A power distribution network active splitting strategy optimization method

Abstract

The application provides a power distribution network active splitting strategy optimization method, comprising the following steps: calculating a reference deviation after obtaining a power grid data set, establishing a transfer mapping relationship table to obtain a measurable deviation interval after obtaining node inverter data, calculating island deviation and obtaining deviation jump amplitude in combination with the reference deviation after obtaining a candidate splitting section set, obtaining a self-checking node based on the deviation jump amplitude, the transfer mapping relationship table and the measurable deviation interval, calculating a self-checking coverage rate, obtaining a signal strength score based on the deviation jump amplitude and the measurable deviation interval, constructing a multi-objective optimization function, performing a Pareto solution to obtain an optimal splitting section set, obtaining a pre-regulation instruction in combination with the power grid data set, the deviation jump amplitude and the measurable deviation interval, obtaining a splitting execution scheme based on the pre-regulation instruction and in combination with the self-checking coverage rate, updating the transfer mapping relationship table based on the splitting execution scheme, and providing a high-reliability, self-checking and implementable active splitting strategy optimization guarantee.

Description

Technical Field

[0001] This invention relates to the field of distributed detection and control, and more specifically, to an optimization method for active disconnection strategy in distribution networks. Background Technology

[0002] In distribution networks, when a fault occurs in the main grid, certain areas need to be actively isolated to form islanded operations to ensure the reliability of power supply to critical loads. The core of active isolation is determining the appropriate isolation sections, i.e., selecting which switches to disconnect to form independent power supply areas. Currently, active isolation strategy optimization methods mainly focus on traditional objectives such as power balance within the island, minimizing the amount of load shedding, and reducing the number of switching operations, while insufficiently considering the autonomous identification capabilities of node inverters after islanding is formed.

[0003] At the moment of disconnection, the change in network topology will cause a jump in the detuning degree of each island. This jump will cause changes in voltage, frequency and phase, which can be detected locally by the node inverter. However, the range of electrical quantity changes that the node inverter can detect is limited. If the electrical quantity change caused by the jump is not within the detection range, the node inverter will not be able to reliably confirm the disconnection status and can only rely on communication, which poses a risk of misjudgment.

[0004] Existing traditional methods do not incorporate the detectability of detuning degree transitions into the optimization of active disconnection strategies, and lack a mechanism to actively adjust the arc suppression coil before disconnection to ensure self-test capability. Therefore, there is an urgent need for an active disconnection strategy optimization method that comprehensively considers self-test coverage, signal strength, and traditional objectives, and ensures reliable and identifiable disconnection status through active pre-adjustment and feedback correction. Summary of the Invention

[0005] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a method for optimizing an active disconnection strategy in a distribution network, the method comprising: S1: After obtaining the power grid dataset, calculate the baseline detuning degree. After obtaining the node inverter data, establish a transfer mapping table and combine it with the baseline detuning degree to obtain the measurable detuning degree range. S2: After obtaining the candidate detuning section set based on the power grid dataset, calculate the island detuning degree and combine it with the benchmark detuning degree to obtain the detuning degree jump amplitude. After obtaining the self-testable nodes based on the detuning degree jump amplitude, the transfer mapping relationship table and the measurable detuning degree interval, calculate the self-test coverage rate. S3: Based on the detuning jump amplitude and the measurable detuning interval, the signal strength score is obtained. After combining the traditional target parameters and the list of key equipment, a multi-objective optimization function is constructed, and Pareto solution is performed to obtain the optimal solution profile set. S4: Based on the optimal disengagement section set and combined with the power grid dataset, detuning degree jump amplitude and measurable detuning degree interval, calculate the reversal current regulation and compensation current regulation to obtain the pre-regulation command, and obtain the disengagement execution scheme based on the pre-regulation command and combined with the self-test coverage. S5: Based on the decoupling execution scheme, obtain the actual jump amplitude value and update the transfer mapping relationship table in combination with the measurable detuning interval.

[0006] Specifically, after acquiring the power grid dataset, the baseline detuning degree is calculated. After acquiring the node inverter data, a transfer mapping table is established and combined with the baseline detuning degree to obtain the measurable detuning degree range, including: Obtain the power grid dataset of the target distribution network, and accumulate the feeder-to-ground current data and coil compensation current data respectively to obtain the total capacitor current and the total compensation current in order to calculate the reference detuning degree; After acquiring the inverter data of each PCC node, a detuning degree jump disturbance is applied and the detuning degree change is recorded. At the same time, the voltage change, frequency offset and phase jump of each PCC node are recorded. The conduction coefficient set is then determined by least squares fitting. Based on the set of conduction coefficients, establish the mapping relationship between the detuning degree change and the voltage change, frequency offset and phase jump variable respectively, and combine them into the corresponding PCC node's transmission mapping relationship table; After establishing the effective range of voltage signal, effective range of frequency signal, and effective range of phase transition based on the inverter data at each node, the measurable detuning range of each PCC node is established by combining the transfer mapping relationship table.

[0007] Specifically, after obtaining the candidate detuning section set based on the power grid dataset, the island detuning degree is calculated, and the detuning degree jump amplitude is obtained by combining it with the benchmark detuning degree. Based on the detuning degree jump amplitude, the transfer mapping relationship table, and the measurable detuning degree interval, the self-testable nodes are obtained, and the self-test coverage rate is calculated, including: The candidate disconnection section set is obtained based on the target distribution network topology data. The candidate disconnection section set represents the set of all possible disconnection sections of the target distribution network when a fault occurs. Each candidate disconnection section contains a set of switches to be disconnected. After identifying the islands divided by each candidate decoupling section and recording the feeders and PCC nodes they contain, the island capacitance current and island compensation current of each island are obtained by combining the feeder-to-ground current data and coil compensation current data to calculate the island detuning degree of the corresponding island. The difference between the island detuning degree and the reference detuning degree is used to obtain the detuning degree jump value corresponding to each island. Based on the detuning jump amplitude value and combined with the transfer mapping relationship table and the measurable detuning interval, detectability is determined to obtain self-testable nodes and calculate the self-test coverage.

[0008] Specifically, based on the detuning jump amplitude and combined with the transfer mapping relationship table and the measurable detuning interval, detectability is determined to obtain self-testable nodes and calculate the self-test coverage, including: Substitute the detuning jump amplitude value into the transfer mapping table to obtain the predicted voltage change, predicted frequency offset, and predicted phase jump of each PCC node in the island. If the predicted voltage change, predicted frequency offset, and predicted phase jump all fall within the corresponding effective range of the voltage signal, the effective range of the frequency signal, and the effective range of the phase jump, respectively, then the PCC node is determined to be a self-testable node. Here, self-testable means that the node inverter at the PCC node can autonomously determine whether it has entered the disconnection state. The number of self-checkable nodes and PCC nodes in each isolated island are counted to calculate the self-check coverage rate of the corresponding candidate disconnection section. The self-check coverage rate is used to characterize the ability of the candidate disconnection section to autonomously identify the disconnection status.

