A power distribution network fault recovery strategy generation method

Abstract

The application provides a power distribution network fault recovery strategy generation method, comprising: collecting reactive current component distribution of each section and transmission section quantity from compensation point to fault boundary in real time, extracting voltage sensitivity of each node, and identifying distribution position of reactive power shortage concentrated area; generating operation position optimization basis suitable for current fault recovery according to path migration performance of reactive power flow under different compensation positions; using the operation position optimization basis, evaluating voltage sensitivity change of each compensation point after migration in the direction away from the shortage area one by one, screening out the compensation point whose voltage sensitivity decreases to the safety range and the reactive power flow direction recovers to normal after migration, and generating a compensation device list that needs to be dynamically adjusted; integrating the reactive over-compensation risk node set and the compensation device list that needs to be dynamically adjusted, eliminating the reverse reactive power flow and reducing the line loss on the reactive power transmission path, and generating a fault recovery strategy scheme with minimized network loss.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for generating a power distribution network fault recovery strategy. Background Technology

[0002] In the operation and management of power systems, research on fault recovery strategies for distribution networks is of paramount importance, directly impacting the reliability and economy of power supply. As a crucial link connecting generation and consumption, the efficiency and stability of the distribution network play a decisive role in the safe operation of the entire power system. Especially after a fault occurs, how to quickly restore power and minimize losses becomes a core issue, affecting users' electricity experience and the economic benefits of the power grid. However, current fault recovery strategies often fall short in addressing the relationship between the location of reactive power compensation devices and network losses. These methods often focus only on superficial distance factors, assuming that placing compensation devices near areas with high reactive power demand will effectively reduce losses, while ignoring the impact of the more complex operating environment. This one-sided consideration often leads to deviations in practical applications, failing to adapt to dynamic changes in power grid operation and thus affecting recovery effectiveness. A deeper technical challenge lies in the fact that the selection of compensation device deployment locations is not only related to distance but also significantly constrained by node voltage sensitivity. Voltage sensitivity refers to the degree to which the voltage at a node responds to changes in reactive power. If the sensitivity is too high, even a small amount of reactive power compensation can lead to significant voltage fluctuations, or even trigger unnecessary reverse power flows. Such flows increase additional losses in the network, contradicting the initial goal of reducing losses. It is precisely because the interference of voltage sensitivity on the effectiveness of reactive power compensation has not been fully considered that the selection of compensation location can, in some cases, exacerbate the problem, creating a difficult-to-balance contradiction. For example, in a distribution network, nodes near the user load center are often areas with high reactive power demand. Conventionally, compensation devices should be deployed there first to shorten transmission distances. However, if the node has high voltage sensitivity, after compensation is implemented, the voltage rises rapidly, triggering reverse power flows. The intended losses are instead increased due to the additional power circulation. This phenomenon is particularly prominent during fault recovery because the post-fault grid state is more complex, and the mutual influence between nodes is further amplified. Therefore, how to comprehensively consider the dual impacts of reactive power demand distribution and node voltage sensitivity during fault recovery, and rationally determine the deployment location of compensation devices to avoid reverse power flows caused by overcompensation and minimize network losses, is a pressing technical problem that needs to be solved. Summary of the Invention

[0003] This invention provides a method for generating a power distribution network fault recovery strategy, which mainly includes:

[0004] Real-time acquisition of reactive current component distribution in each section and number of transmission sections from compensation point to fault boundary; at the same time, extraction of voltage sensitivity of each node to identify the distribution location of reactive power deficit concentration area. Based on the distribution of concentrated reactive power deficit areas, the path with the shortest reactive power transmission distance between the compensation point and the deficit area is extracted as the potential network loss reduction path. Nodes on the potential network loss reduction path whose voltage sensitivity exceeds the safety limit are marked as high-sensitivity nodes, and a list of high-sensitivity nodes is obtained. By comparing the reactive current component of the section where the node is located with the reactive current flow direction of the adjacent section in the list of high-sensitivity nodes, when the compensation point is close to the deficit area and the voltage sensitivity is too high, the node is identified as having the characteristic that the reactive injection exceeds the local demand and is transmitted in the opposite direction. Such nodes are marked as reactive overcompensation risk nodes, forming a set of reactive overcompensation risk nodes. Based on the path migration behavior of reactive power flow at different compensation locations, an optimization basis for the commissioning location applicable to the current fault recovery is generated; Based on the optimization of commissioning location, the voltage sensitivity change of each compensation point after it is moved away from the deficit area is evaluated one by one. Compensation points whose voltage sensitivity drops to a safe range and whose reactive power flow returns to normal after relocation are selected, and a list of compensation devices that need to be dynamically adjusted is generated. By integrating the set of reactive power overcompensation risk nodes with the list of compensation devices that need to be dynamically adjusted, reverse reactive power flow is eliminated and line losses on reactive power transmission paths are reduced, generating a fault recovery strategy scheme that minimizes network losses.

[0005] In one possible implementation, the reactive current component distribution of each section and the number of transmission sections from the compensation point to the fault boundary are collected in real time. Simultaneously, the voltage sensitivity of each node is extracted, and the distribution location of concentrated reactive power deficit areas is identified, including: The distribution network is divided into sections according to the feeder topology. The lines between adjacent switches are taken as independent sections. The voltage amplitude and phase angle of the nodes in each section are collected. The measured data of reactive current components are read from the current transformer. The voltage amplitude, phase angle and reactive current components are collected according to the section number to form a reactive component distribution record for each section. Based on the reactive component distribution record, the reactive power of each section is solved segment by segment to obtain the number of transmission sections traversed from each compensation point to the fault boundary, and the voltage change amplitude caused by injecting a unit reactive power at each node is extracted as the voltage sensitivity. For the reactive current components of each segment, when the reactive current components of multiple consecutive nodes in a segment are all higher than the average reactive current components of the entire network segment, the segment is marked as a reactive current deficit concentration area. The topological locations of all marked segments are summarized to obtain the distribution location of the reactive current deficit concentration area.

[0006] In one possible implementation, based on the distribution location of concentrated reactive power deficit areas, the path with the shortest reactive power transmission distance between the compensation point and the deficit area is extracted as a potential network loss reduction path, including: Extract the topological location of each compensation point and the topological location of each reactive power deficit concentration area from the attribute records of each segment. Count the number of segments passed between the compensation point and each deficit area and accumulate the line impedance values ​​of the segments passed through to obtain the reactive power transmission distance from each compensation point to the corresponding deficit area. For each compensation point, compare its reactive power transmission distance with each deficit area, and select the shortest path as the shortest transmission path for that compensation point. From the shortest transmission paths of all compensation points, the path with the shortest reactive power transmission distance is extracted as the potential path for reducing network loss.

[0007] In one possible implementation, nodes on the potential network loss reduction path whose voltage sensitivity exceeds the safety limit are marked as high-sensitivity nodes to obtain a list of high-sensitivity nodes. This includes: reading the voltage sensitivity value of each node on the potential network loss reduction path, comparing it with the safety limit threshold, and marking nodes whose voltage sensitivity value exceeds the safety limit threshold as high-sensitivity nodes. By summarizing the topological locations of the high-sensitivity nodes and their corresponding voltage sensitivity values, a list of high-sensitivity nodes is obtained.

[0008] In one possible implementation, by comparing the reactive current component of the segment where a node in the high-sensitivity node list is located with the reactive current flow direction of adjacent segments, when the compensation point is close to the deficit area and the voltage sensitivity is too high, the node is identified as having the characteristic of reactive power injection exceeding local demand and being transmitted in the opposite direction. Such nodes are marked as reactive power overcompensation risk nodes, forming a set of reactive power overcompensation risk nodes, including: For each node in the high-sensitivity node list, read the measured value of the reactive current component of the section where the node is located, the reactive power consumption of the local load, and the reactive current flow direction of the upstream and downstream adjacent sections. By comparing the measured value of the reactive current component in the section where the node is located with the reactive power consumption of the local load, when the measured value of the reactive current component is greater than the reactive power consumption of the local load and the reactive current flow direction shows a reverse flow from the node to the power supply side, the node is marked as a reactive power overcompensation risk node. The topological location, voltage sensitivity value, and amplitude of the reverse-transmitted reactive current component of all marked reactive overcompensation risk nodes are summarized to form a set of reactive overcompensation risk nodes.

