A refined early warning method and system for risks of overhead lines in power distribution networks

Abstract

This invention discloses a refined early warning method and system for overhead power distribution lines, comprising: collecting meteorological time-series data and power distribution network topology and line parameter data; preprocessing the data to construct a prediction dataset; establishing a fault probability prediction model based on meteorological data and line parameters; calculating the fault probability using a mathematical model, refining the model, and training it using static topology and dynamic meteorological data to construct a fault probability network model; analyzing the importance of power distribution network nodes and lines, and assessing the node load loss ratio; normalizing the importance indicators; comprehensively calculating the predicted fault probability, node load loss, and importance indicators to obtain branch risk parameters; and providing refined early warning based on a set warning threshold, dynamically adjusting to adapt to actual operating conditions. The advantages of this invention are: providing comprehensive risk assessment, improving the accuracy of fault prediction and the timeliness of risk early warning.

Description

Technical Field

[0001] This invention relates to the field of power system risk early warning technology, and in particular to a refined early warning method and system for overhead lines in distribution networks that takes into account the topology of the network structure. Background Technology

[0002] In modern power distribution networks, the reliability and stability face enormous challenges due to increasing electricity demand and the frequent occurrence of extreme weather events caused by climate change. Under extreme weather conditions, such as strong winds, torrential rain, and snow, overhead lines in the distribution network can suffer direct physical damage, leading to widespread power outages. Therefore, accurate early warning and effective response to these risks have become urgent problems for the current power system.

[0003] Traditional methods for risk assessment and early warning of distribution networks mainly rely on single historical fault data and static risk models. However, these methods often fail to reflect the impact of dynamic weather conditions on the reliability of distribution networks in real time. In addition, existing methods generally lack detailed analysis of the internal structure and node differences of distribution networks, and can only provide a rough global early warning, making it difficult for resource scheduling to efficiently respond to specific fault areas.

[0004] Currently, risk assessments of distribution networks typically employ static topology structures and simple weighted batching methods to calculate the importance of lines and nodes. This approach easily overlooks the load levels, geographical locations, and critical roles of different nodes within the overall network. In practice, due to the volatile nature of weather conditions, static analysis alone is insufficient to accurately pinpoint potential risk points, significantly reducing the accuracy and practicality of early warning systems.

[0005] Furthermore, the lack of quantitative methods for calculating the importance of different nodes and lines within the network makes it difficult for early warning systems to effectively identify and prioritize the protection of critical nodes and important lines during extreme weather events, leading to frequent network-wide blackouts. Therefore, a new method is urgently needed that comprehensively considers meteorological factors, the distribution network topology, and the importance of nodes and lines to achieve more refined and real-time risk warnings. This would enable power dispatch centers to take timely and targeted measures to reduce the scope of power outages and economic losses. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a refined early warning method and system for risks of overhead lines in distribution networks. By considering the importance of edges and nodes in the distribution network, and combining practical needs, the invention classifies the early warning levels of distribution network branches.

[0007] To achieve the above-mentioned objectives, the technical solution adopted by the present invention is as follows:

[0008] A refined early warning method for risks in overhead power distribution lines includes the following steps:

[0009] S1. Collect meteorological data for different typical areas of the power distribution network. The meteorological data includes, but is not limited to, wind speed, maximum wind speed, temperature, humidity, wind direction, and rainfall. Collect topology overhead line parameter data for the corresponding area, and preprocess the data, including: null value interpolation, averaging of repeated observations, and deletion of duplicate data, and establish a dataset.

[0010] S2. Construct a fault probability prediction model for overhead power lines in the distribution network based on time-series data. Based on overhead line parameter data and meteorological observation data, the fault probability is calculated using a mathematical probability model. The mathematical fault probability model is then corrected using actual power grid fault data. A fault probability network model for overhead lines is constructed by fusing static topology data with dynamic meteorological data. Based on the dataset constructed in step S1, the overhead line fault probability network model is trained to obtain the fault probability prediction model. The probability of each overhead line failing under specific meteorological conditions and topology is calculated using the fault probability prediction model.