[0009] Specifically, after obtaining signal strength scores based on the detuning jump amplitude and the measurable detuning interval, and constructing a multi-objective optimization function by combining traditional target parameters and a list of key equipment, Pareto optimization is performed to obtain the optimal solution profile set, including: Obtain traditional target parameters and a list of key equipment. The traditional target parameters include at least the load shedding amount and the number of switching operations. The list of key equipment is a set of key equipment pre-configured according to the importance of the equipment and the protection range. The difference between the detuning jump amplitude and the boundary of the measurable detuning interval is calculated as the margin value of the corresponding PCC node. The margin values ​​of all PCC nodes in the island are weighted and summed to obtain the signal strength score of the candidate solution section. After filtering candidate solution profiles based on the list of key equipment, a multi-objective optimization function is constructed by combining signal strength score, self-test coverage rate and traditional target parameters to obtain the optimal solution profile set.

[0010] Specifically, after screening candidate solution profile sets based on the list of key equipment, a multi-objective optimization function is constructed by combining signal strength scores, self-test coverage, and traditional target parameters to obtain the optimal solution profile set, including: If there is a PCC node in the list of critical devices within the island and it is not marked as a self-testable node, then the candidate disconnection section corresponding to that island will be removed from the candidate disconnection section set. For each candidate segment in the candidate segment set, a multi-objective optimization function is constructed with the optimization objectives of maximizing signal strength score, maximizing self-test coverage, minimizing load shedding, and minimizing the number of switching operations. After generating all non-dominated solutions of the multi-objective optimization function on the Pareto front using a multi-objective optimization algorithm, the optimal solution profile set is obtained. The non-dominated solution indicates that there are no other candidate solution profiles that are better than the candidate solution profiles corresponding to this solution in each optimization objective.

[0011] Specifically, for S4, this includes: Based on the target distribution network topology data, determine all adjacent island pairs for each optimal solution section. If the detuning jump amplitude values ​​between adjacent island pairs have the same sign, then merge the corresponding adjacent island pairs; otherwise, do not perform any operation. Among them, the islands with larger adjustment margins of the arc suppression coils in adjacent islands that cannot be merged are taken as islands to be flipped, and the flipping current adjustment is calculated to flip their detuning jump amplitude value. The intersection of the measurable detuning intervals corresponding to all PCC nodes in each island in the optimal solution section is taken as the island detuning interval of that island. If the detuning degree jump amplitude value corresponding to the island does not fall within the island detuning degree interval, the difference between the detuning degree jump amplitude value and the median value of the island detuning degree interval is calculated to obtain the compensation current adjustment amount; otherwise, no processing is performed. Based on the reversing current regulation and the compensation current regulation, and combined with the power grid dataset, the pre-regulation command is obtained. After the feasibility of the pre-regulation command is tested, the disconnection execution plan is obtained by combining the self-test coverage.

[0012] Specifically, based on the switching current regulation and the compensation current regulation, and combined with the power grid dataset, pre-regulation commands are obtained. After performing feasibility checks on the pre-regulation commands, a tripping execution plan is obtained by combining the self-check coverage, including: The flip current adjustment amount and compensation current adjustment amount are converted into the gear adjustment steps by the coil gear data, and the adjustment direction is recorded and then combined into a pre-adjustment command. After obtaining the current gear, the target gear is obtained by combining the pre-adjustment command. Based on the coil gear data, it is determined whether the target gear exceeds the gear adjustment range of the arc suppression coil. The optimal solution section is discarded according to the judgment result. The optimal solution section with the highest self-inspection coverage is selected as the solution execution scheme.

[0013] Specifically, the method further includes: The power grid dataset includes feeder-to-ground current data, coil compensation current data, coil tap data, and target distribution network topology data; The feeder-to-ground current data is represented as the ground capacitance current value of each feeder in the target distribution network. The coil compensation current data is represented as the compensation current value of each setting of the arc suppression coil; The coil gear data refers to the gear step size and gear adjustment range of the arc suppression coil. The target distribution network topology data is represented by the electrical connection relationships of the target distribution network. The node inverter data refers to the minimum voltage amplitude change, minimum frequency offset, and minimum phase jump that the inverter at the corresponding PCC node can detect, as well as the maximum voltage amplitude change, maximum frequency offset, and maximum phase jump that it can withstand.

[0014] Compared with the prior art, the present invention has the following beneficial effects: This invention establishes a transmission mapping relationship between detuning degree jump and detectable electrical quantities of node inverters, and combines the measurable detuning degree range of each PCC node to incorporate the self-checking capability of the detuning state into the optimization dimension of the active detuning strategy. This enables the determination of whether each node inverter can autonomously identify the detuning state at the detuning section design stage, and solves the technical defect of traditional methods that only discover the detection failure after detuning is completed.

[0015] The design incorporates dual quantitative indicators of self-test coverage and signal strength score, elevating self-testability to an optimization objective on par with traditional targets. This enables a comprehensive evaluation of candidate disconnection sections in terms of both detection coverage and detection reliability. It avoids the problem of focusing solely on power balance and load shedding while neglecting the detection capability of node inverters after islanding formation, thus eliminating the risk of optimal disconnection sections failing to detect.

[0016] A dual pre-adjustment mechanism of switching current adjustment and compensation current adjustment is constructed. Combined with the active adjustment and feasibility detection of the arc suppression coil position, active control of the detuning jump direction and amplitude before disconnection is realized, providing a highly reliable, self-testable and implementable active disconnection strategy optimization guarantee. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the steps of an active disconnection strategy optimization method for a distribution network provided in an embodiment of the present invention. Figure 2 This is a flowchart of a distribution network active disconnection strategy optimization method provided by an embodiment of the present invention. Detailed Implementation

[0018] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating the steps of an active disconnection strategy optimization method for a distribution network provided in an embodiment of the present invention. Figure 2 This is a flowchart of a distribution network active disconnection strategy optimization method provided by an embodiment of the present invention. The following is a detailed description of this distribution network active disconnection strategy optimization method.

[0019] Specifically, an optimization method for an active disconnection strategy in a distribution network includes the following steps: Step S1: After obtaining the power grid dataset, calculate the reference detuning degree. After obtaining the node inverter data, establish a transfer mapping relationship table and combine it with the reference detuning degree to obtain the measurable detuning degree range.

[0020] In this embodiment, step S1 specifically includes steps S1-1 to S1-4: Step S1-1: Obtain the power grid dataset of the target distribution network, and accumulate the feeder-to-ground current data and coil compensation current data to obtain the total capacitor current and total compensation current in order to calculate the reference detuning degree.

