Power distribution network risk management method, system, equipment and medium in extreme weather

Abstract

The invention discloses a power distribution network risk management method, system and device in extreme weather and a medium, and the method comprises the following steps: obtaining meteorological data through detecting weather conditions, and predicting a weather development trend; analyzing the fault risk of each node of the power distribution network according to the meteorological data and the weather development trend; and generating a decision scheme according to the analysis result of the fault risk, and coordinately executing power dispatching and emergency disposal by combining the decision scheme with an emergency communication network for power supply guarantee. According to the invention, the comprehensive management method integrating meteorological monitoring, risk assessment, emergency communication, power dispatching and power supply guarantee is constructed, so that the risk management capability of the power distribution network under the extreme weather condition is remarkably improved; and the precision and timeliness of weather prediction are improved.

Description

Technical Field

[0001] This invention relates to the field of distribution network risk management technology, and in particular to a method, system, equipment and medium for distribution network risk management under extreme weather conditions. Background Technology

[0002] With the intensification of global climate change, the destructive impact of extreme weather events such as typhoons, torrential rains, and icing on power distribution networks is becoming increasingly severe. Traditional risk management methods for power distribution networks are no longer sufficient to meet the safe and reliable operation requirements of modern power systems. Existing technologies mainly suffer from the following problems: meteorological monitoring methods are relatively limited, relying heavily on single data sources for weather forecasting and lacking the ability to fuse multi-source data, resulting in insufficient forecast accuracy and difficulty in accurately grasping the development trend of extreme weather; risk assessment methods are mainly based on static threshold judgments and empirical formulas, lacking dynamic analysis of the vulnerability of the power distribution network topology and failing to effectively predict the risk of cascading failures; power dispatching relies primarily on manual experience for decision-making, resulting in slow response times, difficulty in achieving minute-level rapid responses, and a lack of intelligent resource optimization and allocation capabilities. These technological limitations severely restrict the effectiveness of emergency management of power distribution networks under extreme weather conditions, urgently requiring a more intelligent and automated risk management method. Summary of the Invention

[0003] In view of the aforementioned existing problems, the present invention is proposed.

[0004] Therefore, this invention provides a method, system, equipment, and medium for risk management of distribution networks under extreme weather conditions to address the problems of existing methods lacking multi-source data fusion capabilities, resulting in insufficient prediction accuracy, lacking dynamic analysis of the vulnerability of distribution network topology, and being unable to effectively predict the risk of cascading failures.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] In a first aspect, the present invention provides a method for risk management of power distribution networks under extreme weather conditions, comprising the following steps:

[0007] Meteorological data is obtained by monitoring weather conditions, and weather trends are predicted.

[0008] Based on the meteorological data and weather trends, analyze the fault risks of each node in the power distribution network;

[0009] Based on the analysis of fault risks, a decision-making plan is generated. This plan, combined with the emergency communication network, coordinates the execution of power dispatch and emergency response for power supply assurance.

[0010] As a preferred embodiment of the distribution network risk management method under extreme weather conditions described in this invention, the step of acquiring the meteorological data includes:

[0011] It integrates meteorological satellite, radar and ground sensor data to collect multi-source meteorological data in real time;

[0012] Spatiotemporal alignment and error correction are performed on multi-source meteorological data to generate a typhoon track prediction model with a preset spatial resolution.

[0013] The typhoon path prediction model is deployed on the edge terminals of key nodes, and the monitoring data is uploaded to the monitoring platform through wireless communication protocols.

[0014] The beneficial effects of this preferred technical solution are as follows: by using multi-source meteorological data fusion and spatiotemporal alignment technology, the accuracy and reliability of typhoon path prediction are improved, providing a high-quality data foundation for subsequent risk assessment and effectively solving the problem of large prediction errors from traditional single data sources.

[0015] As a preferred embodiment of the distribution network risk management method under extreme weather conditions described in this invention, the step of analyzing the fault risk of each node in the distribution network includes:

[0016] Feature extraction is performed on historical fault data and meteorological data to predict the fault probability of each distribution network node within a preset time window;

[0017] By analyzing the network topology of vulnerable nodes in the power grid, and combining the failure probability with the equipment tolerance threshold, direct risks and potential risks are calculated to generate risk level assessment results.

[0018] When the meteorological data exceeds a preset threshold, a dynamic early warning is triggered based on the risk level assessment result.

[0019] The beneficial effects of this preferred technical solution are as follows: by using deep learning algorithms combined with network topology analysis technology, it achieves accurate prediction of the probability of faults in distribution network nodes and quantitative assessment of direct and potential risks. Compared with the traditional static threshold judgment method, it improves the accuracy and foresight of risk identification.

[0020] As a preferred embodiment of the distribution network risk management method under extreme weather conditions described in this invention, the establishment of the emergency communication network includes:

[0021] The coverage area of ​​the communication network is determined based on the risk level assessment results, and a frequency domain frequency selection mechanism is initiated using a wireless self-organizing network architecture.

[0022] Physical layer connections are established through multiple-input multiple-output (MIMO) technology, supporting adaptive modulation and coding, and optimizing transmission parameters based on the direct and potential risks.

[0023] Select the optimal relay node and construct a redundant communication link that supports multi-hop transmission based on the risk level assessment results.

[0024] The beneficial effects of this preferred technical solution are as follows: the self-organizing communication network based on the risk level can dynamically adjust the coverage and transmission parameters according to the actual risk situation, and can maintain uninterrupted communication even if some communication facilities are damaged, thereby enhancing the resilience and adaptability of emergency communication.

[0025] As a preferred embodiment of the power distribution network risk management method under extreme weather conditions described in this invention, the power dispatching steps include:

[0026] Dispatch instructions are transmitted through the emergency communication network, and load allocation is adjusted based on the risk level assessment results.