[0009] In one possible implementation, after obtaining the list of high-sensitivity nodes, the process includes: retrieving the measured values ​​of the reactive current components of the sections where each node is located and the reactive current flow direction information of the upstream and downstream adjacent sections in the list of high-sensitivity nodes; analyzing the inflow and outflow difference of reactive current at each node and the reactive power transfer direction between it and the adjacent sections; identifying the node locations where the reactive power injection amount significantly exceeds the local load demand and there is a reverse transmission phenomenon to the power supply side; and marking the amplitude and duration of the reverse transmission of reactive power, thereby forming a set of reactive power overcompensation risk nodes.

[0010] In one possible implementation, based on the path migration performance of reactive power flow at different compensation locations, an optimization basis for the commissioning location applicable to the current fault recovery is generated, including: For each risk node in the set of reactive power overcompensation risk nodes, the compensation point is shifted segment by segment along the feeder topology in a direction away from the reactive power deficit concentration area. After each shift, the node voltage sensitivity and reactive power flow direction at the shift position are recalculated by the power flow calculation unit, and the change in voltage sensitivity before and after the shift is recorded. Based on the number of offset segments obtained from multiple offsets and the corresponding voltage sensitivity values, the relationship between the decrease in voltage sensitivity as the compensation point moves away from the deficit area is determined. The correspondence is matched with the feeder topology and reactive power deficit concentration area distribution in the current fault recovery scenario to obtain the appropriate offset range of each compensation point and the corresponding expected voltage sensitivity change, forming the basis for commissioning location optimization.

[0011] In one implementation, based on the optimization of commissioning location, the voltage sensitivity change of each compensation point after migrating away from the deficit area is evaluated one by one. Compensation points whose voltage sensitivity drops to a safe range and whose reactive power flow returns to normal after migration are selected, generating a list of compensation devices that need dynamic adjustment, including: Based on the optimization criteria for the commissioning location, the appropriate offset range of each compensation point is determined, and the target location is the position with the lowest voltage sensitivity within the specified range. Solve for the voltage sensitivity values ​​and reactive current flow direction of each node after offset; When the voltage sensitivity value drops below the safety upper limit threshold after offset and the reactive current flow direction returns to the positive direction pointing towards the load side, the compensation point is determined as the compensation point to be adjusted. Summarize the current and target positions of the compensation points to be adjusted to form a list of compensation devices that need to be dynamically adjusted.

[0012] In one possible implementation, after forming the commissioning location optimization basis, the process includes: obtaining the corresponding information of the compensation point migration distance and voltage sensitivity reduction in the commissioning location optimization basis; evaluating the voltage sensitivity change and reactive power flow improvement status of each compensation point after migrating to different distances away from the deficit area according to the optimization basis; identifying the compensation point whose voltage sensitivity can be reduced to a safe range and whose reactive power flow direction can be restored from reverse to forward flow after migration, and its optimal migration location; establishing a list of compensation devices that need to be dynamically adjusted and marking the current location and target migration location of each device.

[0013] In one possible implementation, the set of reactive power overcompensation risk nodes is integrated with a list of compensation devices that need to be dynamically adjusted to eliminate reverse reactive power flow and reduce line losses on reactive power transmission paths, generating a fault recovery strategy scheme that minimizes network losses, including: The risk node topology location in the set of risk nodes with no overcompensation is matched with the current location of the compensation device in the list of compensation devices that need to be dynamically adjusted, and the reverse transmission information of the successfully matched risk nodes is merged with the target migration location information of the corresponding compensation device. According to the order of the reactive power amplitude transmitted in reverse from large to small, the matching compensation devices are moved from the current position to the target migration position one by one. After each migration is completed, the reactive power distribution and line loss of the entire network after the migration are calculated. Summarize the target locations of the compensation devices, the direction of reactive power flow, and the line loss values ​​after all migrations are completed to form a fault recovery strategy.

[0014] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: This invention discloses a method for generating a distribution network fault recovery strategy. By real-time acquisition of reactive current component distribution in each section, the number of transmission sections from the compensation point to the fault boundary, and the voltage sensitivity of each node, the location of concentrated reactive power deficit areas is identified. Based on this, the reactive power transmission distance between the compensation point and the deficit area is analyzed, and the potential network loss reduction path with the shortest transmission distance is selected. The voltage sensitivity of each node along this path is evaluated, and high-sensitivity nodes exceeding the safety limit are marked. Furthermore, the reactive current flow direction of the section where the high-sensitivity node is located is compared with that of adjacent sections to identify nodes with reactive power injection exceeding local demand and being transmitted in the opposite direction, indicating a risk of over-compensation. Based on the reactive power flow path migration pattern under different compensation locations, the optimization basis for the decrease in voltage sensitivity after the compensation point moves away from the deficit area is extracted. The changes in voltage sensitivity and the improvement in reactive power flow direction after each compensation device moves away from the deficit area are evaluated one by one. Compensation points whose voltage sensitivity drops to a safe range and whose reactive power flow direction recovers to a positive direction are selected, generating a list of compensation devices that need dynamic adjustment and their optimal migration locations. Ultimately, the reactive power overcompensation risk nodes are integrated with the compensation devices that need adjustment. By gradually migrating high-risk compensation points to locations with lower voltage sensitivity, reverse reactive power flow is effectively eliminated, reactive power transmission distance is shortened, and line losses are reduced, achieving the optimization goal of minimizing network losses during fault recovery. Attached Figure Description

[0015] Figure 1 This is a flowchart of a method for generating a power distribution network fault recovery strategy according to the present invention.

[0016] Figure 2 This is a schematic diagram of a method for generating a power distribution network fault recovery strategy according to the present invention.

[0017] Figure 3 This is another schematic diagram of a method for generating a power distribution network fault recovery strategy according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and thoroughly described below with reference to the accompanying drawings. The described embodiments are merely some embodiments of the present invention.

[0019] like Figure 1 This embodiment provides a method for generating a power distribution network fault recovery strategy, which specifically includes: S101. Real-time acquisition of reactive current component distribution in each section and the number of transmission sections from the compensation point to the fault boundary, while extracting the voltage sensitivity of each node and identifying the distribution location of the reactive power deficit concentration area.

[0020] The distribution network is divided into sections according to the feeder topology, with the lines between each group of adjacent switches considered as independent sections. Voltage monitoring units deployed in each section collect the voltage amplitude and phase angle of each node within that section. Measured reactive current data are read from the current transformers in each section. The voltage amplitude, phase angle, and reactive current components are then aggregated according to section number to form a reactive power distribution record for each section. Based on this record, the reactive power in each section is calculated segment by segment using a power flow calculation unit. The number of transmission sections traversed from each compensation point to the deficit boundary is obtained. At each node, the voltage sensitivity value is extracted based on the voltage change caused by injecting a unit of reactive power. The number of transmission sections and the voltage sensitivity value are then written into the attribute record of each node. For the reactive current components in the attribute records of each node, if the reactive current components of three or more consecutive nodes in a certain section are all higher than 20% of the average reactive current components of all sections of the feeder, then the section is marked as a reactive current deficit concentration area. The topological location of all marked sections and the corresponding reactive current component amplitude are summarized to obtain the distribution location of the reactive current deficit concentration area.

[0021] In one implementation, the distribution network is divided into sections based on the feeder topology, defining the line between each group of adjacent switches as an independent section. Each independent section corresponds to a physically continuous line, and the boundary of the section is determined by the installation location of the sectionalizing switch or tie switch. The section number increases sequentially along the feeder from the substation outlet towards the end load.

