Emergency strategy generation method and system applied to power distribution network disasters

Abstract

The invention provides an emergency strategy generation method and system applied to power distribution network disasters, and the method comprises the steps: obtaining multi-mode sensing data of a power distribution network operation environment, the multi-mode sensing data comprising an equipment state sensing sequence, an environment disturbance sensing sequence and a load response sensing sequence; disaster evolution feature modeling is carried out on the multi-modal sensing data, and disaster evolution feature flow is generated; constructing a strategy generation knowledge graph based on the disaster evolution feature flow; strategy reasoning is carried out through the strategy generation knowledge graph, and emergency strategy candidate streams are generated; and carrying out collaboration verification on the emergency strategy candidate flow to generate a power distribution network disaster emergency strategy. According to the method, the overall effectiveness and implementation feasibility of the emergency strategy can be improved, so that more scientific and reliable strategy support is provided for power distribution network disaster response.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a method and system for generating an emergency strategy for power distribution network disasters. Background Art

[0002] At present, the emergency response strategy for distribution network disasters is obtained by obtaining equipment status data or environmental data, extracting equipment abnormality thresholds or environmental indicators from them, and then generating emergency response strategies based on a preset rule base or a simple mapping table. This method easily leads to one-sided information and is difficult to reflect the dynamic correlation nature of disaster development. Simple rule matching cannot adapt to changes in different disaster scenarios. The generated emergency response strategies are insufficient in scientificity, adaptability and feasibility, and are difficult to meet the actual needs of complex distribution network disaster response. Summary of the Invention

[0003] The present invention provides a method and system for generating an emergency strategy for power distribution network disasters.

[0004] In the first aspect, an embodiment of the present invention provides a method for generating emergency strategies for distribution network disasters, the method comprising: obtaining multimodal perception data of the distribution network operating environment, the multimodal perception data comprising a device status perception sequence, an environmental disturbance perception sequence, and a load response perception sequence; performing disaster evolution feature modeling on the multimodal perception data to generate a disaster evolution feature flow; constructing a strategy generation knowledge graph based on the disaster evolution feature flow; performing strategy reasoning through the strategy generation knowledge graph to generate an emergency strategy candidate flow; performing collaborative verification on the emergency strategy candidate flow to generate a distribution network disaster emergency strategy.

[0005] In a second aspect, an embodiment of the present invention provides a computer system, comprising: a memory storing a computer program; and a processor for loading the computer program to implement the above-mentioned method for generating an emergency strategy for distribution network disasters.

[0006] The emergency strategy generation method for distribution network disasters provided by the present invention integrates three types of perception sequences: equipment status, environmental disturbance, and load response by acquiring multimodal perception data, providing a more comprehensive input basis for disaster analysis and avoiding the problem of information one-sidedness caused by traditional methods relying on only a single data type; by modeling the disaster evolution characteristics, a disaster evolution feature flow including equipment status degradation correlation characteristics, environmental disturbance coupling characteristics, and load response sensitivity characteristics is generated, and traditional static characteristics are transformed into dynamic correlation characteristics reflecting time synchronization, spatial overlap, and causal dependence, which is more in line with the dynamic correlation nature of disaster development and enhances the depth of understanding of the causes of disasters; by constructing a strategy generation knowledge graph, a structured representation of the disaster evolution feature flow is generated. It shows disaster feature entities, strategy entities and association relationships, breaking through the limitation of traditional rule bases or simple mapping tables that can only handle single associations, and providing a more flexible and scalable knowledge basis for strategy generation; through strategy reasoning through knowledge graphs to generate emergency strategy candidate flows, it can dynamically match strategies according to real-time disaster characteristics, adapt to changes in different disaster scenarios, and avoid the rigidity of traditional fixed rule matching; by conducting collaborative verification of candidate strategy flows, comprehensively considering the resource usage, time execution, and effect target synergy between strategies, it avoids strategy conflicts that may be caused by single strategy verification, improves the overall effectiveness and implementation feasibility of emergency strategies, and thus provides more scientific and reliable strategy support for distribution network disaster response. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 This is a flow chart of a method for generating an emergency strategy for distribution network disasters provided by an embodiment of the present invention.

[0008] Figure 2 It is a schematic diagram of the composition of a computer system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0009] See also Figure 1 , Figure 1 A flowchart of a method for generating an emergency strategy for a power distribution network disaster provided by an embodiment of the present invention, which can be executed by a computer system, includes the following steps: Step S100: Acquire multimodal sensing data of the distribution network operating environment, where the multimodal sensing data includes a device state sensing sequence, an environmental disturbance sensing sequence, and a load response sensing sequence.

[0010] Multimodal sensing data is data acquired from multiple dimensions and sensor types, comprehensively reflecting different aspects of the distribution network's operating environment. A device status sensing sequence is a series of data about the operating status of devices in the distribution network, reflecting the various parameters of the devices at different times. For example, the time-varying sequence of parameters such as voltage, current, and temperature. An environmental disturbance sensing sequence is a data sequence that records various disturbance factors that may affect the distribution network. These disturbance factors may include the time-varying variations of environmental factors such as wind speed, precipitation, and thunderstorms. A load response sensing sequence describes the response of the distribution network load under different circumstances. It covers data related to different load states, such as growth, stability, and decline, including information such as load power changes and load fluctuation ranges. For the device status sensing sequence, voltage sensors can be used to collect real-time voltage data from devices, current sensors to collect current data, and temperature sensors to collect temperature data. These sensors collect data at regular intervals to form the device status sensing sequence. For the environmental disturbance sensing sequence, anemometers can be used to measure wind speed, rain gauges to measure precipitation, and lightning monitoring equipment to monitor thunderstorms. Similarly, data is recorded at regular intervals to form the environmental disturbance sensing sequence. For the load response perception sequence, a power sensor can be used to monitor the power changes of the load. By monitoring the load in different areas, the load fluctuation area can be determined, and then a load response perception sequence can be formed.

[0011] Step S200: Modeling the disaster evolution characteristics of the multimodal perception data to generate a disaster evolution characteristic stream.

[0012] Disaster evolution feature modeling analyzes and processes multimodal sensory data to extract characteristic information related to the evolution of distribution network disasters. The disaster evolution feature stream integrates these extracted features according to a specific temporal relationship, forming a data stream that reflects the characteristic changes of distribution network disasters from their onset to their development. Through in-depth analysis of multimodal sensory data, the interrelationships and changing patterns between equipment status, environmental disturbances, and load responses can be discovered.

[0013] As an implementation manner, step S200 may be specifically implemented as the following steps S210 to S240: Step S210: Performing temporal correlation analysis on the device state perception sequence and the environmental disturbance perception sequence to generate device state degradation correlation features. The device state degradation correlation features include temporal synchronization information and change trend dependency information of device parameter changes and environmental disturbances.

[0014] Time series correlation analysis analyzes the temporal relationship between the device state perception sequence and the environmental disturbance perception sequence, identifying the synchronization and dependency of the change trends between the two. The device state degradation correlation feature, obtained through this analysis, is characteristic information that reflects the degradation of the device state due to environmental disturbances. Time series synchronization information reflects the degree of overlap between device parameter change events and environmental disturbance events on the time axis, such as whether an abnormal voltage increase event on the device occurs at the same time as a sudden increase in wind speed. Dependency information on change trends reflects the inherent connection between device parameter change trends and environmental disturbance trends, such as whether the rising trend in device temperature is correlated with the decreasing trend in precipitation.

[0015] As an implementation method, step S210 performs a temporal correlation analysis on the device state perception sequence and the environmental disturbance perception sequence to generate a device state degradation correlation feature, which can be specifically implemented as follows: steps S211 to S215: Step S211: extracting device parameter change event points in the device status perception sequence, where the device parameter change event points include abnormal voltage increase event points, abnormal current decrease event points, and abnormal temperature persistence event points.

[0016] Device parameter change events are key points in the device status perception sequence where abnormal changes in device parameters occur. Abnormal voltage increase events occur when the device voltage exceeds the upper limit of the normal operating range. This may be caused by line faults, sudden load changes, and other factors. Abnormal current decrease events occur when the device current falls below the lower limit of the normal operating range. This may be caused by factors such as device faults and line disconnections. Sustained temperature anomaly events occur when the device temperature remains above the normal temperature threshold for a period of time. Prolonged temperature anomalies may have a serious impact on the performance and lifespan of the device.

[0017] To identify these event points, you can set corresponding thresholds. For abnormal voltage rise events, determine a reasonable upper voltage threshold based on the device's rated voltage and historical operating data. When the device voltage exceeds this threshold, record the time point as the abnormal voltage rise event. For abnormal current decrease events, set a lower current threshold. When the current falls below this threshold, record the corresponding time point. For persistent temperature anomaly events, set a temperature threshold and a duration threshold. When the device temperature exceeds the temperature threshold and lasts for a duration equal to the duration threshold, record the time point.

[0018] Step S212: extracting environmental disturbance event points in the environmental disturbance perception sequence, where the environmental disturbance event points include wind speed sudden increase event points, precipitation sudden increase event points, and thunderstorm occurrence event points.

[0019] An environmental disturbance event point is a critical time point in the environmental disturbance perception sequence when an abnormal change in an environmental factor occurs. A sudden wind speed increase event point is a time point when the wind speed increases sharply to exceed a certain threshold within a short period of time. Strong winds may cause damage to overhead lines, poles, and other equipment in the distribution network. A sudden precipitation increase event point is a time point when precipitation increases significantly within a short period of time, exceeding the normal precipitation range. Excessive precipitation may cause equipment to become damp, short-circuit, and other problems. A thunderstorm occurrence event point is a time point when a thunderstorm is detected. The lightning generated by a thunderstorm may cause direct lightning damage to distribution network equipment.

[0020] To extract these event points, it's necessary to set corresponding thresholds based on environmental monitoring data. For sudden wind speed increases, a wind speed threshold can be determined based on local climate conditions and historical wind speed data. When the wind speed exceeds this threshold within a short period of time, the time point is recorded. For sudden precipitation increases, a precipitation threshold is set. When the precipitation increases by more than this threshold within a short period of time, the time point is recorded. For thunderstorm events, lightning signals can be monitored in real time using lightning monitoring equipment. When a lightning signal is detected, the time point is recorded.

[0021] Step S213: aligning the timestamps of the device parameter change event points and the environmental disturbance event points to obtain timing synchronization information, which includes the overlapping ratio information of the device parameter change event and the environmental disturbance event on the time axis.

[0022] Timestamp alignment arranges device parameter change events and environmental disturbance events in chronological order, allowing them to be compared on the same timeline. Timing synchronization information measures the degree of synchronization between device parameter change events and environmental disturbance events by calculating the overlap ratio between them. A higher overlap ratio indicates a higher likelihood of a correlation between the device parameter change event and the environmental disturbance event.

[0023] In specific implementation, the timestamps of device parameter change event points and environmental disturbance event points can be converted into a unified time format, and then these event points can be traversed to count the number of times device parameter change events and environmental disturbance events occur simultaneously within a certain time window, and then divided by the total number of events to obtain the overlap ratio.