[0027] The emergency communication network is used to coordinate power distribution among various nodes of the power grid and to optimize power supply paths based on the direct and potential risks.

[0028] Based on the aforementioned failure probability, energy storage devices and distributed power sources are dynamically scheduled to maintain stable grid operation.

[0029] As a preferred embodiment of the distribution network risk management method under extreme weather conditions described in this invention, the generation of the decision scheme includes:

[0030] Construct a multi-level risk transmission map to map the risk level assessment results to disaster type, equipment identification, emergency repair team and emergency plan nodes;

[0031] Graph algorithms are used to calculate node importance, and key nodes are identified first based on the direct and potential risks.

[0032] By using a query engine to correlate the failure probability with the emergency plan, a decision-making scheme is generated that includes emergency repair scheduling paths and equipment repair priorities.

[0033] The beneficial effects of this preferred technical solution are as follows: by constructing a multi-level risk transmission map and graph algorithm analysis, key nodes can be quickly identified and the optimal emergency repair scheduling plan can be generated. Compared with traditional manual experience-based decision-making, this shortens the emergency response time and improves the scientificity and efficiency of emergency repair resource allocation.

[0034] As a preferred embodiment of the power distribution network risk management method under extreme weather conditions described in this invention, the power supply guarantee includes:

[0035] The backup power equipment is activated according to the decision-making scheme, and the power resource allocation is coordinated through the emergency communication network.

[0036] The operating mode of the power management system is configured based on the risk level assessment results for automatic switching between primary and backup power supplies;

[0037] When the main power supply is interrupted, backup power is allocated according to the priority of the decision scheme, and the core node is maintained to continue operating through the emergency communication network.

[0038] Secondly, the present invention provides a power distribution network risk management system under extreme weather conditions, including a meteorological monitoring module, a weather forecasting module, a risk assessment module, and a decision support module;

[0039] The meteorological monitoring module is responsible for detecting weather conditions and acquiring meteorological data.

[0040] The weather forecasting module uses meteorological data to predict weather trends.

[0041] The risk assessment module analyzes the fault risks of each node in the power distribution network based on meteorological data and weather trends.

[0042] The decision support module generates a decision plan based on the analysis results of the failure risk.

[0043] Thirdly, the present invention provides an electronic device, comprising:

[0044] Memory and processor;

[0045] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the distribution network risk management method under extreme weather conditions.

[0046] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the power distribution network risk management method under extreme weather conditions.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0048] By constructing a comprehensive management approach integrating meteorological monitoring, risk assessment, emergency communication, power dispatching, and power supply assurance, the risk management capabilities of the distribution network under extreme weather conditions have been significantly improved. Multi-source data fusion technology integrates meteorological satellite, radar, and ground sensor data, improving the accuracy and timeliness of weather forecasts. Deep learning algorithms are used to intelligently analyze historical fault data and real-time meteorological data, enabling accurate prediction of the probability of faults at distribution network nodes. Furthermore, network topology analysis technology dynamically assesses direct and potential risks, providing a scientific basis for emergency decision-making. The emergency communication network, built using a wireless self-organizing network architecture, exhibits excellent resilience and self-healing capabilities, maintaining uninterrupted communication even when some base stations are damaged, ensuring the continuity of emergency command.

[0049] By constructing a multi-level risk transmission map using a graph database and employing intelligent decision-making algorithms, the system can quickly generate optimal decision-making schemes that include emergency repair scheduling paths and equipment repair priorities, shortening emergency response time and improving handling efficiency. Simultaneously, a dynamic power dispatch mechanism based on risk levels can automatically adjust load and power allocation according to real-time risk conditions, maintaining grid stability through the coordinated operation of energy storage devices and distributed power sources, thus reducing the impact of extreme weather on user power supply. The entire system achieves end-to-end automated management from risk warning to emergency response, enhancing the resilience and reliability of the distribution network in the face of extreme weather, and providing strong technical support for ensuring the safe and stable operation of the power system. Attached Figure Description

[0050] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a schematic diagram of the overall process of a power distribution network risk management method under extreme weather conditions according to an embodiment of the present invention. Detailed Implementation

[0052] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0053] Example 1, referring to Figure 1 As an embodiment of the present invention, a method for risk management of power distribution networks under extreme weather conditions is provided, comprising the following steps S1 to S3:

[0054] S1. Obtain meteorological data by detecting weather conditions and predict weather trends;

[0055] S2. Based on the meteorological data and weather trends, analyze the fault risks of each node in the power distribution network;

[0056] S3. Generate a decision plan based on the analysis results of the fault risk. The decision plan is combined with the emergency communication network to coordinate the execution of power dispatch and emergency response for power supply guarantee.

[0057] It should be noted that power distribution networks face severe challenges under extreme weather conditions. Typhoons, torrential rains, and icing can directly impact power equipment, leading to serious faults such as line breaks and tower collapses. Traditional meteorological monitoring relies mainly on a single data source, resulting in limited forecast accuracy and difficulty in accurately grasping the development trajectory and intensity changes of extreme weather. Existing risk assessment methods are mostly based on static thresholds and experience-based judgments, lacking in-depth analysis of the vulnerability of the power distribution network topology and failing to effectively predict the propagation path of cascading faults. Furthermore, emergency communication networks are easily disrupted by base station damage during extreme weather, affecting the timeliness and effectiveness of emergency command. Power dispatch often relies on manual experience, resulting in slow response times and difficulty in completing resource optimization within minutes. These problems severely restrict the safe and stable operation of the power distribution network under extreme weather conditions.