[0011] S3. Based on the distribution network topology, model the importance of nodes and lines in the distribution network. Construct an undirected graph connecting different nodes and lines in the distribution network, considering the key roles of node types and overhead lines in the topology, and calculate the importance of lines and nodes; where node importance includes: node type, load level, and the load loss ratio in the topology when a node fails;

[0012] The importance of a line is determined by considering the connectivity of the network topology and the types of nodes connected to it. Node types include loads, main power sources, and distributed power sources. Node importance is ultimately aggregated and added to the edge importance through the aggregation of adjacent nodes.

[0013] S4. Combine the fault probability of each line predicted in S2 with the importance and load loss ratio of the overhead lines obtained in S3 to calculate the overhead line risk parameter index. Based on the set early warning thresholds, the overhead line risk index is divided into reasonable segments, and early warning strategies are specified according to these segments.

[0014] Further, S1 includes:

[0015] S11: Collect meteorological time-series datasets for different typical regions in the distribution network, including time, region, wind speed (v), temperature (t), wind direction angle (θ), humidity, precipitation, and maximum wind speed. Collect distribution network topology data and line parameter data for this region, and map these data to the IEEE 33-node topology diagram.

[0016] S12: Clean and preprocess the meteorological data. First, remove duplicate data, and replace the observations at the same location and within the same time step with the average value. Use bilinear interpolation to fill in missing values. Group the data for the same region according to the continuity of the time series to construct the dataset required in step S2.

[0017] Further, S2 includes:

[0018] S21: Construct a mathematical model for the probability of faults in overhead power lines of a distribution network, and calculate the probability of faults in overhead lines under different meteorological data. Based on the actual situation and mechanical performance characteristics of conductors in overhead transmission lines, and for the sake of simplifying the line components, the conductors are treated as hinges, the resistance changes caused by changes in conductor length are ignored, and it is assumed that the load on the conductor is uniformly distributed along its length. Therefore, a parabolic curve is used to calculate the relevant parameters. Under the action of horizontal wind v, ignoring the wind load along the line direction, the horizontal wind load W1 perpendicular to the conductor direction on each span of the conductor is:

[0019] W1=0.625α1μ1N′LDv 2 sin 2 θ×10 -3

[0020] In the formula: α1 is the conductor wind pressure non-uniformity coefficient; μ1 is the conductor shape coefficient; N′ is the number of phase conductor splits; L is the length of the conductor under wind load; D is the outer diameter of the conductor; θ is the angle between the wind direction and the conductor direction. In studies of plains or large-scale topographic environments, generally only W1 is considered. However, on hillsides, the vertical wind load F on the conductor should be considered. v for:

[0021]

[0022] In the formula: v v This refers to the upward wind speed along the vertical line.

[0023] G=N′m0gL

[0024] In the formula: m0 is the mass of the conductor per unit length; g is the acceleration due to gravity.

[0025] Under wind load, the total load Q and total specific load γ of this span of conductor are respectively:

[0026]

[0027]

[0028] Let the original length of the conductor at manufacturing temperature t0 be L0, and the length of the conductor under wind load be L.

[0029]

[0030] In the formula: σ av α is the average stress of the conductor; E is the elastic modulus of the conductor; α t t is the coefficient of thermal expansion of the conductor; m Temperature under wind load.

[0031] Based on the principle that the original length of the conductor is equal under windless and wind-loaded conditions, the horizontal stress σ0 of the conductor under wind load is obtained. For conductors in continuous spans, the influence of insulator offset on the elevation difference angle between suspension points is ignored. The horizontal distance l from the lowest point of the conductor to the higher tower is... m for:

[0032]

[0033] In the formula, l is the span of the line, and β is the elevation difference angle between the suspension points of the line.

[0034] The maximum stress σ of this conductor m

[0035]

[0036] The exponential function was used to fit the fault rate P of transmission lines under wind load overload. i With the maximum stress σ of the conductor m Relationship between wind speed v and typhoon speed:

[0037]

[0038] In the formula P i The probability of a utility pole failing is σ. s The ultimate stress that the conductor can withstand is σ0, which is related to the design stress of the conductor and the safety factor μ2 under wind load. K1 and T1 are constants related to the line parameters, and C1 is a constant related to the wind speed parameter.