[0024] Step S214: performing dependency analysis on the change trend of the device parameter change event point and the disturbance trend of the environmental disturbance event point to obtain dependency information of the change trend, which includes covariance information of the device parameter rising trend and the environmental disturbance increasing trend.

[0025] Dependency analysis studies the inherent connection between device parameter trends and environmental disturbance trends. The degree of dependence between the two is measured by calculating covariance. Covariance is a statistic that indicates whether the changing trends of two variables are consistent. A positive covariance indicates a positive correlation between the increasing trend of device parameters and the increasing trend of environmental disturbances, meaning that the two trends are consistent. A negative covariance indicates opposite trends. A covariance close to zero indicates little to no correlation between the two.

[0026] To calculate covariance, we first need to quantify the trends of device parameter change and environmental disturbance event points. This trend can be expressed by calculating the parameter change rate between adjacent event points. For example, for device parameter change event points, we calculate the voltage increase rate between adjacent abnormal voltage increase event points; for environmental disturbance event points, we calculate the wind speed increase rate between adjacent wind speed surge event points. These rates of change are then used as variables in the covariance calculation formula.

[0027] Step S215: combining the time series synchronization information and the dependency information of the change trend into a device state degradation correlation feature.

[0028] Combining time series synchronization information and dependency information on change trends creates a comprehensive feature, the device state degradation correlation feature. This feature more comprehensively reflects the degradation of device state due to environmental disturbances. By combining these two pieces of information, we can more accurately determine whether changes in device state are related to environmental disturbances and the strength of this correlation.

[0029] Step S220: Perform spatial coupling analysis on the environmental disturbance perception sequence and the load response perception sequence to generate an environmental disturbance coupling feature. The environmental disturbance coupling feature includes spatial overlap information of the environmental disturbance area and the load fluctuation area, and gradient correspondence information of the disturbance intensity and the load fluctuation amplitude.

[0030] Spatial coupling analysis analyzes the spatial correlation between environmental disturbance perception sequences and load response perception sequences, studying the spatial overlap between environmental disturbance regions and load fluctuation regions, as well as the gradient correspondence between disturbance intensity and load fluctuation amplitude. The environmental disturbance coupling feature, derived through this analysis, provides characteristic information reflecting the spatial correlation between environmental disturbances and load responses. Spatial overlap information reflects the degree of spatial overlap between environmental disturbance regions and load fluctuation regions, for example, whether high wind speed regions geographically overlap with high load growth regions. The gradient correspondence between disturbance intensity and load fluctuation amplitude reflects the correspondence between changes in environmental disturbance intensity and changes in load fluctuation amplitude, such as the relationship between the attenuation gradient of wind speed from the disturbance center to the boundary and the decreasing gradient of load growth from the load center to the boundary.

[0031] As an implementation manner, step S220 performs spatial coupling analysis on the environmental disturbance perception sequence and the load response perception sequence to generate an environmental disturbance coupling feature, which can be specifically implemented as follows: steps S221 to S227: Step S221: extracting disturbance area distribution information from the environmental disturbance perception sequence, where the disturbance area distribution information includes geographical location information of high wind speed areas, coverage information of heavy rainfall areas, and boundary coordinate information of high-frequency thunderstorm areas.

[0032] The disturbance area distribution information is information that describes the spatial distribution of environmental disturbances, including the specific location and range of different types of environmental disturbance areas. The geographical location information of high wind speed areas clarifies the specific location in geographic space of areas where wind speeds exceed a certain threshold, and can be represented by geographic coordinates. The coverage information of heavy precipitation areas describes the size and shape of areas where precipitation exceeds a certain standard, and can be represented by area, boundary coordinates, etc. The boundary coordinate information of high-frequency thunderstorm areas determines the boundary position of areas with a higher frequency of thunderstorms, and defines the area through a series of coordinate points.

[0033] To extract this information, geographic information systems (GIS) and environmental monitoring data can be used. For the geographic location of high-wind-speed areas, data collected by anemometers can be imported into the GIS system. High-wind-speed areas can be delineated based on wind speed thresholds, and their geographic coordinates can be obtained. For coverage information on heavy rainfall areas, the boundaries of these areas can be mapped in the GIS system using rain gauge distribution and precipitation data, and their coverage calculated. For boundary coordinate information on high-frequency thunderstorm areas, lightning location data recorded by lightning monitoring equipment can be used to calculate thunderstorm frequency in the GIS system, delineate high-frequency thunderstorm areas, and obtain their boundary coordinates.

[0034] Step S222: extracting load fluctuation area distribution information from the load response perception sequence, where the load fluctuation area distribution information includes geographical location information of high load growth areas, coverage information of load drop areas, and boundary coordinate information of load oscillation areas.

[0035] Load fluctuation area distribution information describes the spatial fluctuations of distribution network load, including the specific locations and ranges of different types of load fluctuation areas. The geographic location information for high-load growth areas specifies the specific geographic locations of areas where the load growth rate exceeds a certain threshold and can be represented by geographic coordinates. The coverage information for load drop areas describes the size and shape of areas where the load suddenly drops and can be represented by areas, boundary coordinates, and other information. The boundary coordinate information for load oscillation areas identifies the boundaries of areas where the load fluctuates significantly within a certain range, defining these areas through a series of coordinate points. To extract this information, power load monitoring data and a GIS system can be combined. Data collected by power sensors at each load node is imported into the GIS system. Based on thresholds such as load growth rate, load drop amplitude, and load fluctuation range, high-load growth areas, load drop areas, and load oscillation areas are then delineated. The geographic location information, coverage information, and boundary coordinate information for these areas are then obtained.

[0036] Step S223: performing spatial overlap calculation on the disturbance area distribution information and the load fluctuation area distribution information to obtain spatial overlap information, where the spatial overlap information includes intersection area information of the high wind speed area and the high load growth area.

[0037] Spatial overlap calculations are performed to determine the overlap between disturbance zones and load fluctuation zones in geographic space. Spatial overlap information measures the degree of spatial overlap between environmental disturbance zones and load fluctuation zones by calculating metrics such as the intersection area of ​​these zones. The intersection area of ​​high wind speed zones and high load growth zones is a key component of spatial overlap information, reflecting the specific conditions of high load growth zones in high wind speed environments.

[0038] In practice, the distribution of disturbance areas and load fluctuation areas can be overlaid and analyzed in a GIS system. The geographic data of high wind speed areas and high load growth areas can be imported into the GIS system, and their intersection area can be calculated using the GIS's spatial analysis capabilities.

[0039] Step S224: extracting disturbance intensity gradient information from the environmental disturbance perception sequence, where the disturbance intensity gradient information includes the wind speed attenuation gradient, precipitation decrease gradient, and thunderstorm frequency reduction gradient from the disturbance center to the boundary.

[0040] Disturbance intensity gradient information describes the spatial variation in the intensity of environmental disturbances, reflecting the intensity variation trend of environmental disturbances from the center to the boundary. The wind speed attenuation gradient indicates the degree of wind speed attenuation from the center to the boundary of a high wind speed area, and can be expressed by calculating the ratio of the wind speed difference between adjacent locations to the distance. The precipitation decrease gradient indicates the degree of precipitation reduction from the center to the boundary of a heavy precipitation area, and is also expressed by calculating the ratio of the precipitation difference between adjacent locations to the distance. The thunderstorm frequency reduction gradient indicates the degree of thunderstorm frequency reduction from the center to the boundary of a high-frequency thunderstorm area, and is expressed by calculating the ratio of the thunderstorm frequency difference between adjacent locations to the distance.

[0041] To extract this information, multiple monitoring points can be set up within the environmental disturbance area to collect environmental data at different locations. For the wind speed attenuation gradient, a wind speed monitoring point is set up at regular intervals within the high wind speed area, and the wind speed data at each monitoring point is recorded. The ratio of the wind speed difference between adjacent monitoring points to the distance is then calculated to obtain the wind speed attenuation gradient. For the precipitation decrease gradient, rain gauges are rationally distributed within the heavy precipitation area, precipitation data at different locations is recorded, and the ratio of the precipitation difference to the distance between adjacent locations is calculated. For the thunderstorm frequency reduction gradient, lightning monitoring equipment is used to monitor within the high-frequency thunderstorm area, the frequency of thunderstorms at different locations is counted, and the ratio of the thunderstorm frequency difference to the distance between adjacent locations is calculated.

[0042] Step S225: extracting the load fluctuation amplitude gradient information in the load response perception sequence, where the load fluctuation amplitude gradient information includes a decreasing gradient of the load increase amount and an increasing gradient of the load decrease amount from the load center to the load boundary.

[0043] Load fluctuation gradient information describes the spatial variation of load fluctuations in the distribution network, reflecting the load fluctuation trend from the center to the edge. The load increase gradient indicates the degree to which the load increases gradually from the center to the edge of a high-load growth area. This is expressed by calculating the ratio of the load increase difference between adjacent locations to the distance. The load decrease gradient indicates the degree to which the load decreases gradually from the center to the edge of a load drop area. This is also expressed by calculating the ratio of the load decrease difference between adjacent locations to the distance.

[0044] To extract this information, multiple load monitoring points are set up within the load fluctuation area. Within the high-load growth area, load monitoring points are set up at regular intervals. Load power data at each location is recorded, and the ratio of the load increase difference between adjacent monitoring points to the distance is calculated to determine the load increase gradient. Within the load drop area, load monitoring points are also set up, load power data is recorded, and the ratio of the load reduction difference between adjacent locations to the distance is calculated to determine the load reduction gradient.

[0045] Step S226: performing gradient correspondence analysis on the disturbance intensity gradient information and the load fluctuation amplitude gradient information to obtain gradient correspondence information, which includes linear fitting relationship information between the wind speed attenuation gradient and the load growth amount decreasing gradient.

[0046] Gradient correspondence analysis studies the inherent connection between the gradient of environmental disturbance intensity and the gradient of load fluctuation amplitude. By analyzing the correspondence between them, the relationship between the two can be identified. Linear fitting relationship information is an important manifestation of gradient correspondence information. It indicates whether there is a linear relationship between the wind speed attenuation gradient and the load growth gradient, as well as the specific parameters of this linear relationship.

[0047] To perform gradient correspondence analysis, the extracted disturbance intensity gradient information and load fluctuation amplitude gradient information can be processed. The wind speed decay gradient and the load growth gradient data are used as two variables for fitting analysis using a linear regression algorithm. First, the wind speed decay gradient and the load growth gradient data are organized into a dataset. Then, a linear regression model is trained on this dataset to obtain a linear fitting equation.

[0048] Step S227: combining the spatial overlap information and the gradient correspondence information into an environmental disturbance coupling feature.

[0049] Combining spatial overlap and gradient correspondence information creates a comprehensive feature, the environmental disturbance coupling feature. This feature more comprehensively reflects the spatial correlation and gradient correspondence between environmental disturbances and load responses. By combining these two aspects of information, the extent and manner in which environmental disturbances affect load fluctuations can be more accurately determined.

[0050] Step S230: Perform causal dependency analysis on the load response perception sequence and the device state perception sequence to generate a load response sensitive feature. The load response sensitive feature includes causal chain information of the load change event and the device state change event and transmission path information of the impact of the load change on the device state.