[0021] The power grid dataset is obtained by accessing the power grid data acquisition system of the target power grid. The power grid dataset includes feeder-to-ground current data, coil compensation current data, coil tap data, and target power grid topology data. The feeder-to-ground current data represents the ground capacitance current value of each feeder in the target power grid. The coil compensation current data represents the compensation current value of each tap of the arc suppression coil. The coil tap data represents the tap step size and tap adjustment range of the arc suppression coil. The target power grid topology data represents the electrical connection relationship of the target power grid.

[0022] The total capacitive current of the target distribution network is obtained by summing the feeder-to-ground current data of each feeder in the target distribution network. The total compensation current is obtained by summing the coil compensation current data of each arc suppression coil in the target distribution network. The reference detuning degree is calculated based on the total capacitive current and the total compensation current. The calculation formula is as follows: ; in, Represented as the baseline detuning degree, This is expressed as the total compensation current. It is expressed as the total capacitor current.

[0023] Understandably, the benchmark detuning degree reflects the overall compensation status of the target distribution network under normal grid-connected operation. If the benchmark detuning degree is positive, it indicates that the target distribution network is in an over-compensated state; if the benchmark detuning degree is negative, it indicates that the target distribution network is in an under-compensated state; if the benchmark detuning degree is zero, it indicates that the target distribution network is in a fully compensated state.

[0024] Step S1-2: After obtaining the node inverter data of each PCC node, apply the detuning degree jump disturbance and record the detuning degree change. At the same time, record the voltage change, frequency offset and phase jump of each PCC node, and then determine the conduction coefficient set by least squares fitting.

[0025] Specifically, the node inverter data of each PCC node is obtained. The node inverter data is represented as the minimum voltage amplitude change, minimum frequency offset, and minimum phase jump that the inverter at the corresponding PCC node can detect, as well as the maximum voltage amplitude change, maximum frequency offset, and maximum phase jump that it can withstand.

[0026] By adjusting the tap position of the arc suppression coil in the target distribution network, multiple fixed amplitude detuning degree jump disturbances are applied. The fixed amplitude is preset according to the detuning degree jump range to cover the detuning degree jump range that may be generated in subsequent steps. For example, in a possible embodiment, the detuning degree jump range of each island usually falls between 0.05 and 0.25, so it can be preset to five fixed amplitudes of 0.05, 0.10, 0.15, 0.20 and 0.25.

[0027] Each applied detuning jump disturbance generates a known detuning change. Simultaneously, the local detection module of the inverter at the PCC node records the voltage change, frequency offset, and phase jump caused by the detuning jump disturbance at the corresponding PCC node.

[0028] After establishing the mapping relationship between the detuning degree change and the voltage change, frequency offset and phase jump variable respectively, the voltage conduction coefficient, frequency conduction coefficient and phase conduction coefficient of the PCC node are determined by least squares fitting of the voltage change, frequency offset and phase jump variable corresponding to each fixed amplitude.

[0029] The voltage conduction coefficient, frequency conduction coefficient, and phase conduction coefficient constitute the conduction coefficient set of the corresponding PCC node.

[0030] Step S1-3: Based on the set of conduction coefficients, establish the mapping relationship between the detuning degree change and the voltage change, frequency offset and phase jump variable respectively, and combine them into the corresponding PCC node's transmission mapping relationship table.

[0031] For each PCC node, a detuning-electric mapping relationship is established between the detuning degree change and the voltage change based on its voltage conduction coefficient: ,in, This is expressed as a change in voltage. It is expressed as the voltage conduction coefficient. It is expressed as the change in detuning degree. The detuning-electric mapping relationship is used to characterize the neutral point voltage shift caused by the change in detuning degree at the moment of decoupling.

[0032] Establish a frequency detuning mapping relationship between the detuning degree change and the frequency offset based on the frequency conduction coefficient: ,in, This is expressed as a frequency offset. Expressed as frequency conduction coefficient, It is expressed as the change in detuning degree, and the detuning-frequency mapping relationship is used to characterize the quantitative relationship between the instantaneous change in detuning degree and the frequency offset.

[0033] Based on the phase conduction coefficient, establish the detuning-phase mapping relationship between the detuning degree change and the phase jump variable: ,in, Represented as a phase jump variable, Represented as the phase transmission coefficient, It is expressed as the change in detuning degree. The detuning-phase mapping relationship is used to characterize the quantitative relationship between the impact of the detuning degree jump on the voltage phase and the phase jump at the moment of detuning.

[0034] The detuning-electricity mapping relationship, detuning-frequency mapping relationship, and detuning-phase mapping relationship are combined into a transfer mapping relationship table for the corresponding PCC node. The transfer mapping relationship table transforms the detuning degree change, which was originally not directly perceived by the node inverter, into voltage change, frequency offset, and phase jump variable that can be detected by AD sampling.

[0035] Step S1-4: After establishing the effective range of voltage signal, effective range of frequency signal, and effective range of phase transition based on the node inverter data, establish the measurable detuning range of each PCC node in conjunction with the transfer mapping relationship table.

[0036] First, based on the node inverter data, the effective range of the voltage signal is established with the minimum voltage amplitude change that the node inverter can detect as the lower limit and the maximum voltage amplitude change that it can withstand as the upper limit; the effective range of the frequency signal is established with the minimum frequency offset as the lower limit and the maximum frequency offset as the upper limit; and the effective range of the phase jump is established with the minimum phase jump as the lower limit and the maximum phase jump as the upper limit.

[0037] Then, the effective range of the voltage signal is back-derived to the detuning level through the detuning mapping relationship to obtain the voltage back-derived range, which represents the detuning jump range that can be effectively identified by the voltage detection channel of the node inverter.

[0038] By using the detuning-frequency mapping relationship, the effective range of the frequency signal is back-calculated to the detuning level to obtain the frequency back-calculation range. The frequency back-calculation range represents the detuning jump range that can be effectively identified by the frequency detection channel of the node inverter.

[0039] The effective phase jump range is back-calculated to the detuning degree level by using the phase-detuning mapping relationship to obtain the phase back-calculation range. The phase back-calculation range represents the detuning degree jump range that can be effectively identified by the phase detection channel of the node inverter.

[0040] The measurable detuning interval of the corresponding PCC node is obtained by taking the intersection of the voltage back-calculation interval, the frequency back-calculation interval, and the phase back-calculation interval.

[0041] Step S1 establishes a transfer mapping table and a measurable detuning interval, enabling the node inverter to identify the detuning jump corresponding to the detected voltage change, frequency offset, and phase jump. This achieves cross-domain coupling between two independent technical branches that originally belonged to the power distribution field and the power grid connection field through the transfer mapping table, and brings the detection blind zone problem ignored in traditional decoupling optimization to the decoupling section design stage for resolution.

[0042] Step S2: After obtaining the candidate detuning section set based on the power grid dataset, calculate the island detuning degree and obtain the detuning degree jump amplitude value in combination with the benchmark detuning degree. After obtaining the self-testable nodes based on the detuning degree jump amplitude value, the transfer mapping relationship table and the measurable detuning degree interval, calculate the self-test coverage rate.

