harmonic excess problems at the end during grid operation, and enhance the voltage stability of the substation; Deploy energy storage systems in areas where the load concentration is greater than the preset value to reduce the power supply pressure of the main grid during peak hours and optimize local power supply loads; The energy storage system is installed at the point where distributed photovoltaic and wind power are connected to the grid to store excess photovoltaic power generation during the day and release it during peak load periods to maximize the absorption of new energy. The energy storage system is installed at the junction between substations to provide short-term power supply to critical loads in the event of island operation or emergency power supply restoration in the substation, and to support dynamic load regulation of adjacent substations.

5. The flexible interconnection-based area self-healing optimization control method according to claim 1 is characterized in that: The fault prediction model predicts the potential fault probability based on historical electrical fault data combined with the LSTM model, as follows: Input feature vector X t =[L f ,T r ,I,V,T e ,H, fault label], where I and V are current and voltage respectively; The cell state of the LSTM model is updated: Forget Gate: ; Input Gate: ; Candidate values: ; Cell status update: ; Output Gate: ; Hide status updates: ; in, is the input feature vector at time step t; is the hidden state at time step t; For the Gate of Oblivion; is the input gate; is the candidate value, indicating the candidate memory cell at the current moment; is the cell state, indicating the memory cell state at the current moment; is the output gate; is the hidden state at the current moment; , , , They are the weight matrices of the forget gate, input gate, candidate value, and output gate respectively; , , , are the bias vectors of the forget gate, input gate, candidate value, and output gate, respectively. is the activation function; Train the LSTM model through the training data set and optimize the model parameters so that the model can learn the conditional probability relationship of the fault occurrence at the current time point. Use the cross entropy loss function to measure the error between the actual fault label and the predicted result. After the LSTM model training is completed, the failure probability P of the device at the future moment is predicted. fault : ； Among them, ht is the hidden state of the last layer of the LSTM model; W out , b are the weight and bias of the output layer respectively; sigmoid is the activation function, which maps the output to the range of (0, 1) as the failure probability; The potential failure probability P of the output device fault Finally, the risk levels are divided according to the following rules: Low risk: P fault <0.2, medium risk: 0.2≤P fault <0.5, high risk: P fault ≥0.

5.

6. The flexible interconnection-based area self-healing optimization control method according to claim 1 is characterized in that: The S5 is specifically as follows: Based on the operation monitoring network and combined with the equipment status evaluation model, the equipment health score and failure probability are obtained; Based on flexible interconnected devices, dynamically adjust the load flow or grid topology between stations to meet the needs of different operation scenarios; Optimize load transfer paths and distribution based on equipment status, load distribution, and grid limitations, quickly identify and isolate faults through equipment status and real-time topology status, and reconstruct power supply paths using a flexible interconnection mechanism.

7. A self-healing optimization control system for a substation based on flexible interconnection, characterized in that: It includes a processor, a memory and a computer program stored in the memory. When the processor executes the computer program, it specifically executes the steps in the flexible interconnection-based substation self-healing optimization control method as described in any one of claims 1 to 6.

Images