[0051] Causal dependency analysis studies the causal relationship between load response perception sequences and device state perception sequences, identifying the intrinsic connection between load change events and device state change events, and how load changes affect device states. Causal chain information describes the causal relationship between load change events and device state change events, clarifying which load change events directly or indirectly lead to which device state change events. Impact transfer path information describes the method and path through which load changes affect device states, for example, how load increases are transmitted through current parameters to device temperature.

[0052] Data mining and machine learning algorithms can be used to perform causal dependency analysis. First, data preprocessing is performed on the load response and device status perception sequences to extract key events and parameters. Then, using causal relationship mining algorithms, such as the Granger causality test algorithm, the causal relationship between load change events and device status change events is analyzed to obtain causal chain information. For information on the impact transmission path, a path analysis model can be constructed, combining the physical principles and electrical characteristics of the device to analyze the impact path of the load change on the device status.

[0053] As an implementation method, step S230 performs causal dependency analysis on the load response perception sequence and the device state perception sequence to generate a load response sensitive feature, which can be specifically implemented as follows: steps S231 to S237: Step S231: extracting a load change event chain in the load response sensing sequence, where the load change event chain includes a load increase triggering event, a load stability maintaining event, and a load reduction termination event.

[0054] A load change event chain is a series of events that describe how the load changes over time. A load increase trigger event is the initial event that triggers a load increase, which may be caused by a new load connection or the startup of production equipment. A load stability maintenance event is an event in which the load remains relatively stable for a period of time, with minimal power changes. A load reduction termination event is an event in which the load begins to decrease and eventually stops decreasing, which may be caused by factors such as the shutdown of load equipment or the end of production.

[0055] To extract the load change event chain, event detection can be performed on the load response perception sequence. Load increase trigger events and load decrease termination events are determined by setting load power change thresholds. When the load power increases by more than a certain threshold within a short period of time, it is considered a load increase trigger event; when the load power decreases by more than a certain threshold within a short period of time, it is considered a load decrease termination event. Load stability maintenance events can be determined by determining whether the load power change over a period of time is less than a smaller threshold.

[0056] Step S232: extracting the device state change event chain in the device state perception sequence, where the device state change event chain includes a device normal operation event, a device abnormality warning event, and a device failure shutdown event.

[0057] A device state change event chain is a series of events that describe the transitions between different states of a device. A normal operation event occurs when the device is operating normally, with all parameters within normal ranges. An abnormality warning event occurs when certain parameters of the device are abnormal but not yet at a failure state. In these cases, timely action is required to prevent device failure. A downtime event occurs when a device stops operating due to a fault. The device cannot function properly and requires repair or replacement.

[0058] To extract the chain of device state change events, event detection can be performed on the device state perception sequence. By setting normal ranges and abnormal thresholds for device parameters, normal device operation events, abnormal device warning events, and fault shutdown events can be determined. When all device parameters are within the normal range, this is considered a normal device operation event. When certain device parameters exceed the normal range but do not reach the fault threshold, this is considered a fault warning event. When certain device parameters reach the fault threshold, this is considered a fault shutdown event.

[0059] Step S233: Perform causal relationship mining on the load change event chain and the device state change event chain to obtain causal chain information. The causal chain information includes direct causal relationship information from the load increase trigger event to the device abnormality warning event and indirect causal relationship information from the load stability maintenance event to the device normal operation event.

[0060] Causal relationship mining is the process of identifying the causal relationships between load change event chains and device status change event chains. Direct causal relationship information clarifies the specific circumstances and probabilities of load increase triggering events directly leading to device abnormality warning events. Indirect causal relationship information describes the circumstances in which load stability maintenance events indirectly lead to normal device operation events through some intermediate link.

[0061] To mine causal relationships, causal analysis algorithms, such as causal graph models and Bayesian networks, can be used. First, data on load change event chains and device state change event chains are organized to form a dataset. Then, a causal analysis model is trained on the dataset to identify causal relationships.

[0062] Step S234: extracting load variation information from the load response sensing sequence, where the load variation information includes absolute value information of load increase, absolute value information of load decrease, and relative proportion information of load change.

[0063] Load change information describes the degree of load change, reflecting load changes from different perspectives. The absolute value of load increase is the specific power increase during the load increase, directly reflecting the scale of the load increase. The absolute value of load decrease is the specific power decrease during the load decrease, reflecting the extent of the load decrease. The relative proportion of load change is the ratio of the load change to the initial load value, allowing for a more intuitive comparison of the degree of change in different loads.

[0064] In order to extract the load change information, the load response sensing sequence can be processed to obtain the absolute numerical information of the load increase and the absolute numerical information of the load decrease by calculating the difference between adjacent load data points.

[0065] Step S235: extracting device state impact information from the device state sensing sequence, where the device state impact information includes information on the degree of impact of the load change on the device voltage, the device current, and the device temperature.

[0066] Device status impact information is information that describes the specific impact of load changes on device status, reflecting the effects of load changes on devices from different perspectives. Information on the extent of the load change's impact on device voltage indicates the specific value or ratio of the device voltage change caused by the load change, which can be obtained by measuring the difference or change ratio of the device voltage before and after the load change. Information on the extent of the load change's impact on device current indicates the specific value or ratio of the device current change caused by the load change, which can also be obtained by measuring the difference or change ratio of the device current before and after the load change. Information on the extent of the load change's impact on device temperature indicates the specific value or trend of the device temperature change caused by the load change, which can be determined by monitoring the device temperature change before and after the load change using a temperature sensor.

[0067] Step S236: performing a transmission path analysis on the load change information and the device state impact information to obtain impact transmission path information, where the impact transmission path information includes path information of how the load increase is transmitted to the device temperature via the current parameter.

[0068] Transfer path analysis studies how load changes affect device status through various intermediate parameters and links. Impact transfer path information clarifies the specific paths and methods by which load changes affect device status. A key aspect of impact transfer path information is the path by which load increases are transmitted through current parameters to device temperature. This describes how the load increase affects current and, in turn, device temperature.

[0069] To perform transfer path analysis, circuit principles and the physical characteristics of the device can be combined. First, analyze the relationship between load increase and current. According to Ohm's law, an increase in load increases current. Next, analyze the effect of increased current on device temperature. Current passing through the device's resistance generates heat. According to Joule's law, as current increases, the heat generated by the device increases, leading to an increase in device temperature.

[0070] Step S237: combining the causal chain information and the impact transfer path information into a load response sensitivity feature.

[0071] Combining causal chain information and impact transfer path information creates a comprehensive feature, the load response sensitivity feature. This feature more comprehensively reflects the impact of load changes on device status, including the causal relationship between load change events and device status change events, as well as the impact path of the load change on the device status. By combining these two aspects of information, the extent and method of the load change's impact on the device can be more accurately determined.

[0072] Step S240: Perform feature time series fusion on the equipment state degradation correlation features, environmental disturbance coupling features, and load response sensitivity features to generate a disaster evolution feature flow.

[0073] Feature time series fusion integrates different types of features in chronological order, enabling them to correlate and complement each other over time, forming a continuous feature stream that reflects the evolution of a disaster. This fusion generates a comprehensive feature data stream that encompasses the temporal evolution of multiple aspects of information, including device status, environmental disturbances, and load response. This stream more comprehensively reflects the entire process of a distribution network disaster, from its onset to its development.

[0074] As an implementation method, step S240 performs a time series fusion of the equipment state degradation correlation features, the environmental disturbance coupling features, and the load response sensitivity features to generate a disaster evolution feature stream. Specifically, the steps S241 to S244 are as follows: Step S241: extracting the timing synchronization information from the device state degradation correlation feature, and determining the time weight coefficient of the device feature. The time weight coefficient is positively correlated with the overlap ratio of the device parameter change event and the environmental disturbance event.

[0075] Temporal synchronization is a key component of device state degradation correlation features, reflecting the degree of temporal synchronization between device parameter change events and environmental disturbance events. The time weight coefficient measures the importance of device characteristics in the disaster evolution process and is positively correlated with the overlap ratio between device parameter change events and environmental disturbance events. The higher the overlap ratio, the greater the time weight coefficient.

[0076] To determine the time weight coefficient of the device feature, a calculation can be performed based on the overlap ratio in the timing synchronization information. First, the overlap ratio in the extracted timing synchronization information is normalized so that its value ranges from 0 to 1. Then, the normalized overlap ratio is used as the time weight coefficient.

[0077] Step S242: extracting spatial overlap information from the environmental disturbance coupling feature, and determining a spatial weight coefficient of the environmental feature. The spatial weight coefficient is positively correlated with the intersection area of ​​the environmental disturbance region and the load fluctuation region.

[0078] Spatial overlap is a key component of environmental disturbance coupling, reflecting the degree of spatial overlap between areas of environmental disturbance and areas of load fluctuation. The spatial weight coefficient measures the importance of environmental features in the disaster evolution process and is positively correlated with the intersection area between the environmental disturbance and load fluctuation areas. The larger the intersection area, the greater the spatial weight coefficient. To determine the spatial weight coefficient of an environmental feature, it can be calculated based on the intersection area in the spatial overlap information. First, the intersection area extracted from the spatial overlap information is normalized to a value between 0 and 1. The normalized intersection area is then used as the spatial weight coefficient.

[0079] Step S243: extracting the causal chain information from the load response sensitive feature and determining the causal weight coefficient of the load feature. The higher the clarity of the direct causal relationship of the causal chain information, the greater the causal weight coefficient.

[0080] Causal chain information is a crucial component of load response sensitivity, reflecting the causal relationship between load change events and equipment state changes. The causal weight coefficient measures the importance of load characteristics in the disaster evolution process and is positively correlated with the clarity of the direct causal relationship in the causal chain information. The clearer the direct causal relationship, the greater the causal weight coefficient.

[0081] To determine the causal weight coefficient of the load signature, causal chain information can be quantitatively analyzed. A direct causal clarity index can be defined to measure the clarity of the direct causal relationships in the causal chain information. For example, the direct causal clarity index can be calculated based on factors such as the number of direct causal relationships in the causal chain information and the strength of the causal relationships. The direct causal clarity index is then normalized to a value between 0 and 1, and the normalized index is used as the causal weight coefficient.

[0082] Step S244: Based on the time weight coefficient, space weight coefficient and causal weight coefficient, the equipment state degradation correlation characteristics, environmental disturbance coupling characteristics and load response sensitivity characteristics are weighted and superimposed in time series to generate a disaster evolution characteristic flow with dynamic weights. The dynamic weights are used to characterize the degree of dominant role of different characteristics in the disaster evolution process.

[0083] Weighted time series superposition combines the characteristics associated with equipment degradation, coupled with environmental disturbances, and sensitive load response in chronological order, weighting them according to temporal, spatial, and causal weight coefficients. Dynamic weighting allows the weights of different characteristics to change over time and with the disaster's severity, more accurately reflecting their dominant role in the disaster's evolution.