[0058] Therefore, to address the aforementioned issues such as delayed monitoring and early warning, static risk assessment, frequent communication interruptions, and slow dispatch response, a multi-source data fusion meteorological monitoring system is constructed through steps S1-S3 to achieve accurate prediction of extreme weather development trends; an intelligent risk assessment mechanism based on deep learning and network topology analysis is established to dynamically calculate the failure probability and risk level of each node, enabling quantitative assessment of direct and potential risks; simultaneously, a robust emergency communication network is constructed to ensure reliable transmission of emergency commands, and intelligent decision-making algorithms are combined to quickly generate optimal power dispatch and emergency response plans, achieving end-to-end automated management from risk identification to emergency response, thereby improving the resilience and reliability of the distribution network in the face of extreme weather.

[0059] Example 2, refer to Figure 1 This is one embodiment of the present invention. Based on the above embodiment, a method for risk management of power distribution networks under extreme weather conditions is provided.

[0060] In this embodiment of the application, step S1, the step of acquiring the meteorological data, includes A1 to A3:

[0061] A1. Integrates meteorological satellite, radar and ground sensor data to collect multi-source meteorological data in real time;

[0062] Specifically, multi-source meteorological data acquisition utilizes satellite imagery data with a resolution of 1 km obtained from the Fengyun-4 meteorological satellite, providing extensive cloud imagery and the overall structure of the typhoon; Doppler radar acquires radar echo data with radial velocity accuracy down to ±1 meter per second, enabling precise detection of precipitation intensity and wind field; and ground sensor arrays deployed at key nodes of the towers collect detailed meteorological data in real time, including typhoon path, rainfall, ice thickness, and wind speed. The miniaturized AI edge terminal integrated with the ground sensors includes a temperature sensor with an accuracy of ±0.5 degrees Celsius, a humidity sensor with an accuracy of ±2% relative humidity, an ultrasonic anemometer with an accuracy of ±0.5 meters per second, and an image recognition unit based on the YOLOv5s algorithm, capable of detecting ice thickness with an accuracy of ±1 millimeter and tower tilt angle with an accuracy of ±0.1 degrees.

[0063] A2. Perform spatiotemporal alignment and error correction on multi-source meteorological data to generate a typhoon path prediction model with a preset spatial resolution.

[0064] Specifically, the multi-source data fusion processing employs the Kalman filter algorithm to perform spatiotemporal alignment and error correction on meteorological data from satellites, radar, and ground sensors. The fusion process first establishes a state prediction model, combining historical data and a physical model to predict the meteorological conditions at the next moment; then, it uses real-time observation data to correct the prediction results, optimizing the fusion effect by dynamically adjusting the weights of each data source; finally, it generates a typhoon path prediction model with a preset spatial resolution, which can control the typhoon path prediction error within 3 kilometers.

[0065] A3. Deploy the typhoon path prediction model on the edge terminals of key nodes and upload the monitoring data to the monitoring platform through wireless communication protocols.

[0066] Specifically, the edge terminal deployment uses the LoRaWAN wireless communication protocol, possessing a transmission capability with a coverage radius of 10 kilometers, and uploads real-time monitoring data to the regional monitoring platform at minute-level intervals. Data transmission includes key parameters such as real-time temperature, humidity, wind speed, icing thickness, and tower tilt angle, while integrating data compression and error checking mechanisms to ensure stable data transmission quality even under severe weather conditions. The monitoring platform receives and stores this high-precision meteorological data, providing a reliable data foundation for subsequent risk assessment and decision generation.

[0067] In an optional implementation, the multi-source meteorological data acquisition in step S1 can also integrate X-band weather radar and C-band Doppler radar, significantly improving the ability to identify different precipitation types through dual-polarization radar technology. This technology can accurately distinguish mixed precipitation phenomena such as rain, snow, and hail. By analyzing the differences in horizontal and vertical polarization echoes and combining a comprehensive judgment of reflectivity factor and differential reflectivity factor, it is suitable for precise monitoring under complex weather conditions.

[0068] In another optional implementation, the typhoon track prediction model in step S1 can also employ an ensemble forecasting method. This method quantifies the uncertainty of the prediction by running multiple different numerical weather prediction models and performing ensemble operations. The method weights and averages the results of multiple forecast models and calculates the probability distribution of the prediction results to generate a probabilistic typhoon track prediction that includes uncertainty information. Ensemble forecasts can provide the probability distribution range of possible typhoon tracks, offering more comprehensive and reliable uncertainty quantification information for risk assessment and helping decision-makers develop more robust contingency plans.

[0069] It should be noted that the core reason for choosing the aforementioned multi-source data fusion technology lies in the significant limitations of single data sources: while satellite data has a wide spatial coverage, its temporal resolution is relatively limited, making it difficult to capture rapidly changing weather phenomena; radar data has high temporal resolution, but is easily affected by terrain obstruction and electromagnetic interference; and while ground sensors have high measurement accuracy, their spatial representativeness is insufficient, making it difficult to reflect large-scale weather conditions. Through the state estimation and error correction mechanism of the Kalman filter algorithm, the advantages of each data source can be fully utilized, effectively compensating for the shortcomings of a single data source, achieving complementary advantages and error elimination, significantly improving the accuracy and reliability of typhoon path prediction, and providing high-quality, highly reliable meteorological data support for power distribution network risk management.

[0070] In this embodiment of the application, step S2, which involves analyzing the fault risks of each node in the distribution network, includes steps B1 to B3:

[0071] B1. Extract features from historical fault data and meteorological data to predict the fault probability of each distribution network node within a preset time window;

[0072] B2. Analyze vulnerable nodes in the power grid through network topology analysis, combine the aforementioned failure probability with equipment tolerance threshold, calculate direct risks and potential risks, and generate risk level assessment results;

[0073] B3. When the meteorological data exceeds the preset threshold, a dynamic early warning is triggered based on the risk level assessment result.