[0039] S22: Construct a deep neural network model for fault probability prediction. This network model comprehensively considers overhead line parameter data and time-series meteorological data. Feature extraction method: Time-series data is extracted using a bidirectional LSTM network, and overhead line parameter data is processed through embedding. The two feature vectors are then concatenated into a single feature vector. The fault probability in S21 is used as the ground truth for training the network model. A fault probability prediction model is then trained.

[0040] Furthermore, S3 includes indicators for the importance of nodes and lines in the distribution network based on the static topology.

[0041] S31: Analyze node type weights, including power supply, primary load, secondary load, tertiary load, and important factor vectors.

[0042] S32: Analyze the importance indicators of nodes in the topology, specifically including degree centrality, proximity centrality, betweenness centrality, and eigenvector centrality.

[0043] Degree centrality:

[0044]

[0045] In the formula, deg(v) is the degree of node v, that is, the number of edges connected to node v, and n is the total number of nodes in the network.

[0046] Proximity centrality:

[0047]

[0048] Here, d(v,u) is the shortest path length from node v to node u, and the summation is performed on all nodes u that are different from v.

[0049] Betweenness centrality:

[0050]

[0051] Where σ st σ is the number of shortest paths from node s to node t. st (v) is the number of shortest paths passing through node v. The summation is performed on all pairs of nodes (s,t), where s and t are not equal to v to avoid self-loops.

[0052] Eigenvector centrality.

[0053] Let A be the adjacency matrix of the network, and w be the eigenvector, where w i Let represent the eigenvector centrality score of node i. The eigenvector centrality is calculated as follows:

[0054] w=λA w

[0055] Where A is the adjacency matrix of the network, and w is the eigenvector, where w i Let λ represent the eigenvector centrality score of node i, and λ be the largest eigenvalue of matrix A.

[0056] S33: Considering line importance, line types include lines connected to the power source, important shadow vectors, and line lengths. For the importance indices of edges in static topology, the analysis includes edge betweenness centrality, edge compact centrality, and eigenvector centrality.

[0057] Edge betweenness centrality:

[0058]

[0059] Where σ stσ is the total number of shortest paths from node s to node t. st (u,v) is the number of shortest paths from s to t that pass through edge (u,v). The summation is performed on all pairs of nodes (s,t), where s and t are not equal to v to exclude self-looping and repeating edges.

[0060] Edge-tight centrality:

[0061]

[0062] Among them, C C (u) and C C (v) represents the proximity centrality of the two endpoints u and u', respectively.

[0063] Eigenvector centrality: Calculate the eigenvector corresponding to the largest eigenvalue, and then normalize it so that its sum is 1. The normalization method is Min-Max normalization, which maps the data to the interval [0,1].

[0064]

[0065] Where x is the original data, Min is the minimum value of the data, Max is the maximum value of the data, and x′ is the transformed data.

[0066] Furthermore, S3 further includes: constructing an undirected graph of different nodes and different lines in the distribution network, analyzing the node type weights, and then calculating the importance and correlation of lines and nodes.

[0067] The probability of failure predicted by meteorological data is denoted as P, which ranges from 0 to 1.

[0068] Analyze a branch in a distribution network and calculate its edge importance and node importance.

[0069] Node importance C1: is composed of the weighted sum of degree centrality, proximity centrality, betweenness centrality, and eigenvector centrality.

[0070] The calculated C D (v),C C (v),C B (v) and w are normalized, and then summed to obtain the node importance C1, as shown in the following formula:

[0071] C1 = C D (v)+C C (v)+C B (v)+w

[0072] Edge importance C2: It consists of the weighted sum of edge betweenness centrality, edge compactness centrality, and eigenvector centrality.