[0043] In this embodiment, step S2 specifically includes steps S2-1 to S2-4: Step S2-1: Obtain a set of candidate disconnection sections based on the target distribution network topology data. The set of candidate disconnection sections represents the set of all possible disconnection sections of the target distribution network when a fault occurs. Each candidate disconnection section contains a set of switches to be disconnected.

[0044] Based on the target distribution network topology data, a depth-first search algorithm is used to traverse all on / off combinations of switches in the target distribution network to determine all possible disconnection switch combinations that can divide the target distribution network into multiple independent power supply areas. Each set of switches to be disconnected is taken as a candidate disconnection section. All candidate disconnection sections form a candidate disconnection section set, which represents the set of all possible disconnection sections of the target distribution network when a fault occurs.

[0045] Step S2-2: After determining the islands divided by each candidate decoupling section and recording the feeders and PCC nodes they contain, combine the feeder-to-ground current data and coil compensation current data to obtain the island capacitance current and island compensation current of each island in order to calculate the island detuning degree of the corresponding island.

[0046] When the candidate disconnection section is executed, the target distribution network is divided into multiple independent power supply areas. Each independent power supply area is an island. The islands divided by each candidate disconnection section are determined, and the feeders and PCC nodes contained in each island are recorded.

[0047] After extracting the ground capacitance current values ​​corresponding to all feeders contained in each island from the feeder ground current data, the island capacitance current of the island is obtained by summing them up; after extracting the compensation current values ​​of the current tap of all arc suppression coils in each island from the coil compensation current data, the island compensation current of the island is obtained by summing them up.

[0048] Based on the island capacitor current and island compensation current, and using the calculation method in step S1-1, the island detuning degree of the corresponding island is calculated.

[0049] Step S2-3: Subtract the island detuning degree from the reference detuning degree to obtain the detuning degree jump value corresponding to each island.

[0050] For each island in each candidate detuning section, the detuning degree of the corresponding island is subtracted from the reference detuning degree to obtain the detuning degree jump amplitude value of the island. The detuning degree jump amplitude value represents the amount of change in the detuning degree of the island at the moment of detuning. The larger the absolute value, the more drastic the change in the neutral point grounding state before and after detuning, which is converted into a greater local detectable signal strength.

[0051] Step S2-4: Based on the detuning jump amplitude value and combined with the transfer mapping relationship table and the measurable detuning interval, determine the detectability to obtain self-testable nodes and calculate the self-test coverage.

[0052] Specifically, step S2-4 includes steps S2-4-1 to S2-4-3: Step S2-4-1: Substitute the detuning jump amplitude value into the transfer mapping table to obtain the predicted voltage change, predicted frequency offset, and predicted phase jump of each PCC node in the island.

[0053] Substitute the detuning jump amplitude value corresponding to each island into the transfer mapping table corresponding to each PCC node within its range, and obtain the predicted voltage change, predicted frequency offset, and predicted phase jump of the PCC node through the detuning-electricity mapping relationship, detuning-frequency mapping relationship, and detuning-phase mapping relationship, respectively.

[0054] Step S2-4-2: If the predicted voltage change, predicted frequency offset, and predicted phase jump all fall within the corresponding effective range of the voltage signal, the effective range of the frequency signal, and the effective range of the phase jump, respectively, then the PCC node is determined to be a self-testable node. Here, self-testable means that the node inverter at the PCC node can autonomously determine whether it has entered the disconnection state.

[0055] If the predicted voltage change of the PCC node falls within the effective range of the voltage signal, the predicted frequency offset falls within the effective range of the frequency signal, and the predicted phase jump falls within the effective range of the phase jump, then the PCC node has the effective detection capability for detuning degree jump under the current candidate desegmentation section. At this time, the PCC node is determined to be a self-testable node. Otherwise, it indicates that the PCC node does not have the effective detection capability for detuning degree jump under the current candidate desegmentation section. Here, self-testable means that the node inverter at the PCC node can autonomously determine whether it has entered the desegmentation state by measuring the changes in voltage, frequency, and phase.

[0056] Step S2-4-3: Count the number of self-checkable nodes and PCC nodes in each island to calculate the self-check coverage rate of the corresponding candidate disconnection section. The self-check coverage rate is used to characterize the ability of the candidate disconnection section to autonomously identify the disconnection status.

[0057] The number of self-checkable nodes and the number of PCC nodes in each island under the candidate disconnection section are counted and summed to obtain the total number of self-checkable nodes and the total number of PCC nodes of the candidate disconnection section. The ratio of the total number of self-checkable nodes to the total number of PCC nodes is calculated as the self-check coverage rate of the corresponding candidate disconnection section. The self-check coverage rate is used to characterize the ability of the candidate disconnection section to autonomously identify the disconnection status.

[0058] Step S2 is the bridge between data modeling and quantitative evaluation in this method. Through step S2, each candidate disconnection section obtains a self-test coverage rate that is simultaneously constrained by the signal strength of the detuning degree jump amplitude and the sensitivity constraint of the node inverter detection boundary, and also satisfies the autonomous identification conditions of the disconnection status of each PCC node. This achieves adaptive evaluation of the quantitative index of the self-test capability of each candidate disconnection section, avoiding the problem that traditional disconnection strategies only focus on power balance and load shedding while ignoring whether the node inverter can autonomously detect the disconnection status after islanding. Compared with existing passive islanding detection methods, the self-testability constraint of this method starts from the design stage of the candidate disconnection section, laying the foundation for multi-objective section optimization in subsequent steps.

[0059] Step S3: Based on the detuning jump amplitude and the measurable detuning interval, obtain the signal strength score, and combine it with traditional target parameters and the list of key equipment to construct a multi-objective optimization function. Then, perform Pareto solution to obtain the optimal solution profile set.

[0060] In this embodiment, step S3 specifically includes steps S3-1 to S3-3: Step S3-1: Obtain the traditional target parameters and the list of key equipment. The traditional target parameters include at least the load shedding amount and the number of switching operations. The list of key equipment is a set of key equipment pre-configured according to the importance of the equipment and the protection range.

[0061] Traditional target parameters are obtained, including at least the load shedding amount and the number of switching operations. The load shedding amount represents the total amount of load that loses power from the main grid in each island after the candidate disconnection is executed. The smaller the load shedding amount, the better the effect of the candidate disconnection in maintaining the stable operation of the island. The number of switching operations represents the number of switches required to execute the candidate disconnection. The smaller the number of switching operations, the smaller the disturbance to the target distribution network operation mode and the faster the execution speed of the candidate disconnection.

[0062] Obtain a list of critical devices, which is a set of critical devices pre-configured according to their importance and protection scope. For example, in a specific embodiment, if the protection device corresponding to a certain PCC node still needs to maintain normal operation in the disconnected state and its fault diagnosis depends on the timely confirmation of the disconnected state, then the PCC node is included in the list of critical devices.