[0074] B2. Analyze vulnerable nodes in the power grid through network topology analysis, combine the aforementioned failure probability with equipment tolerance threshold, calculate direct risks and potential risks, and generate risk level assessment results;

[0075] B3. When the meteorological data exceeds the preset threshold, a dynamic early warning is triggered based on the risk level assessment result.

[0076] Specifically, in step B1, feature extraction and fault probability prediction are based on the Long Short-Term Memory (LSTM) deep learning algorithm. The time series length of the input historical fault data is set to 24 hours, and comprehensive analysis is performed by combining real-time meteorological data, including key parameters such as wind speed, icing thickness, rainfall, and temperature. The algorithm first learns historical fault patterns, identifies fault patterns and triggering characteristics under different meteorological conditions, and then, based on the changing trends of current meteorological data, predicts the fault probability of each distribution network node in the next 2 hours.

[0077] Specifically, in step B2, the network topology analysis uses the Apollonius network topology algorithm to identify vulnerable nodes and critical connection paths in the power grid. The analysis process first constructs a topology model of the distribution network, identifying topological characteristics such as connectivity, betweenness centrality, and proximity centrality of each node. Then, combining the fault probability obtained in step B1 with the tolerance threshold parameters of each device, including wind speed tolerance limit, icing bearing capacity, and temperature operating range, the direct and potential risks of each node are calculated. Direct risk is quantified as the expected number of line breaks, reflecting the direct impact of a single node failure; potential risk is assessed as the probability of cascading failures, analyzing the scope and severity of the cascading effects that a single point of failure may trigger. These two types of risk indicators are combined to generate a five-level risk assessment result from low to high.

[0078] Specifically, in step B3, the dynamic early warning mechanism sets multi-level meteorological threshold trigger conditions. The early warning procedure is activated when any of the following conditions are met: wind speed reaches or exceeds 30 meters per second, icing thickness reaches or exceeds 10 millimeters, or continuous rainfall exceeds 50 millimeters per hour. After the early warning is triggered, based on the risk level assessment results generated in step B2, a Level 1 early warning is issued for high-risk areas, and a Level 2 early warning is issued for medium-risk areas. Simultaneously, load shedding instructions and energy storage activation instructions are generated. The early warning response time is controlled within 30 seconds to ensure that emergency measures can be activated promptly, reducing the likelihood and scope of failures.

[0079] In an optional implementation, step S2, the fault probability prediction can also employ a multi-model fusion method, combining various machine learning algorithms such as Support Vector Machine (SVM), Random Forest, and Neural Networks to improve the robustness of the prediction through ensemble learning techniques. This method weights and fuses the prediction results of different algorithms, effectively reducing the bias of a single model and maintaining stable prediction performance under complex weather conditions, making it particularly suitable for fault prediction in extreme weather events.

[0080] In another optional implementation, step S2 can also incorporate graph neural network technology into the network topology analysis. By learning the deep features of the power grid topology, it can automatically identify hidden vulnerability patterns and critical paths. This technology can handle the topology analysis of large-scale complex power grids. Through node embedding and graph convolution operations, it can discover vulnerability features that are difficult to identify by traditional topology analysis methods, thereby improving the comprehensiveness and accuracy of risk identification.

[0081] It should be noted that the main reason for choosing a deep learning-based fault risk analysis method is the significant shortcomings of traditional static threshold judgment methods: traditional methods mainly rely on empirical formulas and fixed thresholds, which cannot adapt to complex and changing meteorological conditions and power grid operating states; they lack the ability to learn from historical fault patterns, making it difficult to capture the potential patterns and triggering mechanisms of fault occurrence; and their predictive ability for cascading faults is limited, failing to effectively assess the cascading impact of single-point faults. By combining the LSTM deep learning algorithm with Apollonius network topology analysis, hidden patterns in historical data can be fully explored, dynamically adapting to different meteorological conditions and network states, achieving accurate quantification of direct and potential risks, improving the accuracy and foresight of distribution network fault risk assessment, and providing a scientific and reliable basis for emergency decision-making.

[0082] In this embodiment of the application, step S3, the establishment of the emergency communication network includes C1 to C3:

[0083] C1. Determine the coverage area of ​​the communication network based on the risk level assessment results, and initiate the frequency domain frequency selection mechanism using a wireless self-organizing network architecture;

[0084] C2. Establish physical layer connections through multiple-input multiple-output (MIMO) technology, support adaptive modulation and coding, and optimize transmission parameters based on the direct and potential risks.

[0085] C3. Select the optimal relay node and construct a redundant communication link that supports multi-hop transmission based on the risk level assessment results.

[0086] Specifically, in step C1, the determination of the communication network coverage is based on the risk level assessment results generated in step S2, identifying high-risk areas and key equipment nodes, and prioritizing communication coverage in these areas. The wireless ad hoc network architecture adopts Ad hoc network technology, enabling direct communication between devices without fixed infrastructure. The frequency domain frequency selection mechanism monitors the interference of each sub-channel in real time within the 5150 to 5850 MHz frequency band, and automatically selects the frequency channel with the least interference and the best signal quality for communication through a channel quality assessment algorithm, ensuring stable communication quality even in complex electromagnetic environments. When a channel quality degradation is detected, channel switching can be completed within milliseconds to ensure communication continuity.