[0022] Specifically, voltage monitoring units are deployed at key nodes within each section to collect the voltage amplitude and phase angle at those nodes. Voltage amplitude refers to the effective value of the bus voltage at that node, and phase angle refers to the angular deviation of the node voltage relative to the substation reference bus voltage. Current transformers are installed on the incoming and outgoing sides of each section to read the instantaneous value of the reactive component in the current flowing through that section. After integration over one power frequency cycle, the reactive current component of that section is obtained. The reactive current component reflects the portion of the current that is orthogonal to the voltage phase, and its magnitude directly corresponds to the reactive power consumption level within that section. The aforementioned voltage amplitude, phase angle, and reactive current component are aggregated according to their respective section numbers to form a reactive component distribution record for each section.

[0023] In one embodiment, after receiving the reactive power component distribution record, the power flow calculation unit performs reactive power flow calculation step by step along the feeder topology from the substation outlet section to the terminal section. The process of solving the power flow section by section is as follows: within each section, the reactive power output value of the outgoing side of the section is obtained by subtracting the reactive power load consumption value of each node in the section from the reactive power input value on the incoming side of the section. This output value is then used as the input value on the incoming side of the next section and is transmitted downstream in sequence. Through this process, the number of sections traversed from each compensation point along the feeder topology to the fault disconnect switch is counted, which is the number of transmission sections from the compensation point to the fault boundary.

[0024] It should be noted that the voltage sensitivity value is extracted after the power flow solution is completed. Voltage sensitivity describes the degree to which the voltage amplitude of a node changes after a unit reactive power is injected.

[0025] For example, a unit amount of reactive power disturbance is applied at a node, and the change in the node's voltage amplitude from the value before the disturbance to the value after the disturbance is observed. The difference between the two values ​​is the voltage sensitivity value of the node. The larger the voltage sensitivity value, the more drastic the response of the node's voltage to reactive power injection. The number of transmission segments and the voltage sensitivity value are both written into the attribute record of each node.

[0026] In one possible implementation, the identification of reactive power deficit concentration areas is based on the reactive current components in the attribute records of each node. The arithmetic mean of the reactive current components of all segments along the same feeder is calculated to obtain the mean value of the reactive current components for that feeder. Each node within a segment is examined one by one. If the reactive current components of multiple consecutive nodes within a segment are all higher than the mean value, then the reactive power consumption level of that segment is determined to be high, and it is marked as a reactive power deficit concentration area.

[0027] S102. Based on the distribution location of the concentrated reactive power deficit area, extract the path with the shortest reactive power transmission distance between the compensation point and the deficit area as the potential network loss reduction path, and mark the nodes on the potential network loss reduction path whose voltage sensitivity exceeds the safety limit as high sensitivity nodes, thus obtaining a list of high sensitivity nodes.

[0028] Based on the distribution of concentrated reactive power deficit areas, the topology location number of each compensation point and the topology location number of each concentrated reactive power deficit area are extracted from the attribute records of each segment. The number of segments traversed between each compensation point and each deficit area is counted along the feeder route, and the line impedance values ​​of each traversed segment are accumulated. The accumulated impedance value is used as the reactive power transmission distance from the compensation point to the corresponding deficit area, forming a transmission distance pairing record between each compensation point and each deficit area. For the transmission distance pairing record, the reactive power transmission distance of each transmission path is compared for each deficit area corresponding to each compensation point. The segment numbers and corresponding reactive current flows are marked along the path with the shortest reactive power transmission distance, obtaining the shortest transmission path for that compensation point. From the shortest transmission paths of each compensation point, the path with the shortest reactive power transmission distance is extracted as a potential network loss reduction path, and the segment numbers and corresponding reactive current flows along the potential network loss reduction path are recorded. The voltage sensitivity values ​​of each node along each segment of the potential network loss reduction path are read one by one. These values ​​are compared with a preset safety upper limit threshold, which is the upper bound of the allowable voltage response amplitude per unit reactive power injection. If the voltage sensitivity value of a node exceeds this safety upper limit threshold, the node is marked as a high-sensitivity node. A list of high-sensitivity nodes is obtained by summarizing the topology location numbers and corresponding voltage sensitivity values ​​of all marked high-sensitivity nodes along the potential network loss reduction path.

[0029] In one implementation, the reactive power transmission distance is measured based on the cumulative line impedance of each segment along the path from the compensation point to the deficit area, rather than the straight-line spatial distance between the compensation point and the deficit area. The line impedance value refers to the combined value of the inductive reactance and resistance of the conductor itself in the reactive power transmission direction within each segment, which reflects the degree of obstruction encountered by the reactive current as it flows through that segment.

[0030] Specifically, when establishing transmission distance pairing records, the process starts from the section where the compensation point is located, proceeding sequentially through adjacent sections along the feeder topology until reaching the section where the deficit area is located. For each section, the pre-stored line impedance value in the section's attribute record is read and added segment by segment. The segment-by-segment addition process is as follows: using the outgoing impedance value of the compensation point's section as the starting value, adding the total impedance value from the incoming to the outgoing side of the next section, then adding the total impedance value of the next section after that, and so on, until reaching the incoming side of the deficit area's section. The final accumulated impedance value is the reactive power transmission distance from the compensation point to the deficit area. If the same compensation point corresponds to multiple deficit areas, the above accumulation process is repeated along each of their respective paths, forming a one-to-one transmission distance pairing record between the compensation point and each deficit area. The reason for using the accumulated impedance value rather than the number of sections as the measure of reactive power transmission distance is that the conductor cross-section, length, and material differ between sections, and the actual impedance loss corresponding to the same number of sections is not the same.

[0031] In one possible implementation, all paths corresponding to the same compensation point in the transmission distance pairing record are arranged in ascending order of reactive power transmission distance, and the path at the top of the list is taken as the shortest transmission path for that compensation point. Along the feeder route of the shortest transmission path, the section number of each segment is marked, and the direction of reactive current flow within each segment is read from the measured data of the current transformers in that segment and recorded in the attribute information of the shortest transmission path.

[0032] For example, in a feeder line comprising eight sections, a compensation point is located in the second section, with its corresponding two deficit areas located in the fifth and seventh sections, respectively. After cumulative impedance calculation, the reactive power transmission distance from the compensation point to the fifth section is less than that to the seventh section. Therefore, the path passing through the third, fourth, and fifth sections is determined as the shortest transmission path for that compensation point. Further, after the shortest transmission paths for all compensation points are determined, the reactive power transmission distances of each shortest path are compared laterally, and the path with the smallest reactive power transmission distance is extracted and identified as the potential network loss reduction path. The section numbers traversed by the potential network loss reduction path and the corresponding reactive current flow direction are recorded together as complete descriptive information for that path.

[0033] Understandably, the potential network loss reduction path indicates the channel with the minimum impedance accumulation during the transmission of reactive power from the compensation point to the deficit area. When reactive power compensation is performed along this channel, the reactive power loss on the line is at its lowest level among all candidate paths. Following the feeder route of the potential network loss reduction path, the voltage sensitivity value of each node in each segment is sequentially read and compared one by one with a preset safety upper limit threshold. The safety upper limit threshold represents the allowable upper bound of the node voltage response amplitude to unit reactive power injection.

[0034] Specifically, if the voltage amplitude change at a node exceeds the allowable voltage deviation range for normal operation of the distribution network after a unit of reactive power is injected, the node is considered to have excessive voltage sensitivity. The safety upper limit threshold is the sensitivity critical value corresponding to this allowable voltage deviation range, which is calculated from the allowable voltage fluctuation range of the nodes specified in the distribution network dispatching and operation procedures.

[0035] For example, five nodes are sequentially distributed along a potential network loss reduction path of a feeder. The voltage sensitivity value of each node is read and compared with a safety upper limit threshold. The voltage sensitivity values ​​of the second and fourth nodes exceed this threshold, and these two nodes are marked as high-sensitivity nodes. The topology location numbers and corresponding voltage sensitivity values ​​of all marked high-sensitivity nodes along the potential network loss reduction path are compiled to form a list of high-sensitivity nodes.