[0063] Step S3-2: Calculate the difference between the detuning jump amplitude and the boundary of the measurable detuning interval as the margin value of the corresponding PCC node, and sum the margin values ​​of all PCC nodes in the islands by weight to obtain the signal strength score of the candidate decoupling section.

[0064] Specifically, for each PCC node within each island, the difference between the detuning degree jump amplitude of the island and the upper and lower boundaries of the measurable detuning degree interval of the PCC node is calculated. The smaller of these values ​​is taken as the margin value of the PCC node. The larger the margin value, the farther the detuning degree jump amplitude of the island is from the detection capability boundary of the corresponding PCC node, and the more accurate the detection result of the PCC node for this detuning degree jump and the stronger its anti-interference capability.

[0065] The weight coefficient of each PCC node is preset based on the list of critical equipment. For example, if a PCC node is in the list of critical equipment, its weight coefficient can be set to 1.5; if a PCC node is not in the list of critical equipment, its weight coefficient can be set to 1.0.

[0066] The signal strength score of the candidate disconnection section is obtained by weighting and summing the margin values ​​of each PCC node in all islands under the candidate disconnection section according to their corresponding weight coefficients.

[0067] Step S3-3: After screening the candidate solution profile set based on the list of key equipment, construct a multi-objective optimization function by combining signal strength score, self-test coverage rate and traditional target parameters to obtain the optimal solution profile set.

[0068] Specifically, step S3-3 includes steps S3-3-1 to S3-3-3: Step S3-3-1: If there is a PCC node in the list of critical devices in the island and it is not marked as a self-testable node, then remove the candidate disconnection section corresponding to the island from the candidate disconnection section set.

[0069] For each candidate disconnection section, traverse all PCC nodes within all the islands it divides, and check whether each PCC node is in the critical equipment list. If an island contains a PCC node from the critical equipment list and that PCC node is not marked as a self-testable node, it indicates that after the corresponding candidate disconnection section is implemented, the island where the critical equipment is located cannot autonomously identify its own disconnection status, and there is a risk of power supply abnormality due to misjudgment of the disconnection status. In this case, the candidate disconnection section is removed from the candidate disconnection section set. If all PCC nodes in the critical equipment list in each island have been marked as self-testable nodes, or if there are no PCC nodes in the critical equipment list in any island, it indicates that the corresponding candidate disconnection section has passed the hard constraint screening of the critical equipment list, and the candidate disconnection section is retained in the candidate disconnection section set.

[0070] Step S3-3-2: For each candidate segment in the candidate segment set, construct a multi-objective optimization function with the optimization objectives of maximizing signal strength score, maximizing self-test coverage, minimizing load shedding, and minimizing the number of switching operations.

[0071] For each candidate disconnection segment retained in the candidate disconnection segment set, its corresponding signal strength score, self-test coverage, load shedding amount, and number of switching operations are extracted. A multi-objective optimization function is then constructed, with the optimization objectives being to maximize the signal strength score, maximize the self-test coverage, minimize the load shedding amount, and minimize the number of switching operations. The expression of the multi-objective optimization function is as follows: ; in, Represented as a multi-objective optimization function, This is represented as a candidate solution section. This is represented as a signal strength score. This is expressed as the self-inspection coverage rate. This represents the amount of load removed. This represents the number of switch operations.

[0072] Understandably, the higher the signal strength score, the closer the detuning degree jump generated by the corresponding candidate decoupling section is to the center region of the measurable detuning degree interval corresponding to each PCC node. At this time, the robustness to measurement error and response deviation is stronger, and the reliability of the decoupling state self-test is higher. Therefore, maximizing the signal strength score is the optimization objective.

[0073] The higher the self-check coverage, the more PCC nodes in the corresponding candidate disconnection section can perform disconnection status self-checks. At this time, the reliability of disconnection status confirmation is high. Therefore, maximizing the self-check coverage is the optimization goal.

[0074] The lower the load shedding amount, the better the corresponding candidate disconnection section is in maintaining the stable operation of the island and the smaller the impact on power supply. Therefore, minimizing the load shedding amount is the optimization objective.

[0075] The lower the number of switching operations, the smaller the disturbance to the target distribution network operation mode and the faster the execution speed of the corresponding candidate disconnection section. This is conducive to quickly completing the disconnection after a fault occurs. Therefore, minimizing the number of switching operations is the optimization objective.

[0076] It should be noted that there are conflicts among the four optimization objectives mentioned above, so there is no single optimal solution. For example, maximizing the self-test coverage may require the candidate solution segment to divide more islands to optimize the signal of detuning transition, but at the same time it will increase the number of switching operations; minimizing the load shedding requires the candidate solution segment to retain more feeders, but at the same time it may reduce the self-test coverage. Therefore, it is necessary to generate a set of non-dominated solutions through Pareto solution.

[0077] Step S3-3-3: After generating all non-dominated solutions of the multi-objective optimization function on the Pareto front using a multi-objective optimization algorithm, the optimal solution profile set is obtained. The non-dominated solution indicates that there are no other candidate solutions that are better than the candidate solution profile corresponding to this solution in each optimization objective.

[0078] A multi-objective optimization algorithm is selected, such as a decomposition-based multi-objective evolutionary algorithm or an intensity Pareto evolutionary algorithm. The multi-objective optimization algorithm is used to perform Pareto solving on the candidate solution profile set to generate all non-dominated solutions of the multi-objective optimization function on the Pareto front. Each non-dominated solution is an optimal solution profile. All optimal solution profiles form the optimal solution profile set. The non-dominated solution indicates that there are no other candidate solutions that are better than the candidate solution profile corresponding to this solution in all optimization objectives.

[0079] Step S3 is a crucial decision-making step in this method, transforming the quantitative evaluation results of candidate disengagement sections into the selection of the optimal disengagement section. Self-inspectability is elevated from a qualitative constraint to an optimization dimension on par with traditional objective parameters. Self-inspection coverage represents the proportion of self-inspectable nodes, while signal strength score represents the comprehensive level of detuning degree jump signal margin. These two complementary aspects constitute the performance of candidate disengagement sections in autonomous disengagement state identification, avoiding the risk of optimal disengagement sections failing to detect in traditional disengagement methods. Simultaneously, a hard constraint mechanism ensures the reliability of key equipment in the disengagement state. Finally, the four optimization objective dimensions of the multi-objective optimization function are clarified, the multi-objective optimization function is constructed, and all non-dominated solutions on the Pareto front are generated, thus forming the optimal disengagement section set. This provides a foundation for determining the optimal disengagement section that takes into account multi-dimensional optimization objectives in subsequent disengagement execution scheme determination.