[0084] In specific implementation, the device state degradation-related features, environmental disturbance coupling features, and load response sensitivity features are first arranged in chronological order. Each feature is then weighted according to the temporal weight coefficient, spatial weight coefficient, and causal weight coefficient. For example, for the device state degradation-related features, the feature value at each time point is multiplied by the temporal weight coefficient; for the environmental disturbance coupling features, the feature value at each time point is multiplied by the spatial weight coefficient; and for the load response sensitivity features, the feature value at each time point is multiplied by the causal weight coefficient. Finally, the weighted features are superimposed to obtain a disaster evolution feature flow with dynamic weights.

[0085] Step S300: Generate a knowledge graph based on a strategy building process of disaster evolution feature flow.

[0086] The strategy generation knowledge graph is a graph structure used to represent the knowledge and relationships related to distribution network disaster emergency response strategies. It associates and integrates various characteristic information in the disaster evolution feature stream with the corresponding emergency response strategies, so that appropriate emergency response strategies can be quickly and accurately generated when a disaster occurs. By constructing a strategy generation knowledge graph, complex disaster characteristics and strategy information can be visualized and structured.

[0087] As an implementation method, step S300, based on the disaster evolution characteristic flow construction strategy to generate a knowledge graph, can be specifically implemented as the following steps S310 to S370: Step S310: extracting equipment state degradation related features from the disaster evolution feature stream, and defining the time series synchronization information and the dependency information of the change trend in the equipment state degradation related features as the first type of disaster feature entity.

[0088] Equipment state degradation correlation features are a key component of the disaster evolution feature stream, reflecting the degradation of equipment status due to environmental disturbances. Key information in these features is temporal synchronization and dependency information on changing trends. These are defined as first-class disaster feature entities for easy representation and association in the knowledge graph.

[0089] To extract features associated with equipment state degradation, the disaster evolution feature stream can be decomposed. By analyzing the device features within the stream, temporal synchronization information and dependency information on change trends can be extracted. For example, within the disaster evolution feature stream, data segments associated with features associated with equipment state degradation are identified. The overlap ratio between equipment parameter change events and environmental disturbance events (temporal synchronization information) and the covariance between the upward trend of equipment parameters and the increasing trend of environmental disturbances (dependency information on change trends) are extracted from these segments. These information are then used as attributes of the first-class disaster feature entity.

[0090] Step S320: extracting the equipment intervention strategy from the preset strategy library, defining the type information and operation process information of the equipment intervention strategy as a first-type strategy entity; and establishing a first association relationship between the first-type disaster feature entity and the first-type strategy entity.

[0091] The preset strategy library is a database that pre-stores various distribution network disaster emergency response strategies. These include equipment intervention strategies, environmental buffering strategies, and load regulation strategies for different disaster scenarios. Equipment intervention strategies are used to address equipment degradation. Their type and operational process information are defined as first-class strategy entities for easy representation and association in the knowledge graph.

[0092] To extract equipment intervention strategies, strategies related to equipment state degradation can be screened from the preset strategy library. For example, strategies for equipment preventive maintenance and equipment emergency isolation are included. The type information (e.g., preventive maintenance, emergency isolation) and operational process information (e.g., maintenance scope determination rules, isolation operation priority rules, etc.) of these strategies are used as attributes of the first-class strategy entity. Then, a first association relationship is established between the first-class disaster feature entity and the first-class strategy entity. This association relationship represents the connection between the equipment state degradation and the corresponding equipment intervention strategy. For example, when the overlap ratio between equipment parameter change events and environmental disturbance events is high, an equipment preventive maintenance strategy may be necessary; when the covariance between the upward trend of equipment parameters and the increasing trend of environmental disturbances is large, an equipment emergency isolation strategy may be necessary.

[0093] Step S330: extracting environmental disturbance coupling features from the disaster evolution feature flow, and defining the spatial overlap information and gradient correspondence information in the environmental disturbance coupling features as a second type of disaster feature entity.

[0094] Environmental disturbance coupling features are another important component of the disaster evolution feature stream, reflecting the spatial correlation and gradient correspondence between environmental disturbances and load responses. Spatial overlap and gradient correspondence are key components of environmental disturbance coupling features. They are defined as second-class disaster feature entities for easy representation and association in the knowledge graph.

[0095] To extract the environmental disturbance coupling characteristics, the disaster evolution characteristic stream can be decomposed. By analyzing the environmental characteristics within the disaster evolution characteristic stream, spatial overlap information and gradient correspondence information can be extracted. For example, within the disaster evolution characteristic stream, data segments related to the environmental disturbance coupling characteristics are identified. From these, the intersection area between the environmental disturbance region and the load fluctuation region (spatial overlap information) and the linear fit relationship between the wind speed attenuation gradient and the load growth gradient (gradient correspondence information) are extracted. This information is then used as attributes of the second-category disaster characteristic entity.

[0096] Step S340: extracting an environmental buffer strategy from a preset strategy library, defining the type information and implementation requirement information of the environmental buffer strategy as a second-type strategy entity; and establishing a second association relationship between the second-type disaster feature entity and the second-type strategy entity.

[0097] The environmental buffer strategy is a strategy used to deal with the impact of environmental disturbances on the distribution network. Its type information and implementation requirement information are defined as the second type of strategy entity so that it can be represented and associated in the knowledge graph.

[0098] To extract environmental buffering strategies, strategies related to environmental disturbances can be screened from the preset strategy library. For example, regional joint protection strategies and gradient-graded protection strategies can be used. The type information (e.g., regional joint protection, gradient-graded protection) and implementation requirements (e.g., rules for the coordinated allocation of protection resources, rules for adjusting protection intensity as the gradient changes, etc.) of these strategies are used as attributes of the second-type strategy entity. Then, a second association relationship is established between the second-type disaster characteristic entity and the second-type strategy entity. This association relationship represents the connection between the environmental disturbance and the corresponding environmental buffering strategy. For example, when the intersection area of ​​the environmental disturbance area and the load fluctuation area is large, a regional joint protection strategy may be necessary; when the linear fit relationship between the wind speed attenuation gradient and the load growth gradient is good, a gradient-graded protection strategy may be necessary.

[0099] Step S350: extracting load response sensitive features from the disaster evolution feature flow, and defining the causal chain information and impact transmission path information in the load response sensitive features as the third type of disaster feature entity.

[0100] Load response sensitive features are another important component of the disaster evolution feature stream, reflecting the impact of load changes on equipment status. Causal chain information and impact transmission path information are key information in load response sensitive features. They are defined as the third type of disaster feature entities to facilitate representation and association in the knowledge graph.

[0101] To extract sensitive load response features, the disaster evolution feature stream can be decomposed. By analyzing the load features within the disaster evolution feature stream, causal chain information and impact transmission path information can be extracted. For example, within the disaster evolution feature stream, data segments related to sensitive load response features can be found. From these segments, the causal chain information between load change events and device state change events, as well as the path information from load increase to device temperature via current parameters, can be extracted. This information can then be used as attributes of the third-category disaster feature entity.

[0102] Step S360: extracting a load regulation strategy from a preset strategy library, defining the type information and execution rule information of the load regulation strategy as a third type strategy entity; and establishing a third association relationship between the third type disaster feature entity and the third type strategy entity.

[0103] The load regulation strategy is a strategy used to deal with the impact of load changes on the distribution network. Its type information and execution rule information are defined as the third type of strategy entity so that it can be represented and associated in the knowledge graph.

[0104] To extract load regulation strategies, policies related to load changes can be filtered from the preset policy library. For example, dynamic load transfer strategies and hierarchical load regulation strategies can be selected. The type information (e.g., dynamic load transfer, hierarchical load regulation) and execution rule information (e.g., rules for filtering transfer load types, rules for classifying load regulation levels, etc.) of these policies are used as attributes of the third-type policy entity. Then, a third association relationship is established between the third-type disaster feature entity and the third-type policy entity. This association relationship represents the connection between the load change and the corresponding load regulation strategy. For example, when the direct causal relationship between the load change event and the device state change event is clear, a dynamic load transfer strategy may be necessary; when the path by which the load increase is transmitted to the device temperature through the current parameter is clear, a hierarchical load regulation strategy may be necessary.

[0105] Step S370: Combine the first type of disaster characteristic entity, the first type of strategy entity and the first association relationship, the second type of disaster characteristic entity, the second type of strategy entity and the second association relationship, and the third type of disaster characteristic entity, the third type of strategy entity and the third association relationship into a strategy generation knowledge graph.

[0106] The various disaster characteristic entities, strategy entities, and association relationships defined above are combined to form a complete strategy generation knowledge graph. In the knowledge graph, the first, second, and third disaster characteristic entities serve as nodes to represent different disaster characteristics. The first, second, and third strategy entities serve as nodes to represent different emergency strategies. The first, second, and third association relationships serve as edges to represent the connection between disaster characteristics and emergency strategies. In this way, the various characteristic information of distribution network disasters and the corresponding emergency strategies are organically integrated to form a visual and structured knowledge graph.

[0107] As an implementation method, the above-mentioned establishment of the first association relationship, the second association relationship, and the third association relationship can be specifically implemented as the following steps S301 to S303: Step S301: For the first association relationship: when the overlapping ratio of the timing synchronization information of the first type of disaster characteristic entity is greater than the first association threshold, an association is established between the first type of disaster characteristic entity and the equipment preventive maintenance strategy; when the covariance value of the dependency information of the change trend of the first type of disaster characteristic entity is greater than the second association threshold, an association is established between the first type of disaster characteristic entity and the equipment emergency isolation strategy; the equipment preventive maintenance strategy includes maintenance scope determination rules and maintenance time window selection rules, and the equipment emergency isolation strategy includes isolation operation priority rules and post-isolation system reconstruction principles.

[0108] The first association threshold and the second association threshold are pre-set thresholds for determining whether to establish an association relationship. The overlapping ratio of the time series synchronization information reflects the degree of synchronization between the equipment parameter change event and the environmental disturbance event in time. When the overlapping ratio is greater than the first association threshold, it indicates that there is a strong temporal correlation between the equipment state degradation and the environmental disturbance. At this time, an association is established between the first type of disaster characteristic entity and the equipment preventive maintenance strategy so that the equipment can be repaired in advance to prevent the occurrence of failures. The covariance value of the dependency information of the change trend reflects the correlation between the rising trend of the equipment parameters and the increasing trend of the environmental disturbance. When the covariance value is greater than the second association threshold, it indicates that there is a strong consistency in the change trend between the equipment state degradation and the environmental disturbance. At this time, an association is established between the first type of disaster characteristic entity and the equipment emergency isolation strategy to quickly isolate the faulty equipment and prevent the failure from expanding.

[0109] To establish these associations, we first need to determine the first and second association thresholds. Appropriate thresholds can be determined through statistical analysis based on historical data and experience. For example, an analysis of past distribution network disaster data revealed that when the overlap ratio between equipment parameter change events and environmental disturbance events is greater than 0.7, implementing a preventive maintenance strategy can effectively reduce equipment failure rates. Therefore, the first association threshold is set to 0.7. When the covariance between the rising trend of equipment parameters and the increasing trend of environmental disturbances is greater than 0.8, implementing an emergency isolation strategy can quickly block fault propagation. Therefore, the second association threshold is set to 0.8. Then, in the knowledge graph, when the overlap ratio of the temporal synchronization information of the first-class disaster characteristic entity is greater than 0.7, an edge is added between the first-class disaster characteristic entity and the equipment preventive maintenance strategy node to indicate the association between them. When the covariance value of the dependency information of the change trend of the first-class disaster characteristic entity is greater than 0.8, an edge is added between the first-class disaster characteristic entity and the equipment emergency isolation strategy node to indicate the association between them.