[0087] Specifically, in step C2, the physical layer connection employs 2×2 multiple-input multiple-output (MIMO-OFDM) technology, using a multi-antenna system to improve data transmission rate and anti-interference capability. It supports multiple adaptive modulation and coding schemes such as BPSK, QPSK, 16QAM, and 64QAM, automatically selecting the optimal modulation scheme based on channel conditions. The transmission parameter optimization mechanism, combined with the direct and potential risk assessment results obtained in step S2, automatically increases transmission power and coding redundancy in high-risk areas, employing more reliable low-order modulation schemes to ensure the accuracy of information transmission; while in low-risk areas, high-order modulation schemes are used to improve transmission efficiency. The channel bandwidth can be flexibly selected between 5MHz, 10MHz, 20MHz, and 40MHz, with a maximum transmission rate of up to 270Mbps, meeting the high-speed data transmission requirements of emergency command.

[0088] Specifically, in step C3, the optimal relay node selection employs a gradient path algorithm, comprehensively considering multiple factors such as the node's geographical location, remaining energy, channel quality, and load. Based on the risk level assessment results from step S2, the algorithm prioritizes nodes located in low-risk areas with good communication conditions as relay nodes, avoiding the establishment of critical communication paths in high-risk areas. The multi-hop transmission mechanism supports data forwarding of more than 5 hops, and when a relay node fails, it can automatically find alternative paths, achieving millisecond-level path switching. The redundant communication link design ensures that each critical node has at least two independent communication paths, and improves communication reliability through path diversity technology, maintaining the connectivity of the entire network even when some nodes fail.

[0089] In an optional implementation, in step S3, the emergency communication network can also integrate a satellite communication link as a backup communication method, switching to satellite communication mode when a large-scale failure occurs in the ground-based ad hoc network. The satellite link uses a C-band radio frequency unit, which has the advantage of being unaffected by ground facilities and can provide reliable long-distance communication capabilities under extreme weather conditions, making it particularly suitable for emergency communication support in large-scale disaster scenarios.

[0090] In another optional implementation, in step S3, the communication security mechanism can also employ 128-bit AES link encryption and IPSec VPN tunneling technology to ensure the secure transmission of control commands and sensitive data. The encryption mechanism uses a tiered encryption strategy for different types of data, employing the highest level of encryption protection for critical control commands and standard encryption methods for general monitoring data, thus ensuring both data security and improved transmission efficiency. It also integrates authentication and access control mechanisms to prevent unauthorized devices from accessing the network.

[0091] It should be noted that the choice of ad hoc wireless ad hoc network architecture for emergency communication technology is primarily based on the vulnerability of traditional fixed base station communication under extreme weather conditions: fixed base stations are easily damaged by physical forces or power outages due to severe weather such as typhoons and rainstorms, causing widespread communication disruptions; traditional communication networks lack self-healing capabilities, and single-point failures can easily trigger cascading effects; fixed network topologies cannot adapt to dynamic changes in demand during disasters. By combining ad hoc wireless ad hoc network architecture with multiple-input multiple-output (MIMO) technology, decentralized distributed communication can be achieved. Each node acts as both a user terminal and a data forwarding device, improving the network's resilience and self-healing capabilities, ensuring that communication connections in critical areas can be maintained even in harsh environments, and providing reliable communication support for distribution network emergency management.

[0092] The power dispatching steps include D1 to D3:

[0093] D1. Transmit dispatch instructions through the emergency communication network and adjust load allocation according to the risk level assessment results;

[0094] D2. Coordinate the power distribution among the nodes of the power grid through the emergency communication network, and optimize the power supply path according to the direct risks and potential risks;

[0095] D3. Dynamically schedule energy storage devices and distributed power sources according to the aforementioned failure probability to maintain stable grid operation.

[0096] Specifically, in step D1, dispatch instructions are transmitted through the emergency communication network established in step S3, and differentiated load adjustment strategies are formulated based on the risk level assessment results generated in step S2. For high-risk areas, a load reduction mode is activated, generating peak shaving and valley filling strategies according to the standard of 10% adjustable air conditioning load and 20% adjustable electric vehicle charging and discharging power, prioritizing power supply for critical equipment and residential use; for medium-risk areas, a moderate adjustment mode is adopted, adjusting interruptible loads through intelligent load controllers to maintain supply and demand balance; for low-risk areas, a normal power supply mode is maintained, with only minor adjustments made to some flexible loads. Load adjustment instructions are sent to each power consumption node through encrypted communication links, achieving a user response rate of up to 92% and a peak load reduction capacity of 4.6MW.

[0097] Specifically, in step D2, the power allocation coordination mechanism is based on a two-layer game model, achieving coordination optimization among multiple entities through the emergency communication network in step S3. The upper-layer optimization strategy dynamically adjusts the electricity pricing strategy and power supply priority based on the direct and potential risk assessment results of step S2, prioritizing power supply to critical loads in high-risk areas and implementing economic dispatch for general loads in low-risk areas. The lower-layer coordination mechanism allocates power trading revenue among nodes through the Nash negotiation algorithm, achieving energy mutual assistance among multiple microgrids. Power supply path optimization automatically avoids high-risk lines and selects safe and reliable transmission paths by real-time monitoring of the load and health status of each line, combined with risk assessment results, ensuring stable power delivery.

[0098] Specifically, in step D3, the dynamic scheduling of energy storage devices and distributed power sources makes intelligent decisions based on the fault probability predicted in step S2. When the fault probability in a certain area exceeds a set threshold, the energy storage devices in that area are activated in advance for charging and reserve, while the output plans of distributed photovoltaic, wind power, and other new energy devices are adjusted. The energy storage scheduling strategy adopts a predictive control algorithm, which optimizes the charging and discharging plans of energy storage devices based on load forecasts and risk assessments for the next two hours, ensuring that backup power can be provided in a timely manner when a fault occurs. Distributed power source scheduling balances the intermittent characteristics of new energy sources through carbon capture and power-to-gas conversion devices, converting excess renewable energy into hydrogen or other chemical energy for storage, and then converting it back into electrical energy when needed, thus realizing the spatiotemporal transfer of energy.