[0073] Calculate C B (u,v),C E (u,v)w1 is normalized, and then summed to obtain the importance C2 of the edge, as shown in the following formula:

[0074] C2 = C B (u,v)+C E (u,v)+w1

[0075] Node load loss C3: This represents the proportion of load lost in the circuit relative to the total load of the distribution network when a node fails, as shown in the following formula:

[0076]

[0077] In the formula L i The load occupied by node i, L s Refers to the total load of the distribution network.

[0078] Therefore, the summation of the above weight indicators after normalization gives the importance C of the branch and node:

[0079] C = P + C1' + C2' + C3'

[0080] Based on the aforementioned importance indicators, refined threshold warnings can be implemented.

[0081] Furthermore, in S4:

[0082] S41: Adjust the importance C using the following formula:

[0083] C=P+ω1C1′+ω2C2′+ω3C3′+C0

[0084] Where ω1, ω2, and ω3 are weighting coefficients, and C0 is the subjective weight.

[0085] S42: Design a tiered early warning mechanism, setting different levels of warning signals based on risk levels and taking corresponding countermeasures. In the event of a high-level warning, the system can automatically notify relevant emergency personnel and suggest activating backup power or taking other emergency measures.

[0086] S43: Utilize historical fault data for model training and optimization to improve the model's responsiveness to unexpected situations. Monitor the distribution network's operational status in real time and feed new data back into the model for iterative optimization to continuously improve model performance.

[0087] S44: Dynamically adjust the weights of each node in the distribution network based on real-time data. Simultaneously, dynamically adjust the early warning level of the distribution network, taking into account special events and peak holiday periods.

[0088] This invention also discloses a refined early warning system for risks of overhead lines in distribution networks. This system can be used to implement the aforementioned refined early warning method for risks of overhead lines in distribution networks, specifically including:

[0089] Data collection module: Collects meteorological data and electrical parameters of the power distribution network, and records the geographical location of the power distribution network lines to obtain complete data input.

[0090] Data preprocessing module: Cleans and preprocesses the collected data, including removing duplicate data, interpolating to handle null values, and standardizing the data.

[0091] Topology Analysis Module: Based on the geographical information and line parameters of the distribution network, construct the static topology structure of the overhead lines of the distribution network and generate a static topology map of the distribution network.

[0092] Model building module: Based on distribution network parameters and meteorological time-series data, a refined risk early warning model is constructed. This includes establishing a fault probability prediction model and an importance analysis model for nodes and edges to assess the risks of the distribution network.

[0093] The calculation and analysis module performs weighted normalization based on the fault prediction probability, node load loss, edge importance, and node importance. It then obtains the importance of branches through weighted summation and provides refined risk warnings based on set thresholds.

[0094] Risk assessment module: Based on the calculation results, assess the risk level of the distribution network, set early warning thresholds, classify risk levels, identify potential risk areas, and issue early warnings.

[0095] Early Warning Notification Module: Issues early warning notifications based on risk levels and recommends corresponding emergency measures. The system automatically notifies relevant emergency personnel for timely response to risk situations.

[0096] Real-time monitoring and optimization module: Monitors the operation of the distribution network in real time and iteratively optimizes the model based on new data. Dynamically adjusts the weights of nodes and edges to adapt to weather changes and special events.

[0097] Historical data management module: Manages and utilizes historical fault data for model training and optimization to improve the model's generalization ability and prediction accuracy. Stores and analyzes historical data to help improve the system's responsiveness to unexpected situations.

[0098] The present invention also discloses a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-mentioned method for refined early warning of risks of overhead lines in power distribution networks.

[0099] The present invention also discloses a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-mentioned method for refined early warning of risks of overhead lines in power distribution networks.

[0100] Compared with the prior art, the advantages of the present invention are as follows:

[0101] This invention collects meteorological data from different typical areas of the power distribution network and constructs a fault probability prediction model based on time-series data, achieving accurate prediction of power distribution network fault probabilities. This method overcomes the limitation of traditional prediction models being insensitive to changes in complex meteorological conditions, improving the accuracy and timeliness of predictions. Simultaneously, by incorporating static topology, fault indicators for overhead lines are corrected, making the power distribution network model more closely reflect actual operating conditions and providing a solid foundation for risk early warning.