[0110] Step S302: For the second association relationship: when the intersection area of ​​the spatial overlapping information of the second type of disaster characteristic entity is greater than the third association threshold, an association is established between the second type of disaster characteristic entity and the regional joint protection strategy; when the linear fitting goodness value of the gradient corresponding information of the second type of disaster characteristic entity is greater than the fourth association threshold, an association is established between the second type of disaster characteristic entity and the gradient graded protection strategy; the regional joint protection strategy includes rules for the coordinated allocation of protection resources and rules for the simultaneous execution of protection measures, and the gradient graded protection strategy includes rules for adjusting the protection intensity as the gradient changes and rules for phased verification of the protection effect.

[0111] The third and fourth association thresholds are pre-set thresholds used to determine whether an association relationship is established. The intersection area of ​​the spatial overlap information reflects the degree of spatial overlap between the environmental disturbance area and the load fluctuation area. When the intersection area is greater than the third association threshold, it indicates that the environmental disturbance and the load fluctuation are strongly spatially correlated. At this time, an association is established between the second-type disaster characteristic entity and the regional joint protection strategy to resist environmental disturbances through resource coordination and synchronization of measures within the region. The linear fit goodness of fit value of the gradient corresponding information reflects the quality of the linear relationship between the wind speed attenuation gradient and the load growth gradient. When the linear fit goodness of fit value is greater than the fourth association threshold, it indicates that there is a strong linear relationship between the two. At this time, an association is established between the second-type disaster characteristic entity and the gradient graded protection strategy to adjust the protection intensity according to the gradient change.

[0112] To establish these associations, the third and fourth association thresholds must first be determined. Appropriate thresholds can be determined through statistical analysis based on historical data and experience. For example, an analysis of past distribution network disaster data revealed that when the intersection of the environmental disturbance area and the load fluctuation area is greater than 12 square kilometers, a regional joint protection strategy can effectively reduce the impact of environmental disturbances on the distribution network. Therefore, the third association threshold is set to 12 square kilometers. When the linear goodness of fit between the wind speed attenuation gradient and the load growth gradient is greater than 0.9, a gradient-graded protection strategy can improve the utilization of protection resources. Therefore, the fourth association threshold is set to 0.9. Then, in the knowledge graph, when the intersection of the spatial overlap information of the second-category disaster characteristic entity is greater than 12 square kilometers, an edge is added between the second-category disaster characteristic entity and the regional joint protection strategy node to indicate the association between them. When the linear goodness of fit of the gradient corresponding information of the second-category disaster characteristic entity is greater than 0.9, an edge is added between the second-category disaster characteristic entity and the gradient-graded protection strategy node to indicate the association between them.

[0113] Step S303: For the third association relationship: when the direct causal relationship clarity of the causal chain information of the third type of disaster characteristic entity is greater than the fifth association threshold, an association is established between the third type of disaster characteristic entity and the load dynamic transfer strategy; when the path clarity of the impact transmission path information of the third type of disaster characteristic entity is greater than the sixth association threshold, an association is established between the third type of disaster characteristic entity and the load hierarchical control strategy; the load dynamic transfer strategy includes transfer load type screening rules and transfer target line selection rules, and the load hierarchical control strategy includes load control level division rules and important load supply guarantee execution rules.

[0114] The fifth and sixth association thresholds are pre-set thresholds for determining whether to establish an association relationship. The direct causal relationship clarity of the causal chain information reflects the clarity of the direct causal relationship between the load change event and the device state change event. When the direct causal relationship clarity is greater than the fifth association threshold, it indicates that the load change has a direct impact on the device state. At this time, an association is established between the third-type disaster characteristic entity and the load dynamic transfer strategy to reduce the burden on the device by transferring the load. The path clarity of the impact transmission path information reflects the clarity of the path of the impact of the load change on the device state. When the path clarity is greater than the sixth association threshold, it indicates that the impact path of the load change on the device state is clear. At this time, an association is established between the third-type disaster characteristic entity and the load hierarchical control strategy to enable hierarchical control based on the importance of the load and the impact path.

[0115] To establish these associations, we first need to determine the fifth and sixth association thresholds. Appropriate thresholds can be determined through statistical analysis based on historical data and experience. For example, an analysis of past distribution network disaster data revealed that when the clarity of the direct causal relationship between a load change event and a device state change event is greater than 0.8, a dynamic load transfer strategy can effectively balance the load. Therefore, the fifth association threshold is set to 0.8. When the clarity of the path from the load increase through current parameters to device temperature is greater than 0.9, a load grading control strategy can better ensure the supply of critical loads. Therefore, the sixth association threshold is set to 0.9. Then, in the knowledge graph, when the clarity of the direct causal relationship of the causal chain information for the third-category disaster characteristic entity is greater than 0.8, an edge is added between the third-category disaster characteristic entity and the load dynamic transfer strategy node to indicate the association between them. When the clarity of the impact transmission path information for the third-category disaster characteristic entity is greater than 0.9, an edge is added between the third-category disaster characteristic entity and the load grading control strategy node to indicate the association between them.

[0116] As another embodiment, the process of establishing the first association relationship, the second association relationship, and the third association relationship may be specifically implemented as follows: Steps S300A to S300F: Step S300A: Obtain effect evaluation data of strategy implementation under historical disaster scenarios, the effect evaluation data including fault mitigation rate information of equipment intervention strategy, disturbance resistance rate information of environmental buffer strategy, and overload relief rate information of load regulation strategy.

[0117] The effectiveness evaluation data for strategy implementation in historical disaster scenarios records and evaluates the effectiveness of various emergency strategies implemented in past distribution network disasters. The fault mitigation rate information represents the effectiveness of the equipment intervention strategy in reducing the incidence of equipment failures, which can be obtained by calculating the reduction in the number of equipment failures after the equipment intervention strategy is implemented. The disturbance resistance rate information represents the effectiveness of the environmental buffer strategy in resisting the impact of environmental disturbances on the distribution network, which can be obtained by calculating the reduction in the loss caused by environmental disturbances to the distribution network after the environmental buffer strategy is implemented. The overload relief rate information represents the effectiveness of the load regulation strategy in relieving overloads in the distribution network, which can be obtained by calculating the reduction in the number of overloaded loads after the load regulation strategy is implemented.

[0118] To obtain these effectiveness evaluation data, relevant information can be extracted from a historical disaster database. Past distribution network disaster events are collated and analyzed, and the emergency response strategies implemented in each event, along with the corresponding effectiveness data, are recorded. Similarly, the disturbance mitigation rate of the environmental buffering strategy and the overload relief rate of the load regulation strategy are calculated.

[0119] Step S300B: Perform strategy effectiveness analysis on the effect evaluation data to obtain the strategy effectiveness coefficient, which includes the historical fault relief rate of the equipment preventive maintenance strategy, the historical fault blocking rate of the equipment emergency isolation strategy, the historical disturbance attenuation rate of the regional joint protection strategy, the historical protection resource utilization rate of the gradient graded protection strategy, the historical load balancing degree of the load dynamic transfer strategy, and the historical important load supply rate of the load graded control strategy.

[0120] Strategy effectiveness analysis involves an in-depth analysis of performance evaluation data to determine the effectiveness of various emergency response strategies. The strategy effectiveness coefficient measures each strategy's effectiveness in historical disaster scenarios. This includes the historical fault mitigation rate for the preventive maintenance strategy, the historical fault blocking rate for the emergency isolation strategy, the historical disturbance attenuation rate for the regional joint protection strategy, the historical protection resource utilization rate for the gradient-based protection strategy, the historical load balancing rate for the dynamic load transfer strategy, and the historical critical load supply rate for the load-level control strategy.

[0121] In order to analyze the effectiveness of the strategy, the effect evaluation data can be statistically analyzed. For the equipment preventive maintenance strategy, calculate its average fault mitigation rate in multiple historical disaster events as the historical fault mitigation rate. For the equipment emergency isolation strategy, calculate its proportion of successfully blocking fault propagation in multiple historical disaster events as the historical fault blocking rate. For the regional joint protection strategy, calculate its average proportion of reducing the impact of environmental disturbances in multiple historical disaster events as the historical disturbance attenuation rate. For the gradient-level protection strategy, calculate its average utilization rate of protection resources in multiple historical disaster events as the historical protection resource utilization rate. For the load dynamic transfer strategy, calculate its degree of load balancing in multiple historical disaster events as the historical load balancing degree. For the load-leveling control strategy, calculate its proportion of ensuring the supply of important loads in multiple historical disaster events as the historical important load supply rate.

[0122] Step S300C: When establishing the first association relationship, the policy effectiveness coefficients of the equipment preventive maintenance strategy and the equipment emergency isolation strategy are used as the first association strength parameter; when establishing the second association relationship, the policy effectiveness coefficients of the regional joint protection strategy and the gradient hierarchical protection strategy are used as the second association strength parameter; when establishing the third association relationship, the policy effectiveness coefficients of the load dynamic transfer strategy and the load hierarchical control strategy are used as the third association strength parameter.

[0123] The association strength parameter is used to measure the closeness of the association between the disaster feature entity and the strategy entity. Using the strategy effectiveness coefficients of the equipment preventive maintenance strategy and the equipment emergency isolation strategy as the first association strength parameter means that when establishing the first association relationship, the higher the strategy effectiveness, the greater the association strength. Similarly, using the strategy effectiveness coefficients of the regional joint protection strategy and the gradient-level protection strategy as the second association strength parameter, and using the strategy effectiveness coefficients of the load dynamic transfer strategy and the load-level control strategy as the third association strength parameter.

[0124] In the knowledge graph, when establishing the first association, edges are added between the first-category disaster characteristic entity and the equipment preventive maintenance strategy and equipment emergency isolation strategy, with the corresponding strategy effectiveness coefficients used as edge weights. Similarly, when establishing the second and third associations, the strategy effectiveness coefficients of the regional joint protection strategy, gradient-level protection strategy, load dynamic transfer strategy, and load-level regulation strategy, respectively, are used as edge weights.

[0125] Step S300D: The first correlation strength parameter, the second correlation strength parameter, and the third correlation strength parameter are respectively positively correlated with the effectiveness coefficients of the corresponding strategies.

[0126] This indicates a positive correlation between the association strength parameter and the effectiveness coefficient of the corresponding strategy: the higher the effectiveness coefficient of the strategy, the larger the association strength parameter. This positive correlation ensures that disaster feature entities are more closely associated with more effective strategies in the knowledge graph, thereby prioritizing more effective strategies during policy reasoning.

[0127] Step S300E: Setting the confidence threshold of the association relationship, the confidence threshold includes the minimum fault mitigation rate requirement associated with the device intervention strategy, the minimum disturbance resistance rate requirement associated with the environment buffer strategy, and the minimum overload relief rate requirement associated with the load regulation strategy.