[0099] In an optional implementation, step S3 can further integrate artificial intelligence optimization algorithms into power dispatching, employing deep reinforcement learning techniques to train dispatching strategies. Through interactive learning with the power grid environment, the optimal dispatching rules are automatically discovered. This method can adapt to different operating scenarios and fault modes, continuously optimizing dispatching performance over long-term operation, and is particularly suitable for power dispatching under complex and variable extreme weather conditions.

[0100] In another optional implementation, a demand response mechanism can be introduced in step S3. This mechanism uses price incentives and contractual agreements to encourage users to actively participate in load regulation. Demand response signals are issued based on real-time risk conditions, and users can respond to regulation requests via smart terminals to receive corresponding economic compensation. This mechanism can fully mobilize user enthusiasm, expand the scale of adjustable resources, and improve the flexibility and resilience of the power grid.

[0101] It should be noted that the choice of a risk-driven dynamic power dispatching method is primarily to address the insufficient adaptability of traditional dispatching methods under extreme weather conditions. Traditional dispatching relies heavily on fixed dispatching rules and human experience, resulting in slow response times and difficulty in quickly adapting to sudden changes in risk. It also lacks foresight regarding risk propagation, often only taking countermeasures after a fault occurs. Furthermore, resource allocation lacks specificity, failing to develop differentiated dispatching strategies based on the risk levels of different regions. By combining risk level assessment results with fault probability prediction, preventative dispatching and precise resource allocation can be achieved, enabling corresponding preventative measures to be taken before faults occur. This significantly improves the safety and stability of the distribution network under extreme weather conditions, minimizes the impact of power outages, and ensures the reliability of power supply for users.

[0102] In this embodiment of the application, step S3, the generation of the decision scheme includes E1 to E3:

[0103] E1. Construct a multi-level risk transmission map to map the risk level assessment results to disaster type, equipment identification, emergency repair team and emergency plan nodes;

[0104] E2. Use graph algorithms to calculate node importance, and prioritize the identification of key nodes based on the direct and potential risks.

[0105] E3. By using a query engine to correlate the fault probability with the emergency plan, a decision scheme is generated that includes emergency repair scheduling paths and equipment repair priorities.

[0106] Specifically, in step E1, the multi-level risk transmission graph is constructed based on the Neo4j graph database architecture, converting the risk level assessment results generated in step S2 into nodes and edge relationships in the graph database. Node types include disaster type nodes such as extreme weather events like typhoons, rainstorms, and icing; equipment identification nodes containing power equipment information such as tower numbers, transformer IDs, and switchgear identifiers; emergency repair team nodes recording the location, professional skills, and available resources of each emergency repair team; and emergency plan nodes storing standard handling procedures corresponding to different fault types. Edge relationships adopt multimodal association rules, including "cause" relationships describing the degree of impact of disaster events on equipment, "impact" relationships quantifying the propagation range of equipment faults, and "handling" relationships connecting fault types with corresponding emergency repair plans. The graph construction process uses the risk level assessment results as node attributes, realizing the effective transmission and association of risk information throughout the entire emergency response system.

[0107] Specifically, in step E2, the node importance calculation uses the PageRank graph algorithm, comprehensively considering multiple dimensions such as node connectivity, centrality, and risk weight. The algorithm first analyzes the topological position and connectivity of each node in the graph, identifying nodes that play a key role in the risk transmission path. Then, combining the direct and potential risk assessment results obtained in step S2, it assigns corresponding risk weights to different nodes, and nodes with a PageRank value greater than or equal to 0.15 are identified as critical nodes. The critical node identification process prioritizes equipment with a significant impact on grid stability, such as hub substations, major transmission lines, and important load centers, while also prioritizing equipment in high-risk areas to ensure that emergency resources are allocated to the most needed locations.

[0108] Specifically, in step E3, the query engine uses the Cypher query language and a multi-hop inference mechanism to intelligently match the fault probability predicted in step S2 with historical emergency plans. The query process first filters relevant emergency plans based on fault type, impact range, and severity. Then, considering currently available repair resources and team distribution, it generates the optimal repair scheduling path. The decision plan includes a detailed equipment repair priority ranking, formulating a strategy of repairing critical equipment first and then handling general faults based on node importance calculations and fault urgency. Repair path planning considers practical constraints such as road conditions, repair team skill matching, and equipment spare parts availability to ensure the plan's feasibility. The entire plan matching process is completed within 186 milliseconds, meeting the real-time requirements of emergency decision-making.

[0109] In one optional implementation, the decision-making process in step S3 can also integrate machine learning algorithms. By analyzing the success rate and effectiveness evaluation of historical emergency response cases, the accuracy of plan matching can be continuously optimized. This allows the system to learn the best handling strategies for different fault scenarios, automatically update and improve the emergency plan library, and enhance the scientific rigor and practicality of the decision-making process.

[0110] In another optional implementation, a collaborative optimization mechanism can be introduced in step S3, which simultaneously considers the repair needs and resource constraints of multiple fault points, and solves the globally optimal resource allocation scheme through an integer programming algorithm. This mechanism can avoid resource conflicts and redundant configurations, maximize repair efficiency, and is particularly suitable for emergency response to large-scale cascading faults.