[0036] S103. Compare the reactive current component of the section where the node is located in the high-sensitivity node list with the reactive current flow direction of the adjacent section. When the compensation point is close to the deficit area and the voltage sensitivity is too high, identify the node as having the characteristic that the reactive injection exceeds the local demand and is transmitted in the opposite direction. Mark such nodes as reactive overcompensation risk nodes and form a set of reactive overcompensation risk nodes.

[0037] For each node in the high-sensitivity node list, the measured value of the reactive current component is read from the current transformer in the section where the node is located. Simultaneously, the reactive power consumption of the local load in that section and the reactive current flow direction of the upstream and downstream adjacent sections are read. The measured value of the reactive current component in the section where the node is located is compared with the reactive current flow direction of the upstream and downstream adjacent sections to determine whether there is a reverse transmission characteristic of reactive current flowing from downstream to upstream at that node. The measured value of the reactive current component, the reactive power consumption of the local load, and the reverse transmission judgment result of each high-sensitivity node are recorded in the reactive current flow direction comparison record. According to the reactive current flow direction comparison record, if the measured value of the reactive current component in the section where a high-sensitivity node is located is greater than the reactive power consumption of the local load in that section, and the reactive current flow direction shows a reverse flow from that node to the power source side, then that node is marked as a reactive power overcompensation risk node. The topology location number, corresponding voltage sensitivity value, and amplitude of the reverse-transmitted reactive current component of all marked reactive overcompensation risk nodes are summarized to form a set of reactive overcompensation risk nodes.

[0038] In one implementation, each node recorded in the high-sensitivity node list is located on a potential network loss reduction path, and its voltage sensitivity value has exceeded the safety upper limit threshold. For each node in the list, the measured value of the reactive current component is read from the current transformer of the section where the node is located, and the reactive power consumption of the local load is read from the metering device of each load connection point in that section.

[0039] It should be noted that the reactive power consumption of local load refers to the total reactive power actually absorbed by all electrical loads in the section under the current operating conditions. The difference between this value and the measured value of the reactive current component reflects whether there is a reactive power surplus in the section after receiving reactive power compensation.

[0040] Specifically, the comparison process for reactive current flow direction is as follows: The reactive current flow direction on the outgoing side of the upstream adjacent section and the incoming side of the downstream adjacent section are read. Under normal operating conditions, reactive current flows unidirectionally from the power source side to the load side along the feeder; that is, the outflow direction of reactive current in the upstream section is consistent with the inflow direction of reactive current in the downstream section, both pointing towards the end of the feeder. If, at a certain high-sensitivity node, the reactive current flow direction of the downstream adjacent section reverses, exhibiting a characteristic of flowing back from that node towards the power source side, it indicates that the reactive power injected at that node has exceeded the absorption capacity of the local load, and the excess is transmitted upstream along the feeder. The measured values ​​of the above reactive current components, the reactive power consumption of the local load, and the reverse transmission judgment results are recorded one by one in the reactive current flow direction comparison record.

[0041] In one possible implementation, each high-sensitivity node is screened one by one based on the reactive current flow direction comparison record. If the measured value of the reactive current component in the section where a node is located is greater than the reactive power consumption of the local load, and the reactive current flow direction comparison result at that node shows a reverse flow towards the power source, then both conditions are met, and the node is marked as a reactive power overcompensation risk node.

[0042] For example, on a potential network loss reduction path of a feeder, the list of high-sensitivity nodes includes three nodes. After comparison, one node's measured reactive current component exceeds the local load's reactive power consumption, and reverse reactive current occurs in the downstream section. This node is marked as a reactive power overcompensation risk node. The other two nodes, although having high voltage sensitivity, do not have reversed reactive current flow and are not marked. The topology location number, corresponding voltage sensitivity value, and amplitude of the reverse-transmitted reactive current component of all marked reactive power overcompensation risk nodes are summarized and collected to form a reactive power overcompensation risk node set. The set fully records the location of each risk node in the feeder topology and the degree of reactive power overcompensation.

[0043] Retrieve the measured values ​​of reactive current components in the sections where each node is located from the high-sensitivity node list and the reactive current flow information of the upstream and downstream adjacent sections. Analyze the inflow and outflow differences of reactive current at each node and the direction of reactive power transfer with adjacent sections. Identify the locations of nodes where the reactive power injection amount significantly exceeds the local load demand and there is a reverse transmission phenomenon to the power supply side. Mark the amplitude and duration of the reverse transmission of reactive power to form a set of reactive power overcompensation risk nodes.

[0044] Retrieve the topology location number of each node segment from the high-sensitivity node list. Read the reactive current value received on the incoming side of that segment as the inflow, and read the reactive current value output on the outgoing side as the outflow. Subtract the outflow from the inflow to obtain the difference. A positive difference indicates that the reactive power absorbed by the segment is greater than the reactive power output; a negative difference indicates that the reactive power output is greater than the reactive power absorbed. Record the difference for each node and the corresponding reactive power transmission direction of the upstream and downstream adjacent segments in the reactive current inflow / outflow difference record. This record integrates reactive current flow direction comparison information and serves as the sole criterion for identifying overcompensated nodes. If the difference for a node is negative and the reactive power transmission direction of the downstream adjacent segment points towards the power source, it is determined that the reactive power injection at that node exceeds the local load demand and there is a reverse transmission to the power source. This node is then marked as a reactive power overcompensation risk node. For reactive power overcompensation risk nodes, the start and end times of the reverse transmission phenomenon are extracted from the continuous sampling data of the current transformer. The interval between the start and end times is taken as the duration, and the amplitude of the reverse transmitted reactive power within this duration is read. The maximum value is taken as the peak power, and the duration and peak power are marked in the attribute record of the risk node. The topology location number, voltage sensitivity value, peak reverse transmission power, and duration of all reactive power overcompensation risk nodes are summarized to form a set of reactive power overcompensation risk nodes.

[0045] In one implementation, each node in the high-sensitivity node list is located on a potential network loss reduction path, and its voltage sensitivity value has exceeded the safety upper limit threshold. For each node in the list, the reactive current value received on the incoming side and the reactive current value output on the outgoing side of the segment where the node is located are read. The incoming side refers to the end of the segment closer to the power source, and the outgoing side refers to the end of the segment closer to the load. The reactive current values ​​at both ends are collected in real time by current transformers installed at the corresponding locations.

[0046] Specifically, the inflow and outflow are defined as follows: Inflow is the effective value of reactive current received by the incoming side of the section from the upstream section at the current sampling time; outflow is the effective value of reactive current output by the outgoing side of the section to the downstream section at the same sampling time. The difference between the two reflects the net absorption or net release of reactive power within the section. The difference is obtained by subtracting the outflow from the inflow. When the difference is positive, it indicates that the reactive current flowing into the section is greater than the reactive current flowing out, meaning that the local load within the section has absorbed some reactive power; when the difference is negative, it indicates that the reactive current flowing out of the section is greater than the reactive current flowing in, meaning that the section has output excess reactive power after receiving reactive power compensation. The difference values ​​at each node, along with the reactive power transmission direction of adjacent upstream and downstream sections, are recorded in the reactive current inflow / outflow difference record.

[0047] It should be noted that the direction of reactive power transmission refers to the direction in which reactive current flows between adjacent sections. Under normal operating conditions of the distribution network, reactive current flows along the feeder from the substation side to the load end, and the transmission direction points towards the load side. When the reactive power compensation device in a certain section invests too much reactive power, the excess reactive current will flow back along the feeder towards the substation, and the transmission direction will reverse, pointing towards the power source side.

[0048] In one possible implementation, each high-sensitivity node is individually assessed based on the reactive current inflow-outflow difference record. The assessment criteria include two conditions: the inflow-outflow difference of the node is negative, and the reactive power transfer direction of the downstream adjacent section points towards the power source. If both conditions are met simultaneously, it indicates that the section containing the node not only has a reactive power surplus, but the surplus reactive power has also formed an actual reverse flow along the feeder towards the power source. In this case, the node is marked as a reactive power overcompensation risk node. If only one condition is met, it is not marked.