[0080] Step S4: Based on the optimal decoupling section set and combined with the power grid dataset, detuning degree jump amplitude and measurable detuning degree interval, calculate the reversing current regulation and compensation current regulation to obtain the pre-regulation command, and obtain the decoupling execution scheme based on the pre-regulation command and combined with the self-test coverage.

[0081] In this embodiment, step S4 specifically includes steps S4-1 to S4-5: Step S4-1: Based on the target distribution network topology data, determine all adjacent island pairs for each optimal solution section. If the detuning jump amplitude values ​​between adjacent island pairs have the same sign, then merge the corresponding adjacent island pairs; otherwise, do not perform any operation.

[0082] Specifically, based on the target distribution network topology data, the adjacency relationship between the islands divided by each optimal solution section in the optimal solution section set is determined, and two islands with a direct electrical connection relationship are regarded as adjacent island pairs.

[0083] The signs of the detuning jump amplitudes of the two islands in an adjacent island pair are compared. If the signs of the detuning jump amplitudes of the two islands in the adjacent island pair are the same, it indicates that at the moment of disconnection, the frequency drift directions of the two islands in the adjacent island pair are the same, and the compensation states are homogeneous. The node inverter cannot identify the directional difference through the frequency detection channel. In this case, the adjacent island pair needs to be merged into a whole. Otherwise, it indicates that at the moment of disconnection, the frequency drift directions of the two islands in the adjacent island pair are complementary, and the compensation states are different. The node inverter can identify the directional difference through the frequency detection channel. Therefore, the two islands are kept to operate independently, and no merging operation is performed.

[0084] Step S4-2: Among them, the island with a larger adjustment margin of the arc suppression coil in the adjacent islands that cannot be merged is taken as the island to be flipped, and the flipping current adjustment amount is calculated to flip its detuning jump amplitude value.

[0085] Understandably, two adjacent islands may be unable to merge due to switch failures, operational limitations, or equipment status. In this case, the adjustment margin of each island is calculated based on the coil compensation current data and the coil speed data. The adjustment margin is expressed as the remaining proportion of the compensation current value that can be further adjusted at the current arc suppression coil speed relative to its total adjustable range. The calculation formula is as follows: ; in, This is expressed as adjustment margin. This is expressed as the maximum compensation current value. This is represented as the current compensation current value. This is represented as the minimum compensation current value.

[0086] Compare the adjustment margins of adjacent islands to the inner islands, and select the island with the larger adjustment margin as the island to be flipped. Take twice the negative value of the detuning jump amplitude corresponding to the island to be flipped, and multiply it by the island capacitor current obtained in step S2-2 to obtain the flipping current adjustment amount. When the flipping current adjustment amount is positive, it indicates that the corresponding arc suppression coil needs to increase the compensation current to achieve the flipping of the detuning jump amplitude; when the flipping current adjustment amount is negative, it indicates that the corresponding arc suppression coil needs to decrease the compensation current to achieve the flipping of the detuning jump amplitude.

[0087] It should be noted that the reason for flipping the detuning jump amplitude value of the island to be flipped is to make the detuning jump amplitude values ​​of the two adjacent islands show opposite changing trends, so that the node inverter can identify the directional difference through the frequency detection channel.

[0088] Step S4-3: Take the intersection of the measurable detuning intervals corresponding to all PCC nodes in each island in the optimal solution section as the island detuning interval of that island.

[0089] For each island divided by the optimal solution section in the optimal solution section set, read the measurable detuning degree interval corresponding to each PCC node in the island, and take the intersection of all the above measurable detuning degree intervals as the island detuning degree interval of the island. It should be noted that taking the intersection is to ensure that when the detuning degree jump amplitude value falls within the interval, all PCC nodes in the island can independently and reliably detect the detuning degree jump amplitude value, so as to avoid communication dependence on the confirmation of the solution status due to the inability of some PCC nodes to detect independently.

[0090] Step S4-4: If the detuning degree jump amplitude value corresponding to the island does not fall within the island detuning degree interval, calculate the difference between the detuning degree jump amplitude value and the median value of the island detuning degree interval to obtain the compensation current adjustment amount; otherwise, no processing is performed.

[0091] Specifically, for each island divided by the optimal solution section in the optimal solution section set, its detuning degree jump amplitude is compared with the island's detuning degree interval. If the detuning degree jump amplitude corresponding to the island does not fall within the island's detuning degree interval, it indicates that there are PCC nodes in the island that cannot effectively detect the detuning degree jump amplitude. In this case, the difference between the detuning degree jump amplitude and the median of the island's detuning degree interval is calculated and multiplied by the island's capacitor current to obtain the compensation current adjustment amount. If the detuning degree jump amplitude corresponding to the island falls within the island's detuning degree interval, it indicates that all PCC nodes in the island can autonomously detect the detuning degree jump amplitude, and no adjustment is required.

[0092] It should be noted that if the island is an island to be flipped, when comparing the detuning degree jump value with the island's detuning degree range, the detuning degree jump value after flipping should be used for comparison with the island's detuning degree range and for the calculation of subsequent compensation current adjustment.

[0093] Step S4-5: Based on the flip current regulation and compensation current regulation and combined with the power grid dataset, obtain the pre-regulation command, perform feasibility testing on the pre-regulation command, and then obtain the disconnection execution plan based on the self-test coverage.

[0094] Specifically, step S4-5 includes steps S4-5-1 to S4-5-3: Step S4-5-1: Convert the reversing current adjustment amount and the compensation current adjustment amount into the number of adjustment steps using the coil gear data, record the adjustment direction, and then combine them into a pre-adjustment command.

[0095] Specifically, based on the gear step size corresponding to each gear of the arc suppression coil in the coil gear data, the reversing current adjustment amount and the compensation current adjustment amount are divided by the gear step size and rounded to obtain the gear adjustment steps for the reversing current adjustment amount and the compensation current adjustment amount respectively. At the same time, the adjustment direction is determined according to the sign of the reversing current adjustment amount and the compensation current adjustment amount. When the sign is positive, the adjustment direction is to increase the gear; when the sign is negative, the adjustment direction is to decrease the gear.

[0096] The gear adjustment steps and adjustment direction are combined to form the pre-adjustment command for the island. If the island is to be flipped, the gear adjustment steps and adjustment direction corresponding to the flipping current adjustment and compensation current adjustment must be recorded respectively.

[0097] Step S4-5-2: After obtaining the current gear, combine it with the pre-adjustment command to obtain the target gear. Based on the coil gear data, determine whether the target gear exceeds the gear adjustment range of the arc suppression coil. Discard the optimal solution section according to the judgment result.

[0098] Read the current gear of the arc suppression coil corresponding to the island. According to the adjustment direction, add or subtract the current gear from the gear adjustment step to obtain the target gear. Determine the gear adjustment range of the corresponding island based on the coil gear data. When the target gear is not within the gear adjustment range, it indicates that the adjustment amount required by the corresponding optimal solution section exceeds the actual adjustable range of the arc suppression coil. At this time, the optimal solution section is discarded. When the target gear falls within the gear adjustment range, the corresponding optimal solution section is retained.