[0102] This invention fully considers the complexity of the distribution network topology and its impact on fault propagation. This innovation enables risk warning to go beyond the fault probability of a single line or node, allowing for a comprehensive assessment of the risk status of the entire network, thereby leading to the development of more scientific and reasonable early warning strategies.

[0103] This invention comprehensively considers multi-dimensional information such as distribution network structure parameters, fault probability predicted by meteorological data, line level, and load level. By normalizing the weights of different magnitudes, it ultimately obtains the fault warning level of the distribution network. This method effectively avoids the one-sidedness of single-factor assessment and improves the comprehensiveness and accuracy of risk assessment. Simultaneously, by conducting rationality assessments and formulating warning strategies based on set warning thresholds, the warnings become more timely and accurate, facilitating the advance scheduling of appropriate resources and reducing fault losses.

[0104] This invention utilizes node importance analysis technology to differentiate between different nodes in a distribution network, fully considering the criticality and scope of influence of each node within the network, as well as its importance in real-world scenarios. This innovation makes the formulated fault early warning strategy more targeted and better able to address the risks posed by uncertainties. When implementing the early warning strategy, priority can be given to ensuring the power supply stability of critical nodes, reducing the cascading effects and escalating losses caused by failures at key nodes. Attached Figure Description

[0105] Figure 1 This is a flowchart of the method for refined early warning of risks in overhead lines of power distribution networks according to an embodiment of the present invention.

[0106] Figure 2 This is a power distribution network topology diagram according to an embodiment of the present invention. Detailed Implementation

[0107] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and examples.

[0108] like Figure 1 As shown, this embodiment uses an improved IEEE 33-node distribution network system as a test case, with the specific structure as follows: Figure 2 As shown. The system contains 33 nodes, 32 sectionalizing switches, and 5 tie switches. Distributed power sources are configured at nodes 10, 18, 21, 24, and 31. Nodes 10, 20, and 24 are photovoltaic power generation units equipped with energy storage devices, node 18 is directly equipped with energy storage devices, and node 31 is equipped with a diesel generator. The specific steps of the refined early warning method and system for overhead line risks in distribution networks considering the topology structure described in this invention are as follows:

[0109] S1. Collect meteorological data (including but not limited to wind speed, maximum wind speed, temperature, humidity, wind direction, rainfall, etc.) of different typical areas of the power distribution network, collect topology overhead line parameter data (such as line length, material, etc.) of the corresponding area, and preprocess the data, including null value interpolation, averaging of repeated observations, and deleting duplicate data.

[0110] The sub-steps of S1 are as follows:

[0111] S11: Collect meteorological time-series datasets for different typical regions in the power distribution network, including time, region, wind speed v, temperature t, wind direction angle θ, humidity, precipitation, and maximum wind speed. Also collect topology data and line parameter data for these regions. Transfer this data to the IEEE 33-node topology map.

[0112] S12: Cleaning and preprocessing the meteorological data. First, duplicate data are removed, and observations at the same location and within the same time step are replaced with the average value. Missing values ​​are filled using bilinear interpolation. Data from the same region are grouped according to temporal continuity. The dataset required for S2 is then constructed.

[0113] S2. Construct a fault probability prediction model for overhead power lines in the distribution network based on time-series data. Based on overhead line parameter data and meteorological observation data, the fault probability is calculated using a mathematical probability model. The mathematical fault probability model is then corrected using actual power grid fault data. A fault probability network model for overhead lines is constructed by fusing static topology with dynamic meteorological data. Based on the dataset constructed in step S1, the overhead line fault probability network model is trained to obtain the fault probability prediction model.

[0114] The sub-steps of S2 are as follows:

[0115] S21: Based on time-series data, a fault probability prediction model for overhead lines in a distribution network is constructed. Under wind load, the total load Q and total specific load γ of a certain span of conductor are respectively:

[0116]

[0117]

[0118] Based on the principle that the original length of the conductor is equal under windless and wind-loaded conditions, the horizontal stress σ0 of the conductor under wind load can be obtained. For conductors in continuous spans, the influence of insulator offset on the elevation difference angle between suspension points is ignored. The horizontal distance l from the lowest point of the conductor to the higher tower is... m for:

[0119]

[0120] In the formula, l is the span of the line, and β is the elevation difference angle between the suspension points of the line.