[0128] The confidence threshold is used to determine whether an association is valid. The minimum fault mitigation rate requirement associated with a device intervention policy indicates that the association is valid only when the device intervention policy's fault mitigation rate reaches this threshold. The minimum disturbance resistance rate requirement associated with an environmental buffer policy indicates that the association is valid only when the environmental buffer policy's disturbance resistance rate reaches this threshold. The minimum overload relief rate requirement associated with a load regulation policy indicates that the association is valid only when the load regulation policy's overload relief rate reaches this threshold.

[0129] To set these confidence thresholds, they can be determined based on actual needs and experience.

[0130] Step S300F: A valid first association relationship is established only when the first association strength parameter is greater than the confidence threshold associated with the device intervention strategy; a valid second association relationship is established only when the second association strength parameter is greater than the confidence threshold associated with the environmental buffer strategy; a valid third association relationship is established only when the third association strength parameter is greater than the confidence threshold associated with the load regulation strategy; the valid association relationship is used to preferentially trigger the policy type that has been historically verified to be valid.

[0131] In the knowledge graph, an association is considered valid only when its association strength parameter exceeds the corresponding confidence threshold. For example, when the first association strength parameter (historical fault mitigation rate) of the equipment preventive maintenance strategy exceeds the minimum fault mitigation rate requirement (0.6) associated with the equipment intervention strategy, a valid first association is established between the first-type disaster characteristic entity and the equipment preventive maintenance strategy. Similarly, the second and third associations are judged according to the corresponding confidence threshold. Valid associations prioritize historically validated policy types during policy inference, thereby improving the effectiveness and reliability of emergency response strategies.

[0132] Step S400: Perform policy reasoning through the policy generation knowledge graph to generate an emergency policy candidate flow.

[0133] Policy reasoning utilizes the knowledge and relationships in the policy-generated knowledge graph to infer appropriate emergency response strategies based on current disaster characteristics. The emergency strategy candidate stream is a chronological list of possible emergency response strategies, encompassing multiple candidate strategies tailored to the current disaster situation.

[0134] As an implementation method, step S400 performs policy reasoning through the policy generation knowledge graph to generate an emergency policy candidate flow, which can be specifically implemented as follows: steps S410 to S440: Step S410: Extract the device state degradation associated feature instance at the current moment from the disaster evolution feature stream, query the disaster feature entity that matches the device state degradation associated feature instance in the strategy generation knowledge graph, obtain the device intervention strategy entity associated with the disaster feature entity, and generate the device intervention strategy unit.

[0135] The device state degradation feature instance is the specific feature information in the disaster evolution feature stream that reflects the degradation of the device state due to environmental disturbances at the current moment. Searching for matching disaster feature entities in the strategy generation knowledge graph involves finding the first-class disaster feature entity that is most similar to the current device state degradation feature instance. Obtaining the device intervention strategy entity associated with this disaster feature entity involves finding the equipment preventive maintenance strategy or equipment emergency isolation strategy, etc., that is connected to this disaster feature entity through the first association relationship.

[0136] In order to extract the current device state degradation-related feature instances, the disaster evolution feature stream can be monitored in real time. At the current moment, data related to the device state degradation-related features are extracted from the disaster evolution feature stream, such as the overlap ratio between device parameter change events and environmental disturbance events, and the covariance between the device parameter rising trend and the environmental disturbance strengthening trend. Then, in the strategy generation knowledge graph, these data are compared with the attributes of the first-class disaster feature entity to find the most similar disaster feature entity. For example, by calculating the similarity between the device state degradation-related feature instance and each first-class disaster feature entity in the knowledge graph, the entity with the highest similarity is selected as the matching entity. The device intervention strategy entity associated with the matching entity is obtained, and its related information is organized into an equipment intervention strategy unit.

[0137] As an implementation method, in step S410, a disaster feature entity matching an instance of a feature associated with equipment state degradation is searched in the strategy generation knowledge graph, an equipment intervention strategy entity associated with the disaster feature entity is obtained, and an equipment intervention strategy unit is generated. Specifically, the steps S411 to S415 are as follows: Step S411: performing similarity calculation on the temporal synchronization information of the equipment state degradation associated feature instance and the temporal synchronization information of the disaster feature entity in the strategy generation knowledge graph to obtain a first similarity value.

[0138] Similarity calculation compares the temporal synchronization information of the device state degradation-related feature instances with the temporal synchronization information of the disaster feature entities in the knowledge graph. The first similarity value is a metric that measures this similarity and can be calculated using various methods, such as Euclidean distance and cosine similarity.

[0139] To calculate similarity, we first preprocess the temporal synchronization information of the device state degradation-related feature instances (such as the overlap ratio between device parameter change events and environmental disturbance events) and the temporal synchronization information of the disaster feature entities in the knowledge graph to ensure they have the same format and scope. We then calculate the distance between the two using the Euclidean distance formula and convert the distance into a similarity value.

[0140] Step S412: performing similarity calculation on the dependency information of the change trend of the equipment state degradation associated feature instance and the dependency information of the change trend of the disaster feature entity in the strategy generation knowledge graph to obtain a second similarity value.

[0141] Similar to step S411, a similarity calculation is performed between the dependency information of the changing trends of the device state degradation-related feature instances (e.g., the covariance between the rising trend of device parameters and the increasing trend of environmental disturbances) and the dependency information of the changing trends of the disaster feature entities in the knowledge graph to obtain a second similarity value. Similarly, the data is preprocessed to ensure that they have the same format and scope. Then, a suitable similarity calculation method, such as cosine similarity, is used to calculate the similarity between the two.

[0142] Step S413: Select the disaster feature entity with the largest sum of the first similarity value and the second similarity value as the matching entity.

[0143] The first similarity value and the second similarity value are added together to obtain a comprehensive similarity score between each disaster feature entity and the equipment state degradation associated feature instance. The disaster feature entity with the largest comprehensive similarity score is selected as the matching entity because it is the most similar to the equipment state degradation associated feature instance.

[0144] Step S414: Acquire the equipment intervention policy entity associated with the matching entity, where the equipment intervention policy entity includes an equipment preventive maintenance policy and an equipment emergency isolation policy.

[0145] In the policy generation knowledge graph, find the device intervention policy entities connected to the matching entity through the first association relationship. These entities may be device preventive maintenance policies or device emergency isolation policies. Obtain their related information, such as type information and operation process information.

[0146] Step S415: According to the severity information of the device state degradation associated feature instance, the policy with the highest priority is selected from the device intervention policy entity as the device intervention policy unit. The severity information includes the urgency and impact range of the device parameter change.

[0147] Severity information is used to determine the severity of equipment degradation, including the urgency of equipment parameter changes and the scope of impact. The urgency of equipment parameter changes can be measured by the speed and magnitude of the change, while the scope of impact can be determined by the number of devices affected and the area.

[0148] Device intervention policy entities are filtered based on severity information. If the device state degradation is urgent and has a wide impact, the device emergency isolation policy may be selected as the highest priority policy. If the device state degradation is relatively mild, the device preventive maintenance policy may be selected. For example, if device parameters change rapidly and a large number of devices are affected, the device state degradation is considered severe, and the device emergency isolation policy is selected as the device intervention policy unit.

[0149] Step S420: Extract the environmental disturbance coupling feature instance at the current moment from the disaster evolution feature stream, query the disaster feature entity that matches the environmental disturbance coupling feature instance in the strategy generation knowledge graph, obtain the environmental buffer strategy entity associated with the disaster feature entity, and generate an environmental buffer strategy unit.

[0150] The environmental disturbance coupling feature instance is the specific feature information in the disaster evolution feature stream that reflects the spatial correlation and gradient correspondence between environmental disturbances and load fluctuations at the current moment. Querying the strategy generation knowledge graph for matching disaster feature entities involves finding the second-category disaster feature entity that is most similar to the current environmental disturbance coupling feature instance. Obtaining the environmental buffer strategy entity associated with this disaster feature entity involves finding the regional joint protection strategy or gradient-based protection strategy, etc., that is connected to this disaster feature entity through the second-category association relationship.

[0151] To extract instances of environmental disturbance coupling characteristics at the current moment, the disaster evolution characteristic stream can be monitored in real time. At the current moment, data related to the environmental disturbance coupling characteristics is extracted from the disaster evolution characteristic stream, such as the intersection area of ​​the environmental disturbance region and the load fluctuation region, and the linear goodness of fit between the wind speed attenuation gradient and the load growth gradient. Then, in the strategy generation knowledge graph, this data is compared with the attributes of the second-category disaster characteristic entity to identify the most similar disaster characteristic entity. The environmental buffer strategy entity associated with this disaster characteristic entity is obtained, and its related information is organized into environmental buffer strategy units.

[0152] Step S430: Extract the load response sensitive feature instance at the current moment from the disaster evolution feature stream, query the disaster feature entity that matches the load response sensitive feature instance in the strategy generation knowledge graph, obtain the load regulation strategy entity associated with the disaster feature entity, and generate a load regulation strategy unit.

[0153] A load response sensitive feature instance is the specific characteristic information in the disaster evolution feature stream that reflects the impact of load changes on equipment status at the current moment. Searching for matching disaster feature entities in the strategy generation knowledge graph involves finding the third-category disaster feature entity that is most similar to the current load response sensitive feature instance. Obtaining the load regulation policy entity associated with this disaster feature entity involves finding a load dynamic transfer strategy or load tiered regulation strategy, such as one connected to the disaster feature entity through a third-category relationship.

[0154] To extract the load response sensitive feature instances at the current moment, the disaster evolution feature stream can be monitored in real time. At the current moment, data related to the load response sensitive features are extracted from the disaster evolution feature stream, such as the causal chain information of load change events and device state change events, and the path information of the load increase transmitted to the device temperature through current parameters, etc. Then, in the policy generation knowledge graph, these data are compared with the attributes of the third type of disaster feature entities to find the most similar disaster feature entities. The load regulation policy entities associated with the disaster feature entities are obtained, and their relevant information is sorted into load regulation policy units.

[0155] Step S440: Arrange the device intervention policy units, environmental buffer policy units, and load regulation policy units in chronological order of timestamps to generate a candidate emergency policy stream.

[0156] Arrange the generated device intervention policy units, environmental buffer policy units, and load regulation policy units in chronological order of timestamps to form a continuous candidate emergency policy stream. The chronological order of timestamps reflects the execution sequence of these policies during the disaster evolution process. By arranging them in chronological order of timestamps, it can ensure that the implementation of the emergency policies has a reasonable timing sequence, thus more effectively coping with the distribution network disasters.

[0157] To achieve the arrangement in chronological order of timestamps, first, each policy unit needs to be marked with its corresponding timestamp. For the device intervention policy units, environmental buffer policy units, and load regulation policy units, according to the time information of the disaster features they target in the disaster evolution feature stream, the corresponding timestamps are assigned to them. For example, when the device state degradation associated feature instance at the current moment is extracted from the disaster evolution feature stream and a device intervention policy unit is generated, this moment is used as the timestamp of the device intervention policy unit.