[0049] For example, there are four highly sensitive nodes on a potential network loss reduction path of a feeder. After reading them one by one, the inflow-outflow difference of the first node is positive, indicating that the section is still absorbing reactive power, and it is not marked. The difference of the second node is negative, but the reactive power transmission direction of its downstream adjacent section is still towards the load side, indicating that the excess reactive power is absorbed by the downstream load and no reverse flow is formed, so it is also not marked. The difference of the third node is negative and the transmission direction of its downstream adjacent section is towards the power source side, satisfying both conditions, and it is marked as a reactive power overcompensation risk node. The determination process for the fourth node is similar to that of the third node; if both conditions are met, it is marked; otherwise, it is not marked. Further, for the marked reactive power overcompensation risk nodes, the start and end times of the reverse transmission phenomenon are extracted from the continuous sampling data of the current transformer in the section where the node is located. The start time is the time point when the inflow-outflow difference of the node changes from positive to negative, and the end time is the time point when the difference recovers from negative to positive. The interval between the two time points is the duration of the reverse transmission. During the duration, the reactive power amplitude transmitted in reverse at each sampling moment is read, and the maximum value is taken as the peak power transmitted in reverse at the risk node. The length of the duration and the magnitude of the peak power together reflect the severity of reactive power overcompensation at the node. The longer the duration, the more stable the reverse transmission phenomenon is in the time dimension; the larger the peak power, the more significant the deviation of the reverse transmission in the amplitude dimension. The duration and reactive power amplitude are marked in the attribute record of the risk node.

[0050] Preferably, after the attribute records of all reactive power overcompensation risk nodes are labeled, the topology location number, voltage sensitivity value, reactive power amplitude transmitted in reverse, and duration of each risk node are summarized and collected to form a set of reactive power overcompensation risk nodes. Each record in the set corresponds to a complete description of a risk node, including the node's location in the feeder topology, its voltage response to reactive power injection, the magnitude of the reverse power transmission, and the duration of the reverse transmission.

[0051] In one embodiment, the set of reactive power overcompensation risk nodes is stored in the form of a list. Each record in the list is arranged from largest to smallest according to the magnitude of the reactive power transmitted in reverse, which facilitates prioritizing the handling of nodes with more severe reverse transmission during the subsequent adjustment of the position of the compensation device.

[0052] S104. Based on the path migration performance of reactive power flow at different compensation locations, generate optimization criteria for the commissioning location applicable to the current fault recovery.

[0053] For each risk node in the reactive power overcompensation risk node set, the compensation point is gradually shifted away from its current position along the feeder topology, moving away from the deficit area by one segment. After each shift, the reactive power distribution of each node along the path of the compensation point is recalculated using the power flow calculation unit. The changes in the reactive power flow direction of each node before and after the shift are recorded, resulting in a reactive power flow path migration record for each compensation point at different shift positions. Based on the reactive power flow path migration record, the voltage sensitivity value of the node at the new position of the compensation point after each shift is extracted. Each shift position and its corresponding voltage sensitivity value are arranged in ascending order of the number of shift segments. The trend of the voltage sensitivity value gradually decreasing as the compensation point moves away from the deficit area is identified, resulting in a table of correspondence between the migration distance and the sensitivity reduction of each compensation point. From the table, the shift position intervals where the voltage sensitivity value drops below the safe upper limit threshold are selected, and the correspondence between the shift distance and the corresponding sensitivity reduction within each shift position interval is determined as the migration law. The offset distance and sensitivity reduction of each compensation point in the migration pattern are matched with the feeder topology and the distribution location of the reactive power deficit concentration area under the current fault recovery scenario. The matching rule is to calculate the similarity. , where d is the standardized difference between the offset distance and the distribution of the missing area. If S is greater than 0.8, it is considered a successful match. The appropriate offset range of each compensation point in the current feeder topology and the corresponding expected value of sensitivity reduction are marked to form the basis for optimizing the commissioning location suitable for the current fault recovery.

[0054] In one implementation, the set of reactive power overcompensation risk nodes records the location of the compensation point corresponding to each risk node. For the compensation point corresponding to each risk node, it is offset along the feeder topology from its current segment to the next adjacent segment away from the concentrated area of ​​reactive power deficit, i.e., moved one segment away from the power source side. After the offset is completed, the position number of the compensation point in the feeder topology changes, and its reactive power injection point changes accordingly.

[0055] Specifically, after each offset, the power flow calculation unit uses the offset compensation point location as the new reactive power injection point and recalculates the reactive power distribution of each segment along the path of that compensation point. The recalculation process is the same as the aforementioned segment-by-segment power flow calculation, starting from the substation outlet segment and calculating the reactive power inflow and outflow segment by segment along the feeder topology until the fault boundary is reached. After the solution is completed, the differences between the reactive power flow direction of each node before and after the offset are recorded, including whether the flow direction has reversed and the increase or decrease in reactive power amplitude. The above information is recorded one by one according to the number of offsets, forming a reactive power flow path migration record of the compensation point at different offset positions. The step-by-step offset process is repeated multiple times, each time offsetting one segment further away from the deficit area, until the compensation point reaches the offsettable boundary position on the feeder.

[0056] It should be noted that the offset boundary position refers to the section position closest to the power supply side that the compensation point can reach in the feeder topology. Beyond this position, the compensation point will be out of the coverage area of ​​the potential network loss reduction path.

[0057] In one possible implementation, the voltage sensitivity value of the new location node after each offset of the compensation point is extracted based on the reactive power flow path migration record. The extraction method of the voltage sensitivity value is the same as described above, that is, the change in voltage amplitude at the new location node is observed after applying a unit reactive power disturbance. The number of segments offset each time is taken as the migration distance, and the decrease in voltage sensitivity value before and after the offset is taken as the sensitivity reduction. These are arranged in ascending order of migration distance to form a correspondence table between migration distance and sensitivity reduction. In the correspondence table, each row records the migration distance value and sensitivity reduction value corresponding to one offset, and the number of rows equals the total number of offsets. By observing the changing trend of sensitivity reduction values ​​in each row, it is identified whether the voltage sensitivity value of the compensation point shows a continuous decreasing characteristic as it moves away from the deficit area, and this is used as the basis for optimizing the commissioning location.

[0058] For example, in a certain feeder, a compensation point is initially located in the sixth segment and gradually shifts towards the power supply side to the fifth, fourth, and third segments. After shifting to the fifth segment, the voltage sensitivity of the node at that location decreases compared to the initial location; after shifting to the fourth segment, the sensitivity decreases further; and after shifting to the third segment, the sensitivity drops below the safe upper limit threshold. A correspondence table records the sensitivity reduction for shift distances of one segment, two segments, and three segments, respectively. Further, the shift position intervals where the voltage sensitivity drops below the safe upper limit threshold are selected from the correspondence table. Within this interval, the voltage sensitivity of the node where the compensation point is located no longer exceeds the safe range, and the voltage fluctuation caused by reactive power compensation is at a controllable level. The correspondence between each shift distance and the corresponding sensitivity reduction within the shift position interval is determined as the migration law, which describes the expected reduction in voltage sensitivity after the compensation point shifts by a certain number of segments under the current feeder topology.

[0059] It is understandable that different feeders have different line parameters and load distributions, and the sensitivity reduction caused by the same offset distance is not the same on different feeders. The migration pattern reflects the characteristics of a specific feeder under the current fault recovery scenario.

[0060] In one embodiment, the offset distance and sensitivity reduction of each compensation point in the migration pattern are matched with the feeder topology and the distribution location of the reactive power deficit concentration area under the current fault recovery scenario. The matching process is as follows: for each compensation point, based on the minimum offset distance in the migration pattern that reduces sensitivity to below the safety upper limit threshold, the starting and ending segments of the appropriate offset for that compensation point are marked in the feeder topology. At the same time, the expected sensitivity reduction value corresponding to each position within the offset range is marked, forming a basis for optimizing the commissioning location suitable for the current fault recovery scenario. The basis for optimizing the commissioning location includes the current position number of each compensation point, the range of the appropriate offset segment, and the expected sensitivity reduction value corresponding to each offset position.