[0121] The maximum stress σ of this conductor m

[0122]

[0123] The exponential function was used to fit the fault rate P of transmission lines under wind load overload. i With the maximum stress σ of the conductor m Relationship between wind speed v and typhoon speed:

[0124]

[0125] S22: Construct a deep neural network model for fault probability prediction. This network model comprehensively considers overhead line parameter data and time-series meteorological data. Feature extraction method: Time-series data is extracted using a bidirectional LSTM network, and overhead line parameter data is processed through embedding. The two feature vectors are then concatenated into a single feature vector. The fault probability in S21 is used as the ground truth for training the network model. A fault probability prediction model is then trained.

[0126] S3. Based on the distribution network topology, model the importance of nodes and lines in the distribution network. Construct an undirected graph connecting different nodes and lines in the distribution network, fully considering the key roles of node types and overhead lines in the topology, and then calculate the importance of lines and nodes. Node importance includes node type, load level, and the load loss ratio in the topology when a node fails. Edge importance mainly considers the connectivity of the topology network and the types of nodes connected to it. Node types include loads, main power sources, and distributed power sources. Finally, node importance is aggregated onto the edge importance through the aggregation of adjacent nodes.

[0127] S3 specifically includes:

[0128] Indicators of node importance and line importance in the distribution network based on static topology.

[0129] First, the node type weights are analyzed, including power sources, primary loads, secondary loads, tertiary loads, and importance factor vectors. The node importance indices in the topology are analyzed, including degree centrality, proximity centrality, betweenness centrality, and eigenvector centrality. The degree centrality C is calculated. D (v) (For undirected graphs):

[0130]

[0131] In the formula, deg(v) is the degree of node v, that is, the number of edges connected to node v.

[0132] Closeness centrality is a metric used in network analysis to measure the central position of a node. It reflects the average shortest path length from a node to all other nodes in the network. The higher the closeness centrality of a node, the shorter the path to other nodes, and thus the more central its position in the network.

[0133]

[0134] Here, d(v,u) is the shortest path length from node v to node u, and the summation is performed on all nodes u that are different from v.

[0135] Betweenness centrality is a metric in network analysis used to measure the importance of a node on the shortest path between all pairs of nodes in a network.

[0136]

[0137] Where σ st σ is the number of shortest paths from node s to node t. st (v) is the number of shortest paths passing through node v. The summation is performed on all pairs of nodes (s,t), where s and t are not equal to v to avoid self-loops.

[0138] Eigenvector centrality is a method for measuring node importance based on node connections. It considers not only the number of connections (degree) of a node but also the importance of its neighbors. It is calculated through an iterative process that propagates a node's score to its neighbors.

[0139] Let A be the adjacency matrix of the network, and w be the eigenvector, where w i Let represent the eigenvector centrality score of node i. The eigenvector centrality is calculated as follows:

[0140] w=λA w

[0141] Where λ is the largest eigenvalue of matrix A.

[0142] Secondly, the importance of the lines is considered. Line types include: the important shadow vector of lines connected to the power source, and line length. For the analysis of edge importance in static topology, indicators include: edge betweenness centrality, edge compact centrality, and eigenvector centrality.

[0143] Edge betweenness centrality is calculated based on the number of times an edge appears in the shortest path between all pairs of nodes. For an edge (u,v) in the network, its edge betweenness centrality is...

[0144]

[0145] Where σ st σ is the total number of shortest paths from node s to node t. st (u,v) is the number of shortest paths from s to t that pass through edge (u,v). The summation is performed on all pairs of nodes (s,t), where s and t are not equal to v to exclude self-looping and repeating edges.