[0099] If the pre-adjustment instruction includes the gear adjustment steps and adjustment direction corresponding to the reversing current adjustment amount and the compensation current adjustment amount, then the current gear is first combined with the gear adjustment steps corresponding to the reversing current adjustment amount to obtain the first target gear. Then, the first target gear is combined with the gear adjustment steps corresponding to the compensation current adjustment amount to obtain the second target gear. At this time, both the first target gear and the second target gear must fall within the gear adjustment range to retain the corresponding optimal solution section; otherwise, the optimal solution section is discarded.

[0100] Step S4-5-3: Select the optimal solution section with the highest self-check coverage as the solution execution scheme.

[0101] Traverse all optimal solution segments retained after feasibility testing, compare their corresponding self-check coverage rates, and select the optimal solution segment with the highest self-check coverage rate as the solution execution plan.

[0102] Through the flip-compensation dual adjustment mechanism and feasibility detection process in step S4, the optimal decoupling section is concentrated on the optimal decoupling section that does not yet meet the frequency detection conditions in the detuning jump amplitude dimension. By adjusting the corresponding arc suppression coil position, it is transformed into the optimal decoupling section that meets the frequency detection conditions. This avoids the limitation of the traditional active decoupling method, which only passively judges the island boundary after decoupling occurs and lacks active adjustment before decoupling. It forms a complete guarantee for the self-test of decoupling status from two orthogonal dimensions of direction and amplitude. At the same time, by converting the current adjustment amount into the number of adjustment steps and performing feasibility detection, the problem of the pre-adjustment command being unable to be executed due to the limited actual adjustment capability of the arc suppression coil is solved, so that the final decoupling execution scheme has both detectability and feasibility.

[0103] Step S5: Obtain the actual jump amplitude based on the decoupling execution scheme and update the transfer mapping relationship table in combination with the measurable detuning interval.

[0104] After implementing the decoupling execution scheme, the actual voltage change, actual frequency offset, and actual phase jump value detected by the node inverter at each PCC node in each island are collected at the instant of decoupling. Combined with the transfer mapping relationship corresponding to each PCC node, the three different detuning degree jump amplitude values ​​of that PCC node are inferred. A weighted average of the three different detuning degree jump amplitude values ​​is taken as the actual jump amplitude value of that PCC node. The weights of the weighted average are determined based on the signal-to-noise ratio (SNR). For example, in a specific embodiment, the SNR of the voltage detection channel of the node inverter at the PCC node is [value missing]. The signal-to-noise ratio of the frequency detection channel is The signal-to-noise ratio of the phase detection channel is Then the weight of the detuning jump amplitude corresponding to the actual voltage change can be calculated as follows: .

[0105] The actual jump amplitude is compared with the island detuning interval corresponding to the island. If the actual jump amplitude falls within the island detuning interval, it indicates that the detuning jump detection result of the PCC node in the island for this decoupling execution scheme is basically consistent with the expectation. At this time, the actual voltage change, actual frequency offset, and actual phase jump variable detected by each PCC node are incorporated into the voltage change, frequency offset, and phase jump variable collected in step S1-2. Then, the least squares fitting process in step S1-2 is repeated to correct the voltage conduction coefficient, frequency conduction coefficient, and phase conduction coefficient, and the transfer mapping relationship table of the PCC node is updated accordingly.

[0106] If the actual jump amplitude does not fall within the island detuning degree range, it indicates that the detuning degree jump detection result of the PCC node in the island for this decoupling execution scheme deviates from the expectation. At this time, an error flag is output and the update of the transmission mapping table is stopped.

[0107] In summary, this method, through the synergy of the above five steps, constructs a complete closed-loop active disconnection strategy optimization method, from detuning degree jump variable modeling, candidate disconnection section detectability assessment, multi-objective optimization, to active adjustment and execution feedback correction. It expands the selection criteria for disconnection sections from traditional dimensions such as load shedding and number of switching operations to a multi-dimensional assessment of disconnection status self-checking capability and signal detectability strength. Furthermore, through active pre-adjustment methods such as flip current adjustment and compensation current adjustment, it transforms disconnection schemes that were originally difficult to detect into executable schemes. Finally, it corrects and updates the transmission mapping relationship table using measured data, effectively avoiding the risk of secondary faults or power outages caused by misjudgments of disconnection status, and improving the accuracy, executability, and adaptability of active disconnection decisions in the distribution network.

[0108] It should be understood that, in the embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0109] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for optimizing an active disconnection strategy in a distribution network, characterized in that, include: S1: After obtaining the power grid dataset, calculate the baseline detuning degree. After obtaining the node inverter data, establish a transfer mapping table and combine it with the baseline detuning degree to obtain the measurable detuning degree range. S2: After obtaining the candidate detuning section set based on the power grid dataset, calculate the island detuning degree and combine it with the benchmark detuning degree to obtain the detuning degree jump amplitude. After obtaining the self-testable nodes based on the detuning degree jump amplitude, the transfer mapping relationship table and the measurable detuning degree interval, calculate the self-test coverage rate. S3: Based on the detuning jump amplitude and the measurable detuning interval, the signal strength score is obtained. After combining the traditional target parameters and the list of key equipment, a multi-objective optimization function is constructed, and Pareto solution is performed to obtain the optimal solution profile set. S4: Based on the optimal disengagement section set and combined with the power grid dataset, detuning degree jump amplitude and measurable detuning degree interval, calculate the reversal current regulation and compensation current regulation to obtain the pre-regulation command, and obtain the disengagement execution scheme based on the pre-regulation command and combined with the self-test coverage. S5: Based on the decoupling execution scheme, obtain the actual jump amplitude value and update the transfer mapping relationship table in combination with the measurable detuning interval.

2. The method for optimizing an active disconnection strategy in a distribution network according to claim 1, characterized in that, After acquiring the power grid dataset, the baseline detuning degree is calculated. After acquiring the node inverter data, a transfer mapping table is established and combined with the baseline detuning degree to obtain the measurable detuning degree range, including: Obtain the power grid dataset of the target distribution network, and accumulate the feeder-to-ground current data and coil compensation current data respectively to obtain the total capacitor current and the total compensation current in order to calculate the reference detuning degree; After acquiring the inverter data of each PCC node, a detuning degree jump disturbance is applied and the detuning degree change is recorded. At the same time, the voltage change, frequency offset and phase jump of each PCC node are recorded. The conduction coefficient set is then determined by least squares fitting. Based on the set of conduction coefficients, establish the mapping relationship between the detuning degree change and the voltage change, frequency offset and phase jump variable respectively, and combine them into the corresponding PCC node's transmission mapping relationship table; After establishing the effective range of voltage signal, effective range of frequency signal, and effective range of phase transition based on the inverter data at each node, the measurable detuning range of each PCC node is established by combining the transfer mapping relationship table.