[0158] Then, sorting algorithms are used to sort these policy units. Simple bubble sort, quick sort, etc. can be adopted to sort the device intervention policy units, environmental buffer policy units, and load regulation policy units in ascending order according to the chronological order of timestamps. Suppose the timestamp of the device intervention policy unit is t1, the timestamp of the environmental buffer policy unit is t2, and the timestamp of the load regulation policy unit is t3, and t1 < t2 < t3, then the order of the candidate emergency policy stream after sorting is the device intervention policy unit, the environmental buffer policy unit, and the load regulation policy unit.

[0159] In this way, the generated candidate emergency policy stream can clearly show the emergency policies to be taken at different time points, providing clear guidance for subsequent emergency handling.

[0160] Step S500: Conduct a coordination verification on the candidate emergency policy stream to generate the distribution network disaster emergency policy.

[0161] Collaboration verification checks whether the various policy units in the candidate emergency strategy flow can cooperate and work together to avoid resource conflicts, time conflicts, or offsetting effects. This collaboration verification ensures that the resulting distribution network disaster emergency strategy is efficient and reliable, enabling better response to distribution network disasters.

[0162] As an implementation method, step S500 performs collaborative verification on the candidate emergency strategy flows to generate a distribution network disaster emergency strategy, which can be specifically implemented as follows: steps S510 to S540: Step S510: Perform resource occupancy collaborative verification on the device intervention strategy unit and the environment buffer strategy unit in the emergency strategy candidate flow to obtain a resource occupancy collaborative verification result. The resource occupancy collaborative verification result includes overlapping conflict information and complementary utilization information of human and material resources required for strategy implementation.

[0163] Resource occupancy collaboration verification analyzes the human and material resource demands of the equipment intervention strategy unit and the environmental buffer strategy unit during implementation to determine whether there is a resource conflict or whether complementary resource utilization is possible. Overlapping conflict information for human and material resources indicates whether the two strategy units will compete for the same resources at the same time, such as whether they require the same team of maintenance personnel or the same batch of protective materials. Complementary utilization information indicates whether the two strategy units can share certain resources or achieve better results through the rational allocation of resources. For example, the transport vehicles required for the equipment intervention strategy can be used to allocate materials for the environmental buffer strategy after completing equipment transportation.

[0164] To conduct resource utilization collaborative verification, we first identify the required human and material resources for the equipment intervention strategy unit and the environmental buffer strategy unit. For the equipment intervention strategy unit, its human resources requirements may include maintenance personnel and technicians, while its material requirements may include repair tools and spare equipment. For the environmental buffer strategy unit, its human resources requirements may include protection personnel and monitoring personnel, while its material requirements may include protective materials and monitoring equipment. Next, we compare the two lists to analyze resource overlap and potential complementarity. For example, analysis of the lists reveals that both the equipment intervention strategy unit and the environmental buffer strategy unit require cranes for material lifting, resulting in a resource overlap conflict. However, the remaining protective materials in the equipment intervention strategy unit can be used for protective work in the environmental buffer strategy unit, demonstrating resource complementarity. Based on this analysis, we obtain the results of resource utilization collaborative verification.

[0165] Step S520: Perform time execution collaborative verification on the environment buffer strategy unit and the load adjustment strategy unit in the emergency strategy candidate flow to obtain a time execution collaborative verification result. The time execution collaborative verification result includes overlapping conflict information of the strategy start time and duration and timing connection information.

[0166] Time execution coordination verification checks the rationality of the timing of the environmental buffering strategy unit and the load regulation strategy unit, whether there are any time conflicts, or whether good timing coherence can be achieved. Overlapping conflict information for strategy startup times and durations indicates whether the startup times and durations of the two strategy units will interfere with each other. For example, whether the environmental buffering strategy unit and the load regulation strategy unit will be started at the same time, causing confusion in resource allocation or mutual impact on the effectiveness. Timing coherence information indicates whether the two strategy units can be executed sequentially in an orderly manner. For example, whether the load regulation strategy can be started after the environmental buffer strategy has established certain environmental conditions to achieve better results.

[0167] To perform temporal execution co-verification, it is necessary to obtain the startup time and duration information of the environmental buffering strategy unit and the load regulation strategy unit. This information can be estimated based on the nature of the strategies and the actual situation. For example, the environmental buffering strategy unit may need to be activated immediately after a disaster occurs, lasting several hours; the load regulation strategy unit may need to be activated after the environmental buffering strategy has achieved a certain effect, lasting several days. Then, the schedules of the two strategy units are compared to analyze whether there is a time conflict or the possibility of temporal connection. For example, if the duration of the environmental buffering strategy unit overlaps the startup time of the load regulation strategy unit, there is a time conflict. However, if the load regulation strategy unit can be activated immediately after the environmental buffering strategy unit ends, and the effects of the two can complement each other, then a good temporal connection is achieved. Based on the analysis results, the temporal execution co-verification results are obtained.

[0168] Step S530: Perform effect target collaborative verification on the load regulation strategy unit and the equipment intervention strategy unit in the emergency strategy candidate flow to obtain an effect target collaborative verification result. The effect target collaborative verification result includes the target consistency information of the strategy for disaster relief and the effect superposition or offset information.

[0169] Effect target collaborative verification is to evaluate whether the effect targets of the load regulation strategy unit and the equipment intervention strategy unit are consistent when responding to distribution network disasters, and whether their implementation effects will be superimposed or offset. The target consistency information of the strategies for disaster mitigation indicates whether the two strategy units are working towards the same direction of alleviating distribution network disasters, such as whether they both aim to reduce the incidence of equipment failures, reduce load overloads, etc. The effect superposition or offset information indicates whether the implementation effects of the two strategy units can enhance or weaken each other. For example, the load regulation strategy reduces the burden on the equipment by shifting the load, while the equipment intervention strategy performs preventive maintenance on the equipment. The effects of the two are superimposed on each other and can better mitigate disasters. Conversely, if the implementation of the load regulation strategy causes the operating status of the equipment to change, and the equipment intervention strategy does not take this change into account, the effects of the two may offset each other.

[0170] To perform collaborative validation of effectiveness objectives, it's necessary to clearly define the effectiveness objectives of the load regulation strategy unit and the device intervention strategy unit. The effectiveness objectives of the load regulation strategy unit might be achieving load balancing or relieving overloads, while the effectiveness objectives of the device intervention strategy unit might be preventing equipment failures or repairing faulty equipment. Next, analyze whether the effectiveness objectives of the two strategy units are consistent, and the relationship between their implementation effects. Simulation experiments and historical data statistics can be used to assess the potential for additive or offsetting effects. For example, an analysis of historical data revealed that, in some cases, the combined implementation of the load regulation strategy and the device intervention strategy can significantly reduce equipment failure rates, indicating that the two effects are additive. In other cases, improper implementation of the load regulation strategy can impact the effectiveness of the device intervention strategy, indicating that the two effects are offsetting. Based on these analysis results, the results of the collaborative validation of effectiveness objectives are obtained.

[0171] Step S540: Based on the resource occupancy collaborative verification results, the time execution collaborative verification results and the effect target collaborative verification results, the emergency strategy candidate flows are conflict resolved and collaboratively optimized to generate a conflict-free and collaboratively enhanced distribution network disaster emergency strategy.

[0172] Conflict resolution and collaborative optimization are to adjust and optimize the candidate flows of emergency strategies based on the collaborative verification results of resource occupancy, time execution and effect target, eliminate conflicts between strategy units and enhance the collaborative effects between strategies.

[0173] Resource reallocation is used to resolve overlapping conflicts identified during collaborative resource occupancy verification. For example, if both the equipment intervention strategy unit and the environmental buffering strategy unit require the use of a crane, crane usage schedules and task allocations are rationally arranged based on strategy priorities and actual needs to ensure efficient resource utilization. This complementary resource utilization information can be used to further optimize resource allocation, enabling shared and collaborative use, and improving resource utilization efficiency.

[0174] For timing conflicts discovered during the time-execution co-verification, adjust the strategy's startup time and duration. For example, if there's a timing conflict between the environmental buffering strategy unit and the load regulation strategy unit, delay the startup of one strategy or shorten its duration to avoid the conflict. For timing cohesion information, optimize the execution order of the strategies based on their interrelationships to ensure they execute sequentially and achieve good timing cohesion.

[0175] If the results of the collaborative verification of effect targets reveal offsetting effects, the strategies will be adjusted and improved. For example, if the effects of the load regulation strategy and the device intervention strategy are offsetting, the cause will be analyzed and the strategy parameters or implementation methods will be adjusted to enhance their effects. Furthermore, based on the consistency information of the effect targets, the common goals of the strategies will be further clarified, and the combination of strategies will be optimized to improve the overall effectiveness of the strategies.

[0176] Through the above conflict resolution and collaborative optimization process, the candidate emergency strategy flows are comprehensively adjusted and optimized, ultimately generating a conflict-free and collaboratively enhanced distribution network disaster emergency strategy. This strategy can fully leverage the advantages of each strategy unit, coordinately respond to distribution network disasters, and improve the reliability and stability of the distribution network.

[0177] It is understandable that the various algorithms involved in the above-mentioned introductions of the embodiments of the present invention, such as the Euclidean distance algorithm, the cosine distance algorithm, the causal analysis algorithm, etc., can all be learned from the relevant content in the prior art. In order to save space, they will not be expanded too much in the embodiments of the present invention. In addition, when implementing the scheme of the present invention, those skilled in the art can supplement the details according to the common knowledge in this field. For example, according to the common knowledge in this field, normalization can be used to eliminate dimensional conflicts before feature fusion, interpolation can be used to eliminate dimensional differences, and thresholds can be reasonably set based on historical data, experience or business scenario requirements. The model can be trained based on a general model training method, and the number of layers in the model structure can be set based on actual needs, the activation function can be selected, etc. The present invention will no longer provide redundant introductions to the overly detailed implementation process.

[0178] See also Figure 2 , Figure 2This is a schematic diagram of the structure of a computer system provided in an embodiment of the present invention. The computer system includes at least a processor 101, a communication interface 102, and a memory 103. The processor 101, communication interface 102, and memory 103 may be connected via a bus or other means. The processor 101 (also known as the Central Processing Unit (CPU)) is the computing and control core of the computer system, capable of parsing various instructions within the computer system and processing various data within the computer system. The communication interface 102 may optionally include a standard wired interface or a wireless interface (such as Wi-Fi, a mobile communication interface, etc.), which can be used to send and receive data under the control of the processor 101. The communication interface 102 may also be used for data transmission and interaction within the computer system. The memory 103 is a storage device in the computer system for storing programs and data. It is understood that the memory 103 herein may include both the built-in memory of the computer system and, of course, the extended memory supported by the computer system. The memory 103 provides storage space, which stores the computer system's operating system, but this is not limited to this in the present invention.

[0179] In one embodiment, the processor 101 executes the emergency strategy generation method for power distribution network disasters provided in the above embodiment of the present invention by running the computer program in the memory 103 .