[0061] S105. Based on the optimization of commissioning location, evaluate the change in voltage sensitivity of each compensation point after it is moved away from the deficit area, screen out the compensation points whose voltage sensitivity drops to a safe range and whose reactive power flow returns to normal after the migration, and generate a list of compensation devices that need to be dynamically adjusted.

[0062] Based on the optimization of commissioning locations, the current location number, suitable offset range, and expected sensitivity reduction value corresponding to each offset position of each compensation point are read one by one. Each compensation point is offset from its current position away from the deficit area to the position with the lowest sensitivity value corresponding to the expected sensitivity reduction value within the specified range, which is then taken as the target position. The voltage sensitivity value and reactive current flow direction of each node after offset are recalculated using a power flow calculation unit, resulting in a sensitivity and flow direction evaluation record for each compensation point after offset. According to the sensitivity and flow direction evaluation record, if the voltage sensitivity value of the node where a compensation point is offset to the target position drops below the safe upper limit threshold, and the reactive current flow direction of the downstream adjacent section of that node recovers from reverse flow to forward flow towards the load side, then that compensation point is determined as a compensation point to be adjusted. The current location numbers and target migration location numbers of all compensation points to be adjusted are summarized to form a list of compensation devices that require dynamic adjustment.

[0063] In one implementation, the deployment location optimization criteria record the current location number of each compensation point, the suitable offset range, and the expected sensitivity reduction value corresponding to each offset position. For each compensation point, the offset position with the lowest sensitivity reduction value is selected from the expected sensitivity reduction values ​​corresponding to each offset position within the range as the target location for that compensation point.

[0064] Specifically, the selection process for the target location is as follows: Read the expected sensitivity reduction values ​​corresponding to each offset position within the suitable offset range of a compensation point. Subtract the expected sensitivity reduction values ​​for each offset position from the voltage sensitivity value at the current position to obtain the expected sensitivity values ​​for each offset position. Take the offset position with the smallest expected sensitivity value as the target location for that compensation point. After offsetting the compensation point from the current position to the target position, the power flow calculation unit uses the target position as the new reactive power injection point and recalculates the voltage sensitivity values ​​and reactive current flow direction of each node along the path, obtaining the sensitivity and flow direction evaluation records for each compensation point after offset.

[0065] It should be noted that the voltage sensitivity value obtained by resolving is the actual value after offset, and deviations are allowed between it and the expected value in the basis of the commissioning location optimization. The final selection shall be based on the actual solution result.

[0066] In one possible implementation, each compensation point is screened individually based on sensitivity and flow direction evaluation records. The screening criteria include two conditions: the voltage sensitivity value of the node where the compensation point is located after shifting to the target position drops below the safe upper limit threshold; and the reactive current flow direction of the downstream adjacent section of the node recovers from reverse flow to forward flow towards the load side. Forward flow refers to the normal state of reactive current transmission along the feeder from the power source side to the load end. If a compensation point meets both of the above conditions, it is identified as a compensation point to be adjusted; if only one condition is met or neither condition is met, the compensation point is not included in the adjustment range.

[0067] For example, in a certain feeder, three compensation points are evaluated. After offsetting and resolving, the sensitivity of the first compensation point drops below the safety threshold and the flow direction returns to positive. The sensitivity of the second compensation point decreases but remains above the safety threshold. The sensitivity of the third compensation point drops below the threshold, but the flow direction does not return to positive. Only the first compensation point is identified as the compensation point to be adjusted. Further, the current location numbers and target migration location numbers of all compensation points to be adjusted are summarized and compiled to form a list of compensation devices that need dynamic adjustment. Each record in the list corresponds to a compensation point to be adjusted, indicating the current installation location of the compensation device in the feeder topology and the target location to which it should be migrated.

[0068] Obtain the corresponding information of the compensation point migration distance and voltage sensitivity reduction in the commissioning location optimization basis, evaluate the voltage sensitivity change and reactive power flow improvement status after each compensation point is migrated to a different distance away from the deficit area according to the optimization basis, identify the compensation point whose voltage sensitivity can be reduced to a safe range and the reactive power flow can be restored from reverse to forward flow after migration, and its optimal migration position, establish a list of compensation devices that need to be dynamically adjusted and mark the current position and target migration position of each device.

[0069] like Figure 2The current location number and suitable offset range of each compensation point are obtained from the commissioning location optimization basis. For each compensation point, it is offset segment by segment along the feeder topology away from the deficit area. At each offset position, the voltage sensitivity value of the node at that position and the reactive current flow direction of the downstream adjacent segment are recalculated to obtain the measured sensitivity value and reactive current flow direction of each compensation point at different migration distances. Based on the measured sensitivity value and reactive current flow direction, the measured voltage sensitivity value of each offset position is compared to see if it drops below the safe upper limit threshold. At the same time, it is determined whether the reactive current flow direction at the offset position has recovered from reverse to forward flow towards the load side. If an offset position meets both conditions of sensitivity dropping to the safe range and flow direction recovering to the forward direction, then the offset position is marked as a candidate migration position for that compensation point. All candidate migration positions for each compensation point are summarized to obtain a set of candidate migration positions. For multiple candidate migration positions corresponding to the same compensation point in the candidate migration position set, the measured voltage sensitivity values ​​of each candidate position are compared, and the candidate position with the lowest measured sensitivity value is selected as the optimal migration position for that compensation point. Summarize all compensation points with optimal migration positions, and mark the current position number and optimal migration position number of each compensation point to establish a list of compensation devices that need to be dynamically adjusted.

[0070] In one implementation, the optimization criteria for the commissioning location have recorded the current location number of each compensation point, the suitable offset range, and the expected sensitivity reduction value corresponding to each offset location. For each compensation point, starting from its current location, it is offset segment by segment along the feeder topology in a direction away from the concentrated area of ​​reactive power deficit, i.e., towards the power supply side.

[0071] Specifically, the step-by-step offset process is as follows: After offsetting the compensation point by one segment from its current position, the offset position is used as the new reactive power injection point. The power flow calculation unit, based on the feeder topology, load data, and compensation distribution as input, uses step-by-segment power flow iterative calculation to solve for the reactive power distribution of each node along the path. The output is the voltage sensitivity S (the ratio of voltage to reactive power change) of the node at the offset position and the reactive current flow direction I (forward or reverse) of the downstream adjacent segment. After completing one offset and solution, the compensation point is offset by another segment, and the above solution process is repeated until the offset distance reaches the upper limit of the suitable offset segment range. The voltage sensitivity value obtained after each offset is the measured sensitivity value at that position, distinguished from the expected sensitivity reduction value in the optimization basis of the commissioning position. The measured sensitivity value and the corresponding reactive current flow direction at each offset position are recorded in the measured sensitivity value and reactive current flow direction measurement status record. The measured sensitivity value is obtained by solving the power flow calculation unit under the actual feeder operating conditions after the offset, while the expected sensitivity reduction value is an estimate calculated based on historical offset data according to the aforementioned migration pattern. A numerical discrepancy is allowed between the two, and the final selection is based on the measured sensitivity value.

[0072] In one possible implementation, such as Figure 3 As shown, based on the measured sensitivity values ​​and the measured reactive power flow records, each offset location is screened under two conditions. The first condition is whether the measured voltage sensitivity value at the offset location has dropped below the safe upper limit threshold. The safe upper limit threshold is the same as the threshold used in the aforementioned high-sensitivity node screening, i.e., the allowable upper bound of the node voltage's response amplitude to unit reactive power injection. The second condition is whether the reactive current flow direction in the downstream adjacent section at the offset location has recovered from reverse to forward flow towards the load side. Forward flow refers to the transmission of reactive current along the feeder from the power source side to the load end, which is the reactive current flow direction under normal operating conditions of the distribution network. If an offset location simultaneously meets both conditions—sensitivity dropping to the safe range and flow direction recovering to the forward direction—then the offset location is marked as a candidate relocation location for the compensation point.