[0111] It should be noted that the choice of a knowledge graph-based intelligent decision generation method is primarily to overcome the problems of information silos and decision lags in traditional emergency decision-making processes. Traditional emergency decision-making mainly relies on human experience and static plans, lacking a deep understanding of complex risk transmission relationships; information between different departments and systems lacks effective integration, making it difficult to form a unified situational awareness; and the plan matching process is time-consuming, failing to meet the timeliness requirements of emergency response. By constructing a multi-level risk transmission graph and using graph algorithm analysis, it is possible to achieve structured organization and correlation analysis of risk information, quickly identify key risk points and optimal handling strategies, significantly improve the scientific nature, timeliness, and effectiveness of emergency decision-making, and provide intelligent decision support for the emergency management of power distribution networks during extreme weather.

[0112] In step S3, power supply assurance includes: activating backup power equipment according to the decision scheme, coordinating power resource allocation through the emergency communication network; configuring the working mode of the power management system according to the risk level assessment results for automatic switching between primary and backup power supplies; and allocating backup power according to the priority of the decision scheme when the primary power supply is interrupted, and maintaining the continuous operation of the core node through the emergency communication network.

[0113] Specifically, the backup power equipment is automatically activated based on the priority order in the decision scheme generated in step E3. First, the intelligent cable reel device is activated, providing 5 kW of power and a 1 gigabit per second fiber optic communication link to the tethered drone via a fiber optic composite cable. This enables the drone to provide continuous video surveillance transmission and emergency power supply services while hovering at a height of 50 meters. The backup power supply consists of a wide-temperature-range lithium battery pack composed of lithium iron phosphate cells, capable of maintaining over 80% of its capacity performance in extreme temperature environments ranging from -30°C to 80°C. Power resource allocation is coordinated through the emergency communication network established in step C3. Backup power resources are intelligently allocated based on the importance level of each node and current load demand, prioritizing the power supply needs of critical equipment and important loads.

[0114] The power management system's operating mode configuration is dynamically adjusted based on the risk level assessment results generated in step B2. For high-risk areas, it switches to a high-reliability mode, enabling redundant battery packs and bidirectional DC / DC modules, supporting hot-swapping, and controlling interruption time to within 10 milliseconds. For medium-risk areas, it adopts a standard standby mode to maintain a certain power reserve. For low-risk areas, it maintains normal power supply mode. The automatic primary / backup power switching mechanism monitors the voltage, frequency, and power quality parameters of the primary power supply in real time. When an anomaly or interruption of the primary power supply is detected, it can automatically switch to the backup power supply within milliseconds to ensure the continuity and stability of power supply.

[0115] The priority allocation of backup power is strictly executed according to the critical node importance ranking determined in step E3. When the main power supply is interrupted, priority is given to ensuring power supply to critical nodes with a PageRank value greater than or equal to 0.15, including hub substations, important load centers, and emergency command centers; then, power resources are allocated to secondary nodes tier by tier based on remaining power capacity. The continuous operation of core nodes is coordinated and managed through the emergency communication network of step C3, which monitors the power status and load changes of each node in real time and dynamically adjusts the power allocation strategy. C-band radio frequency units provide remote communication support, ensuring that remote monitoring and control of core nodes can be maintained even if ground communication facilities are damaged. Backup power can guarantee uninterrupted power supply to core nodes for more than 8 hours, providing sufficient time for fault repair and emergency response.

[0116] In an optional implementation, the power supply guarantee in step S3 can also integrate a mobile emergency power vehicle and portable generators as a supplement to the fixed backup power supply. The mobile power equipment can be quickly deployed to the fault site to provide temporary power support, and is particularly suitable for emergencies involving large-scale power outages or severe equipment damage.

[0117] In another optional implementation, a microgrid islanding mode can also be introduced in step S3. When the connection with the main grid is interrupted, important areas can form an independent microgrid system through distributed power sources and energy storage devices to maintain the local area's independent power supply and improve power supply reliability and resilience.

[0118] It should be noted that the selection of multi-level backup power supply technology is primarily to address the multiple challenges faced by power systems under extreme weather conditions: main power sources are prone to interruption due to line damage, equipment failure, etc.; traditional backup power systems lack intelligent management and cannot dynamically allocate resources according to actual needs; fixed backup power sources may be damaged simultaneously in large-scale disasters, lacking flexible backup solutions. By integrating intelligent wire retraction and deployment devices, wide-temperature-range lithium battery packs, and multi-mode power management systems, a three-dimensional power supply protection system can be constructed, providing reliable power support under various fault scenarios, ensuring the continuous operation of the core functions of the distribution network, and minimizing the impact of extreme weather on power supply.

[0119] In summary, by constructing a comprehensive management approach integrating meteorological monitoring, risk assessment, emergency communication, power dispatching, and power supply assurance, the risk management capabilities of the distribution network under extreme weather conditions have been significantly improved. Multi-source data fusion technology integrates meteorological satellite, radar, and ground sensor data, improving the accuracy and timeliness of weather forecasts. Deep learning algorithms are used to intelligently analyze historical fault data and real-time meteorological data, enabling accurate prediction of the probability of faults at distribution network nodes. Furthermore, network topology analysis technology dynamically assesses direct and potential risks, providing a scientific basis for emergency decision-making. The emergency communication network built using a wireless self-organizing network architecture exhibits excellent resilience and self-healing capabilities, maintaining uninterrupted communication even when some base stations are damaged, ensuring the continuity of emergency command.

[0120] By constructing a multi-level risk transmission map using a graph database and employing intelligent decision-making algorithms, optimal decision-making schemes that include emergency repair scheduling paths and equipment repair priorities can be quickly generated, shortening emergency response time and improving handling efficiency. Simultaneously, a dynamic power dispatch mechanism based on risk levels can automatically adjust load and power allocation according to real-time risk conditions, maintaining grid stability through the coordinated operation of energy storage devices and distributed power sources, thus reducing the impact of extreme weather on user power supply. This achieves fully automated management across the entire chain from risk warning to emergency response, enhancing the resilience and reliability of the distribution network in the face of extreme weather, and providing strong technical support for ensuring the safe and stable operation of the power system.