Claims

1. A method for generating an emergency strategy for a distribution network disaster, characterized in that: The method comprises: Acquire multimodal perception data of the distribution network operating environment, wherein the multimodal perception data includes a device state perception sequence, an environmental disturbance perception sequence, and a load response perception sequence; Performing disaster evolution feature modeling on the multimodal sensing data to generate a disaster evolution feature stream; Generating a knowledge graph based on the disaster evolution characteristic flow construction strategy; Generate a knowledge graph through the strategy to perform strategy reasoning and generate a candidate flow of emergency strategies; The candidate emergency strategy flows are collaboratively verified to generate a distribution network disaster emergency strategy.

2. The method according to claim 1, wherein The performing disaster evolution feature modeling on the multimodal sensing data to generate a disaster evolution feature stream includes: Performing a time series correlation analysis on the device state sensing sequence and the environmental disturbance sensing sequence to generate a device state degradation correlation feature, wherein the device state degradation correlation feature includes time series synchronization information and change trend dependency information of device parameter changes and environmental disturbances; Performing spatial coupling analysis on the environmental disturbance perception sequence and the load response perception sequence to generate an environmental disturbance coupling feature, wherein the environmental disturbance coupling feature includes spatial overlap information of the environmental disturbance area and the load fluctuation area and gradient correspondence information of the disturbance intensity and the load fluctuation amplitude; Performing causal dependency analysis on the load response perception sequence and the device state perception sequence to generate a load response sensitive feature, wherein the load response sensitive feature includes causal chain information of a load change event and a device state change event and transmission path information of the impact of the load change amount on the device state; The equipment state degradation correlation feature, the environmental disturbance coupling feature and the load response sensitivity feature are fused in a time series manner to generate the disaster evolution feature flow.

3. The method according to claim 2, wherein The performing time series correlation analysis on the device state sensing sequence and the environmental disturbance sensing sequence to generate a device state degradation correlation feature includes: Extracting device parameter change event points in the device state perception sequence, and extracting environmental disturbance event points in the environmental disturbance perception sequence; Performing time stamp alignment on the device parameter change event point and the environmental disturbance event point to obtain time series synchronization information, wherein the time series synchronization information includes information on the overlap ratio of the device parameter change event and the environmental disturbance event on the time axis; Performing a dependency analysis on the change trend of the device parameter change event point and the disturbance trend of the environmental disturbance event point to obtain dependency information of the change trend, wherein the dependency information of the change trend includes covariance information of the device parameter rising trend and the environmental disturbance increasing trend; The time series synchronization information and the dependency information of the change trend are combined into the device state degradation correlation feature.

4. The method according to claim 2, wherein The performing spatial coupling analysis on the environmental disturbance perception sequence and the load response perception sequence to generate an environmental disturbance coupling feature includes: Extracting disturbance area distribution information from the environmental disturbance perception sequence, wherein the disturbance area distribution information includes geographical location information of high wind speed areas, coverage information of heavy rainfall areas, and boundary coordinate information of high-frequency thunderstorm areas; Extracting load fluctuation area distribution information from the load response sensing sequence, wherein the load fluctuation area distribution information includes geographical location information of a high load growth area, coverage information of a load drop area, and boundary coordinate information of a load oscillation area; Performing spatial overlap calculation on the disturbance region distribution information and the load fluctuation region distribution information to obtain spatial overlap information, wherein the spatial overlap information includes intersection area information of the high wind speed region and the high load growth region; Extracting disturbance intensity gradient information from the environmental disturbance perception sequence, wherein the disturbance intensity gradient information includes a wind speed attenuation gradient, a precipitation decrease gradient, and a thunderstorm frequency decrease gradient from the disturbance center to the boundary; Extracting load fluctuation amplitude gradient information from the load response perception sequence, wherein the load fluctuation amplitude gradient information includes a decreasing gradient of a load increase and an increasing gradient of a load decrease from a load center to a load boundary; Performing gradient correspondence analysis on the disturbance intensity gradient information and the load fluctuation amplitude gradient information to obtain gradient correspondence information, wherein the gradient correspondence information includes linear fitting relationship information of the wind speed attenuation gradient and the load growth amount decreasing gradient; The spatial overlap information and the gradient correspondence information are combined into the environmental disturbance coupling feature.

5. The method according to claim 2, wherein The performing causal dependency analysis on the load response perception sequence and the device state perception sequence to generate a load response sensitive feature includes: Extracting a load change event chain from the load response perception sequence, wherein the load change event chain includes a load increase trigger event, a load stability maintenance event, and a load reduction termination event; Extracting a device state change event chain from the device state perception sequence, wherein the device state change event chain includes a device normal operation event, a device abnormality warning event, and a device failure shutdown event; Performing causal relationship mining on the load change event chain and the device state change event chain to obtain causal chain information, wherein the causal chain information includes direct causal relationship information from the load increase trigger event to the device abnormality warning event and indirect causal relationship information from the load stability maintenance event to the device normal operation event; Extracting load variation information from the load response perception sequence, wherein the load variation information includes absolute numerical value information of load increase, absolute numerical value information of load decrease, and relative proportion information of load change; Extracting device state impact information from the device state sensing sequence, wherein the device state impact information includes information on the degree of impact of load change on device voltage, device current, and device temperature; Performing a transfer path analysis on the load change information and the device state impact information to obtain impact transfer path information, wherein the impact transfer path information includes path information of how the load increase is transferred to the device temperature via the current parameter; The causal chain information and the impact transfer path information are combined into the load response sensitive feature.

6. The method according to claim 1, wherein The generating of the knowledge graph based on the disaster evolution characteristic flow construction strategy includes: Extracting equipment state degradation correlation features from the disaster evolution feature stream, and defining the time series synchronization information and the dependency information of the change trend in the equipment state degradation correlation features as a first type of disaster feature entity; Extracting an equipment intervention strategy from a preset strategy library, defining the type information and operation process information of the equipment intervention strategy as a first-type strategy entity; establishing a first association relationship between the first-type disaster feature entity and the first-type strategy entity; Extracting environmental disturbance coupling features from the disaster evolution characteristic flow, and defining spatial overlap information and gradient correspondence information in the environmental disturbance coupling features as a second type of disaster characteristic entity; Extracting an environmental buffer strategy from a preset strategy library, defining the type information and implementation requirement information of the environmental buffer strategy as a second-type strategy entity; establishing a second association relationship between the second-type disaster feature entity and the second-type strategy entity; Extracting load response sensitive features from the disaster evolution feature flow, and defining the causal chain information and impact transmission path information in the load response sensitive features as a third type of disaster feature entity; Extracting a load regulation strategy from a preset strategy library, defining the type information and execution rule information of the load regulation strategy as a third type strategy entity; establishing a third association relationship between the third type disaster feature entity and the third type strategy entity; The first type of disaster characteristic entity, the first type of strategy entity and the first association relationship, the second type of disaster characteristic entity, the second type of strategy entity and the second association relationship, the third type of disaster characteristic entity, the third type of strategy entity and the third association relationship are combined to generate a knowledge graph for the strategy.

7. The method according to claim 6, wherein The first association relationship, the second association relationship, and the third association relationship are established in the following manners: Regarding the first association relationship: when the overlapping ratio of the time series synchronization information of the first type of disaster characteristic entity is greater than the first association threshold, an association is established between the first type of disaster characteristic entity and the equipment preventive maintenance strategy; when the covariance value of the dependency information of the change trend of the first type of disaster characteristic entity is greater than the second association threshold, an association is established between the first type of disaster characteristic entity and the equipment emergency isolation strategy; the equipment preventive maintenance strategy includes a maintenance scope determination rule and a maintenance time window selection rule, and the equipment emergency isolation strategy includes an isolation operation priority rule and a post-isolation system reconstruction principle; For the second association relationship: when the intersection area of ​​the spatial overlapping information of the second type of disaster characteristic entity is greater than the third association threshold, an association is established between the second type of disaster characteristic entity and the regional joint protection strategy; when the linear fit goodness value of the gradient corresponding information of the second type of disaster characteristic entity is greater than the fourth association threshold, an association is established between the second type of disaster characteristic entity and the gradient graded protection strategy; the regional joint protection strategy includes a collaborative allocation rule for protection resources and a synchronous execution rule for protection measures, and the gradient graded protection strategy includes an adjustment rule for protection intensity changing with gradient and a staged verification rule for protection effect; Regarding the third association relationship: when the direct causal relationship clarity of the causal chain information of the third type of disaster characteristic entity is greater than the fifth association threshold, an association is established between the third type of disaster characteristic entity and the load dynamic transfer strategy; when the path clarity of the impact transmission path information of the third type of disaster characteristic entity is greater than the sixth association threshold, an association is established between the third type of disaster characteristic entity and the load grading control strategy; the load dynamic transfer strategy includes transfer load type screening rules and transfer target line selection rules, and the load grading control strategy includes load control level division rules and important load supply guarantee execution rules.

8. The method according to claim 1, wherein The step of generating a knowledge graph through the strategy to perform strategy reasoning and generate an emergency strategy candidate flow includes: Extracting a device state degradation-related feature instance at the current moment from the disaster evolution feature stream, searching the strategy generation knowledge graph for a disaster feature entity that matches the device state degradation-related feature instance, obtaining a device intervention strategy entity associated with the disaster feature entity, and generating a device intervention strategy unit; Extracting the environmental disturbance coupling feature instance at the current moment from the disaster evolution feature stream, querying the disaster feature entity matching the environmental disturbance coupling feature instance in the strategy generation knowledge graph, obtaining the environmental buffer strategy entity associated with the disaster feature entity, and generating an environmental buffer strategy unit; Extracting a load response sensitive feature instance at the current moment from the disaster evolution feature stream, querying a disaster feature entity matching the load response sensitive feature instance in the strategy generation knowledge graph, obtaining a load regulation strategy entity associated with the disaster feature entity, and generating a load regulation strategy unit; The device intervention strategy unit, the environment buffer strategy unit, and the load adjustment strategy unit are arranged in a timestamp order to generate an emergency strategy candidate flow.

9. The method according to claim 8, wherein The step of searching the strategy generation knowledge graph for a disaster feature entity that matches the device state degradation associated feature instance, obtaining the device intervention strategy entity associated with the disaster feature entity, and generating a device intervention strategy unit includes: Performing similarity calculation on the time series synchronization information of the device state degradation associated feature instance and the time series synchronization information of the disaster feature entity in the strategy generation knowledge graph to obtain a first similarity value; performing similarity calculation on the dependency information of the change trend of the equipment state degradation associated feature instance and the dependency information of the change trend of the disaster feature entity in the strategy generation knowledge graph to obtain a second similarity value; Selecting the disaster feature entity with the largest sum of the first similarity value and the second similarity value as the matching entity; Acquire an equipment intervention policy entity associated with the matching entity, wherein the equipment intervention policy entity includes an equipment preventive maintenance policy and an equipment emergency isolation policy; According to the severity information of the device state degradation associated feature instance, the policy with the highest priority is screened out from the device intervention policy entity as the device intervention policy unit, wherein the severity information includes the urgency and impact range information of the device parameter change.

10. A computer system, characterized in that: include: a memory storing a computer program; A processor is used to load the computer program to implement the emergency strategy generation method applied to distribution network disasters as described in any one of claims 1 to 9.

Images