3. The method for optimizing an active disconnection strategy in a distribution network according to claim 1, characterized in that, After obtaining the candidate detuning section set based on the power grid dataset, the island detuning degree is calculated, and the detuning degree jump amplitude is obtained by combining it with the benchmark detuning degree. Based on the detuning degree jump amplitude, the transfer mapping table, and the measurable detuning degree interval, the self-testable nodes are obtained, and the self-test coverage rate is calculated, including: The candidate disconnection section set is obtained based on the target distribution network topology data. The candidate disconnection section set represents the set of all possible disconnection sections of the target distribution network when a fault occurs. Each candidate disconnection section contains a set of switches to be disconnected. After identifying the islands divided by each candidate decoupling section and recording the feeders and PCC nodes they contain, the island capacitance current and island compensation current of each island are obtained by combining the feeder-to-ground current data and coil compensation current data to calculate the island detuning degree of the corresponding island. The difference between the island detuning degree and the reference detuning degree is used to obtain the detuning degree jump value corresponding to each island. Based on the detuning jump amplitude value and combined with the transfer mapping relationship table and the measurable detuning interval, detectability is determined to obtain self-testable nodes and calculate the self-test coverage.

4. The method for optimizing an active disconnection strategy in a distribution network according to claim 3, characterized in that, Based on the detuning jump amplitude value and combined with the transfer mapping relationship table and the measurable detuning interval, detectability is determined to obtain self-testable nodes and calculate the self-test coverage rate, including: Substitute the detuning jump amplitude value into the transfer mapping table to obtain the predicted voltage change, predicted frequency offset, and predicted phase jump of each PCC node in the island. If the predicted voltage change, predicted frequency offset, and predicted phase jump all fall within the corresponding effective range of the voltage signal, the effective range of the frequency signal, and the effective range of the phase jump, respectively, then the PCC node is determined to be a self-testable node. Here, self-testable means that the node inverter at the PCC node can autonomously determine whether it has entered the disconnection state. The number of self-checkable nodes and PCC nodes in each isolated island are counted to calculate the self-check coverage rate of the corresponding candidate disconnection section. The self-check coverage rate is used to characterize the ability of the candidate disconnection section to autonomously identify the disconnection status.

5. The method for optimizing an active disconnection strategy in a distribution network according to claim 1, characterized in that, After obtaining signal strength scores based on the detuning jump amplitude and the measurable detuning interval, and constructing a multi-objective optimization function by combining traditional target parameters and a list of key equipment, Pareto optimization is performed to obtain the optimal solution profile set, including: Obtain traditional target parameters and a list of key equipment. The traditional target parameters include at least the load shedding amount and the number of switching operations. The list of key equipment is a set of key equipment pre-configured according to the importance of the equipment and the protection range. The difference between the detuning jump amplitude and the boundary of the measurable detuning interval is calculated as the margin value of the corresponding PCC node. The margin values ​​of all PCC nodes in the island are weighted and summed to obtain the signal strength score of the candidate solution section. After screening candidate solution profiles based on the list of key equipment, a multi-objective optimization function is constructed by combining signal strength score, self-test coverage rate and traditional target parameters to obtain the optimal solution profile set.

6. The method for optimizing an active disconnection strategy in a distribution network according to claim 5, characterized in that, After screening candidate solution profiles based on the list of key equipment, a multi-objective optimization function is constructed by combining signal strength scores, self-test coverage, and traditional target parameters to obtain the optimal solution profile set, including: If there is a PCC node in the list of critical devices in the island and it is not marked as a self-testable node, then the candidate disconnection section corresponding to the island is removed from the candidate disconnection section set. For each candidate segment in the candidate segment set, a multi-objective optimization function is constructed with the optimization objectives of maximizing signal strength score, maximizing self-test coverage, minimizing load shedding, and minimizing the number of switching operations. After generating all non-dominated solutions of the multi-objective optimization function on the Pareto front using a multi-objective optimization algorithm, the optimal solution profile set is obtained. The non-dominated solution indicates that there are no other candidate solution profiles that are better than the candidate solution profiles corresponding to this solution in each optimization objective.

7. The method for optimizing an active disconnection strategy in a distribution network according to claim 1, characterized in that, For S4, including: Based on the target distribution network topology data, determine all adjacent island pairs for each optimal solution section. If the detuning jump amplitude values ​​between adjacent island pairs have the same sign, then merge the corresponding adjacent island pairs; otherwise, do not perform any operation. Among them, the islands with larger adjustment margins of the arc suppression coils in adjacent islands that cannot be merged are taken as islands to be flipped, and the flipping current adjustment is calculated to flip their detuning jump amplitude value. The intersection of the measurable detuning intervals corresponding to all PCC nodes in each island in the optimal solution section is taken as the island detuning interval of that island. If the detuning degree jump amplitude value corresponding to the island does not fall within the island detuning degree interval, the difference between the detuning degree jump amplitude value and the median value of the island detuning degree interval is calculated to obtain the compensation current adjustment amount; otherwise, no processing is performed. Based on the reversing current regulation and the compensation current regulation, and combined with the power grid dataset, the pre-regulation command is obtained. After the feasibility of the pre-regulation command is tested, the disconnection execution plan is obtained by combining the self-test coverage.

8. The method for optimizing the active disconnection strategy of a distribution network according to claim 7, characterized in that, Based on the flip current regulation and compensation current regulation, and combined with the power grid dataset, pre-regulation commands are obtained. After feasibility testing of the pre-regulation commands, and combined with the self-test coverage, a disconnection execution plan is obtained, including: The flip current adjustment amount and compensation current adjustment amount are converted into the gear adjustment steps by the coil gear data, and the adjustment direction is recorded and then combined into a pre-adjustment command. After obtaining the current gear, the target gear is obtained by combining the pre-adjustment command. Based on the coil gear data, it is determined whether the target gear exceeds the gear adjustment range of the arc suppression coil. The optimal solution section is discarded according to the judgment result. The optimal solution section with the highest self-inspection coverage is selected as the solution execution scheme.

9. The method for optimizing an active disconnection strategy in a distribution network according to claim 1, characterized in that, The method further includes: The power grid dataset includes feeder-to-ground current data, coil compensation current data, coil tap data, and target distribution network topology data; The feeder-to-ground current data is represented as the ground capacitance current value of each feeder in the target distribution network. The coil compensation current data is represented as the compensation current value of each setting of the arc suppression coil; The coil gear data refers to the gear step size and gear adjustment range of the arc suppression coil. The target distribution network topology data is represented by the electrical connection relationships of the target distribution network. The node inverter data refers to the minimum voltage amplitude change, minimum frequency offset, and minimum phase jump that the inverter at the corresponding PCC node can detect, as well as the maximum voltage amplitude change, maximum frequency offset, and maximum phase jump that it can withstand.

Images