[0121] Example 3 illustrates a schematic scheme for a distribution network risk management method under extreme weather conditions. It should be noted that the technical solution of this distribution network risk management system under extreme weather conditions is based on the same concept as the technical solution of the aforementioned distribution network risk management method under extreme weather conditions. Details not described in detail in this example can be found in the description of the technical solution of the aforementioned distribution network risk management method under extreme weather conditions.

[0122] This embodiment also provides a distribution network risk management system under extreme weather conditions, including a meteorological monitoring module, a weather forecasting module, a risk assessment module, and a decision support module;

[0123] The meteorological monitoring module is responsible for detecting weather conditions and acquiring meteorological data.

[0124] The weather forecasting module uses meteorological data to predict weather trends.

[0125] The risk assessment module analyzes the fault risks of each node in the power distribution network based on meteorological data and weather trends.

[0126] The decision support module generates a decision plan based on the analysis results of the failure risk.

[0127] This embodiment also provides an electronic device suitable for distribution network risk management under extreme weather conditions, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the distribution network risk management method under extreme weather conditions as proposed in the above embodiment.

[0128] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements the method for managing distribution network risks under extreme weather conditions as proposed in the above embodiments.

[0129] The storage medium proposed in this embodiment and the method for implementing distribution network risk management under extreme weather conditions proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0130] Based on the above description of the implementation methods, those skilled in the art will clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0131] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for risk management of power distribution networks under extreme weather conditions, characterized in that, Includes the following steps: Meteorological data is obtained by monitoring weather conditions, and weather trends are predicted. Based on the meteorological data and weather trends, analyze the fault risks of each node in the power distribution network; Based on the analysis of fault risks, a decision-making plan is generated. This plan, combined with the emergency communication network, coordinates the execution of power dispatch and emergency response for power supply assurance.

2. The method for risk management of power distribution networks under extreme weather conditions as described in claim 1, characterized in that, The steps for obtaining the meteorological data include: It integrates meteorological satellite, radar and ground sensor data to collect multi-source meteorological data in real time; Spatiotemporal alignment and error correction are performed on multi-source meteorological data to generate a typhoon track prediction model with a preset spatial resolution. The typhoon path prediction model is deployed on the edge terminals of key nodes, and the monitoring data is uploaded to the monitoring platform through wireless communication protocols.

3. The method for risk management of power distribution networks under extreme weather conditions as described in claim 2, characterized in that, The steps for analyzing the fault risks at each node of the distribution network include: Feature extraction is performed on historical fault data and meteorological data to predict the fault probability of each distribution network node within a preset time window; By analyzing the network topology of vulnerable nodes in the power grid, and combining the failure probability with the equipment tolerance threshold, direct risks and potential risks are calculated to generate risk level assessment results. When the meteorological data exceeds a preset threshold, a dynamic early warning is triggered based on the risk level assessment result.

4. The method for risk management of power distribution networks under extreme weather conditions as described in claim 3, characterized in that, The establishment of the emergency communication network includes: The coverage area of ​​the communication network is determined based on the risk level assessment results, and a frequency domain frequency selection mechanism is initiated using a wireless self-organizing network architecture. Physical layer connections are established through multiple-input multiple-output (MIMO) technology, supporting adaptive modulation and coding, and optimizing transmission parameters based on the direct and potential risks. Select the optimal relay node and construct a redundant communication link that supports multi-hop transmission based on the risk level assessment results.

5. The method for risk management of power distribution networks under extreme weather conditions as described in claim 4, characterized in that, The power dispatching steps include: Dispatch instructions are transmitted through the emergency communication network, and load allocation is adjusted based on the risk level assessment results. The emergency communication network is used to coordinate power distribution among various nodes of the power grid and to optimize power supply paths based on the direct and potential risks. Based on the aforementioned failure probability, energy storage devices and distributed power sources are dynamically scheduled to maintain stable grid operation.

6. The method for risk management of power distribution networks under extreme weather conditions as described in claim 5, characterized in that, The generation of the decision scheme includes: Construct a multi-level risk transmission map to map the risk level assessment results to disaster type, equipment identification, emergency repair team and emergency plan nodes; Graph algorithms are used to calculate node importance, and key nodes are identified first based on the direct and potential risks. By using a query engine to correlate the failure probability with the emergency plan, a decision-making scheme is generated that includes emergency repair scheduling paths and equipment repair priorities.

7. The method for risk management of power distribution networks under extreme weather conditions as described in claim 6, characterized in that, The power supply guarantee includes: The backup power equipment is activated according to the decision-making scheme, and the power resource allocation is coordinated through the emergency communication network. The operating mode of the power management system is configured based on the risk level assessment results for automatic switching between primary and backup power supplies; When the main power supply is interrupted, backup power is allocated according to the priority of the decision scheme, and the core node is maintained to continue operating through the emergency communication network.

8. A distribution network risk management system under extreme weather conditions, employing the method described in any one of claims 1-7, characterized in that, It includes a meteorological monitoring module, a weather forecasting module, a risk assessment module, and a decision support module; The meteorological monitoring module is responsible for detecting weather conditions and acquiring meteorological data. The weather forecasting module uses meteorological data to predict weather trends. The risk assessment module analyzes the fault risks of each node in the power distribution network based on meteorological data and weather trends. The decision support module generates a decision plan based on the analysis results of the failure risk.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the distribution network risk management method under extreme weather conditions as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the distribution network risk management method under extreme weather conditions as described in any one of claims 1 to 7.

Images