[0073] For example, a compensation point has four offset positions within a suitable offset range. After solving each position, the measured sensitivity value of the first offset position is still higher than the safety upper limit threshold, thus not satisfying the first condition; the measured sensitivity value of the second offset position drops below the threshold but the flow direction is still reversed, thus not satisfying the second condition; the third and fourth offset positions satisfy both conditions and are marked as candidate migration positions. After summing all the candidate migration positions for compensation points, a set of candidate migration positions is obtained. Further, for multiple candidate migration positions corresponding to the same compensation point in the candidate migration position set, the measured voltage sensitivity values ​​of each candidate position are compared one by one. In the above example, the third and fourth offset positions are both candidates. If the measured sensitivity value of the fourth offset position is lower than that of the third, then the fourth offset position is selected as the optimal migration position for that compensation point. If a compensation point has only one candidate migration position, then that position is directly selected as the optimal migration position.

[0074] Understandably, the optimal migration position is the offset position where the measured voltage sensitivity is at its lowest level, under the premise of satisfying both sensitivity safety and normal flow direction. The reactive power compensation device at this position is furthest from the risk of overcompensation.

[0075] Preferably, after the optimal migration positions of all compensation points are determined, the current position number and the optimal migration position number of each compensation point are marked one by one and summarized to establish a list of compensation devices that need to be dynamically adjusted. Each record in the list contains the current installation position number of a compensation device and the target position number to which it should be migrated, indicating where the device should be migrated from and to in the feeder topology.

[0076] S106. Integrate the set of reactive power overcompensation risk nodes with the list of compensation devices that need to be dynamically adjusted, eliminate reverse reactive power flow and reduce line losses on reactive power transmission paths, and generate a fault recovery strategy scheme that minimizes network losses.

[0077] The topology location number, voltage sensitivity value, reverse reactive power amplitude, and duration of each risk node are read from the set of reactive power overcompensation risk nodes. The current location number and target migration location number of each compensation device are read from the list of compensation devices requiring dynamic adjustment. These are matched one by one according to the topology location number. If the topology location number of a risk node matches the current location number of a compensation device, the reverse power transmission information of that risk node and the target migration location information of that compensation device are merged into a single record, resulting in an integrated adjustment record for risk nodes and compensation devices. Based on this integrated adjustment record, each risk node is arranged from largest to smallest reactive power amplitude. Each compensation device is migrated from its current location to its corresponding target migration location in descending order. After each compensation device migration is completed, the reactive power flow distribution and line losses of each segment along the entire path are recalculated using the power flow calculation unit. If the reactive current flow direction of each segment returns to a positive flow towards the load side after migration, and the line losses are lower than before migration, the migration result is recorded, resulting in a record of the overall network reactive power flow distribution and line losses after the gradual migration. The reactive power flow distribution of the entire network after gradual migration is summarized along with the target migration location of each compensation device in the line loss record, the reactive power flow direction of each section after migration, and the corresponding line loss value. The network loss Ploss (Ploss is the line loss power) of different migration combinations is calculated iteratively, the Ploss values ​​of each combination are compared, and the summary result with the smallest Ploss is selected to form a fault recovery scheme with minimal network loss.

[0078] In one implementation, the reactive power over-compensation risk node set records the topology location number, voltage sensitivity value, reverse reactive power amplitude, and duration of each risk node. The list of compensation devices requiring dynamic adjustment records the current location number and target migration location number of each compensation device. The integration and matching process is as follows: read the topology location number of each risk node in the risk node set one by one, search for the compensation device record in the compensation device list whose current location number matches the topology location number, merge the information of the two into one record, and obtain the integration and adjustment record of risk nodes and compensation devices.

[0079] Specifically, each record in the integration and adjustment record includes the magnitude of the reverse reactive power transmitted to that node, the duration, and the target location number to which the corresponding compensation device should be relocated. The integration and adjustment records are arranged from largest to smallest according to the magnitude of the reverse reactive power transmitted. The larger the magnitude, the more severe the reactive power overcompensation at that node, and these records are processed first during the relocation process.

[0080] It should be noted that the migration process is carried out sequentially according to the arrangement order, rather than migrating all compensation devices simultaneously. After each compensation device completes its relocation from its current location to the target relocation location, the power flow calculation unit, using the changed distribution of compensation devices across the entire network as a condition, recalculates the reactive power flow distribution and line losses for each segment along the entire path. The recalculation process is the same as the aforementioned segment-by-segment power flow calculation, starting from the substation outlet segment and calculating the inflow and outflow of reactive power segment by segment along the feeder. If, after the migration, the reactive current flow direction in each segment returns to a positive flow towards the load side, and the line losses are lower than before the migration, then the migration result is recorded, and the process continues with the next compensation device in the sequence.

[0081] In one possible implementation, if, after the relocation of a compensation device, some sections still experience unrestored reactive current flow or unreduced line losses, the relocation result is retained, and the abnormal section location is marked in the record. The next compensation device in the sequence is then processed. After all compensation devices have been relocated, the reactive power flow distribution and line loss records for the entire network after the gradual relocation are obtained. Further, the target relocation locations of each compensation device, the reactive power flow direction of each section after relocation, and the corresponding line loss values ​​in the gradually relocated reactive power flow distribution and line loss records are summarized and collected. The summarized record includes information on the location changes of all compensation devices before and after relocation, as well as the reactive power flow direction and line loss level of each section of the entire network after relocation.

[0082] Preferably, the summary records are arranged by feeder, and the migration information of compensation devices on the same feeder, the reactive power flow direction of each section of the feeder, and the line loss are stored together to form a fault recovery scheme with minimal network loss.

[0083] If the technical solution of this application involves personal information, the product using this solution has clearly informed the user of the personal information processing rules and obtained the user's voluntary consent before processing the personal information. If sensitive personal information is involved, the user's separate consent has been obtained before processing, and the "express consent" requirement is met. For example, a clear sign is placed at the collection device such as a camera to inform the user that they have entered the collection area, and the user's voluntary entry is considered as consent; or the processing device clearly indicates the processing rules and obtains authorization through pop-up windows or by asking the user to upload information themselves. The personal information processing rules include the processor, the purpose of processing, the processing method, and the types of personal information.

[0084] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for generating a power distribution network fault recovery strategy, characterized in that, The method includes: Real-time acquisition of reactive current component distribution in each section and number of transmission sections from compensation point to fault boundary; at the same time, extraction of voltage sensitivity of each node to identify the distribution location of reactive power deficit concentration area. Based on the distribution of concentrated reactive power deficit areas, the path with the shortest reactive power transmission distance between the compensation point and the deficit area is extracted as the potential network loss reduction path. Nodes on the potential network loss reduction path whose voltage sensitivity exceeds the safety limit are marked as high-sensitivity nodes, and a list of high-sensitivity nodes is obtained. By comparing the reactive current component of the section where the node is located with the reactive current flow direction of the adjacent section in the list of high-sensitivity nodes, when the compensation point is close to the deficit area and the voltage sensitivity is too high, the node is identified as having the characteristic that the reactive injection exceeds the local demand and is transmitted in the opposite direction. Such nodes are marked as reactive overcompensation risk nodes, forming a set of reactive overcompensation risk nodes. Based on the path migration behavior of reactive power flow at different compensation locations, an optimization basis for the commissioning location applicable to the current fault recovery is generated; Based on the optimization of commissioning location, the voltage sensitivity change of each compensation point after it is moved away from the deficit area is evaluated one by one. Compensation points whose voltage sensitivity drops to a safe range and whose reactive power flow returns to normal after relocation are selected, and a list of compensation devices that need to be dynamically adjusted is generated. By integrating the set of reactive power overcompensation risk nodes with the list of compensation devices that need to be dynamically adjusted, reverse reactive power flow is eliminated and line losses on reactive power transmission paths are reduced, generating a fault recovery strategy scheme that minimizes network losses.

2. The method for generating a power distribution network fault recovery strategy according to claim 1, characterized in that, The real-time acquisition of reactive current component distribution in each section and the number of transmission sections from the compensation point to the fault boundary, along with the extraction of voltage sensitivity at each node and identification of the distribution location of concentrated reactive power deficit areas, includes: The distribution network is divided into sections according to the feeder topology. The lines between adjacent switches are taken as independent sections. The voltage amplitude and phase angle of the nodes in each section are collected. The measured data of reactive current components are read from the current transformer. The voltage amplitude, phase angle and reactive current components are collected according to the section number to form a reactive component distribution record for each section. Based on the reactive component distribution record, the reactive power of each section is solved segment by segment to obtain the number of transmission sections traversed from each compensation point to the fault boundary, and the voltage change amplitude caused by injecting a unit reactive power at each node is extracted as the voltage sensitivity. For the reactive current components of each segment, when the reactive current components of multiple consecutive nodes in a segment are all higher than the average reactive current components of the entire network segment, the segment is marked as a reactive current deficit concentration area. The topological locations of all marked segments are summarized to obtain the distribution location of the reactive current deficit concentration area.

3. The method for generating a power distribution network fault recovery strategy according to claim 1, characterized in that, The step of extracting the path with the shortest reactive power transmission distance between the compensation point and the deficit area as a potential network loss reduction path based on the distribution location of the concentrated reactive power deficit area includes: Extract the topological location of each compensation point and the topological location of each reactive power deficit concentration area from the attribute records of each segment. Count the number of segments passed between the compensation point and each deficit area and accumulate the line impedance values ​​of the segments passed through to obtain the reactive power transmission distance from each compensation point to the corresponding deficit area. For each compensation point, compare its reactive power transmission distance with each deficit area, and select the shortest path as the shortest transmission path for that compensation point. From the shortest transmission paths of all compensation points, the path with the shortest reactive power transmission distance is extracted as the potential path for reducing network loss.

4. The method for generating a power distribution network fault recovery strategy according to claim 1, characterized in that, The step of marking nodes whose voltage sensitivity exceeds the safety limit on the potential network loss reduction path as high-sensitivity nodes and obtaining a list of high-sensitivity nodes includes: reading the voltage sensitivity value of each node on the potential network loss reduction path, comparing it with the safety limit threshold, and marking nodes whose voltage sensitivity value exceeds the safety limit threshold as high-sensitivity nodes. By summarizing the topological locations of the high-sensitivity nodes and their corresponding voltage sensitivity values, a list of high-sensitivity nodes is obtained.

5. The method for generating a power distribution network fault recovery strategy according to claim 1, characterized in that, The reactive current component of the segment where the node in the high-sensitivity node list is located is compared with the reactive current flow direction of the adjacent segment. When the compensation point is close to the deficit area and the voltage sensitivity is too high, the node is identified as having the characteristic of reactive power injection exceeding local demand and being transmitted in the opposite direction. Such nodes are marked as reactive power overcompensation risk nodes, forming a set of reactive power overcompensation risk nodes, including: For each node in the high-sensitivity node list, read the measured value of the reactive current component of the section where the node is located, the reactive power consumption of the local load, and the reactive current flow direction of the upstream and downstream adjacent sections. By comparing the measured value of the reactive current component in the section where the node is located with the reactive power consumption of the local load, when the measured value of the reactive current component is greater than the reactive power consumption of the local load and the reactive current flow direction shows a reverse flow from the node to the power supply side, the node is marked as a reactive power overcompensation risk node. The topological location, voltage sensitivity value, and amplitude of the reverse-transmitted reactive current component of all marked reactive overcompensation risk nodes are summarized to form a set of reactive overcompensation risk nodes.

6. The method for generating a power distribution network fault recovery strategy according to claim 4, characterized in that, After obtaining the list of high-sensitivity nodes, the following steps are taken: retrieve the measured values ​​of reactive current components in the sections where each node is located and the reactive current flow direction information of the upstream and downstream adjacent sections; analyze the inflow and outflow difference of reactive current at each node and the reactive power transfer direction between it and the adjacent sections; identify the node locations where the reactive power injection amount significantly exceeds the local load demand and there is a reverse transmission phenomenon to the power supply side; and mark the amplitude and duration of the reverse transmission of reactive power to form a set of reactive power overcompensation risk nodes.

7. The method for generating a power distribution network fault recovery strategy according to claim 1, characterized in that, The process of generating optimization criteria for the commissioning location applicable to the current fault recovery based on the path migration performance of reactive power flow at different compensation locations includes: For each risk node in the set of reactive power overcompensation risk nodes, the compensation point is shifted segment by segment along the feeder topology in a direction away from the reactive power deficit concentration area. After each shift, the node voltage sensitivity and reactive power flow direction at the shift position are recalculated by the power flow calculation unit, and the change in voltage sensitivity before and after the shift is recorded. Based on the number of offset segments obtained from multiple offsets and the corresponding voltage sensitivity values, the relationship between the decrease in voltage sensitivity as the compensation point moves away from the deficit area is determined. The correspondence is matched with the feeder topology and reactive power deficit concentration area distribution in the current fault recovery scenario to obtain the appropriate offset range of each compensation point and the corresponding expected voltage sensitivity change, forming the basis for commissioning location optimization.

8. The method for generating a power distribution network fault recovery strategy according to claim 1, characterized in that, The optimization of commissioning location is used as a basis to evaluate the voltage sensitivity change of each compensation point after it is moved away from the deficit area. Compensation points whose voltage sensitivity drops to a safe range and whose reactive power flow returns to normal after relocation are selected, generating a list of compensation devices that require dynamic adjustment, including: Based on the optimization criteria for the commissioning location, the appropriate offset range of each compensation point is determined, and the target location is the position with the lowest voltage sensitivity within the specified range. Solve for the voltage sensitivity values ​​and reactive current flow direction of each node after offset; When the voltage sensitivity value drops below the safety upper limit threshold after offset and the reactive current flow direction returns to the positive direction pointing towards the load side, the compensation point is determined as the compensation point to be adjusted. Summarize the current and target positions of the compensation points to be adjusted to form a list of compensation devices that need to be dynamically adjusted.

9. The method for generating a power distribution network fault recovery strategy according to claim 7, characterized in that, After establishing the basis for optimizing the commissioning location, the following steps are taken: obtaining the corresponding information of the compensation point migration distance and voltage sensitivity reduction in the commissioning location optimization basis; evaluating the voltage sensitivity change and reactive power flow improvement status of each compensation point after migrating it to a different distance away from the deficit area according to the optimization basis; identifying the compensation points whose voltage sensitivity can be reduced to a safe range and whose reactive power flow direction can be restored from reverse to forward flow after migration, and their optimal migration locations; establishing a list of compensation devices that need to be dynamically adjusted and marking the current location and target migration location of each device.

10. The method for generating a power distribution network fault recovery strategy according to claim 1, characterized in that, The process of integrating the set of reactive power overcompensation risk nodes with the list of compensation devices that need dynamic adjustment, eliminating reverse reactive power flow and reducing line losses on reactive power transmission paths, and generating a fault recovery strategy scheme that minimizes network losses includes: The risk node topology location in the set of risk nodes with no overcompensation is matched with the current location of the compensation device in the list of compensation devices that need to be dynamically adjusted, and the reverse transmission information of the successfully matched risk nodes is merged with the target migration location information of the corresponding compensation device. According to the order of the reactive power amplitude transmitted in reverse from large to small, the matching compensation devices are moved from the current position to the target migration position one by one. After each migration is completed, the reactive power distribution and line loss of the entire network after the migration are calculated. Summarize the target locations of the compensation devices, the direction of reactive power flow, and the line loss values ​​after all migrations are completed to form a fault recovery strategy.

Images