[0146] Edge compact centrality is defined based on the compact centrality of the nodes at both ends of an edge. It takes into account the shortest path length from the two end nodes to all other nodes and is defined as follows:

[0147]

[0148] Eigenvector centrality can be determined by constructing the adjacency matrix of the network, calculating the eigenvector corresponding to its largest eigenvalue, and finally normalizing it so that its sum is 1.

[0149] In step S3, an undirected graph of different nodes and lines in the distribution network is constructed, the node type weights are analyzed, and then the importance and correlation of lines and nodes are calculated:

[0150] First, we denot the fault probability predicted by meteorological data as P, which ranges from 0 to 1. We analyze one branch of the distribution network and calculate its edge importance and node importance. The node importance is denoted as C1, and its degree centrality is denoted as C0. D (v), proximity centrality C C (v), betweenness centrality C B The weighted sum of (v) and the eigenvector centrality w can be calculated by first calculating C. D (v),C C (v),C B (v) and w are normalized and then summed to obtain the node importance C1.

[0151] C1 = CD (v)+C C (v)+C B (v)+w

[0152] Similarly, the importance of an edge, C2, is denoted as the edge betweenness centrality C. B (u,v), edge-close centrality C E (u,v), the weighted sum of eigenvector centrality w1, can be calculated first by... B (u,v),C E (u,v)w1 is normalized and then summed to obtain the importance of the edge C2.

[0153] C2 = C B (u,v)+C E (u,v)+w1

[0154] At the same time, we need to consider the load loss C3 of the node, that is, the proportion of the load lost in the circuit to the total load of the distribution network when this node fails.

[0155]

[0156] In the formula L i The load occupied by node i, L s Refers to the total load of the distribution network.

[0157] Therefore, the summation of the above weight indicators after normalization gives the importance C of the branch and node:

[0158] C=P+ω1C1′+ω2C2′+ω3C3′

[0159] In the formula, C1′ is the normalized node importance, C2′ is the normalized edge importance, and C3′ is the normalized node unload. Based on this importance, a refined threshold warning can be implemented.

[0160] S4: The model fully considers the practical significance of power distribution networks in real life. For example, the power distribution networks of hospitals and research institutes are of high importance regardless of electricity consumption. To prepare for emergency needs in actual production and daily life, the practicality of the area where the power distribution network branches are located is taken into account in the model, and is also used as a subjective weight C0 to affect the fault warning level of the power distribution network. That is, the importance C is modified accordingly.

[0161] C=P+ω1C1′+ω2C2′+ω3C3′+C0

[0162] By taking into account the distribution network structure parameters, the prediction probability obtained from meteorological data, the load loss of each node, the node level, and the load level, the generalization ability and robustness of the model can be improved.

[0163] In another embodiment of the present invention, a refined early warning system for risks of overhead lines in a distribution network is provided. This system can be used to implement the above-mentioned refined early warning method for risks of overhead lines in a distribution network, specifically including:

[0164] Data collection module: Collects meteorological data (including wind speed, wind direction, temperature, humidity, and rainfall) and electrical parameters (including line impedance, admittance parameters, etc.) of the power distribution network, and records the geographical location of the power distribution network lines to obtain complete data input.

[0165] Data preprocessing module: Cleans and preprocesses the collected data, including removing duplicate data, interpolating to handle null values, and standardizing data, to ensure data quality and consistency and provide a reliable foundation for subsequent analysis.

[0166] Topology Analysis Module: Based on the geographical information and line parameters of the distribution network, this module constructs the static topology structure of the overhead lines of the distribution network and generates a static topology map of the distribution network, providing a topological foundation for model construction.

[0167] Model building module: Based on distribution network parameters and meteorological time-series data, a refined risk early warning model is constructed. This includes establishing a fault probability prediction model and an importance analysis model for nodes and edges to assess the risks of the distribution network.

[0168] The calculation and analysis module performs weighted normalization based on the fault prediction probability, node load loss, edge importance, and node importance. It then obtains the importance of branches through weighted summation and provides refined risk warnings based on set thresholds.

[0169] Risk assessment module: Based on the calculation results, assess the risk level of the distribution network, set early warning thresholds, and classify the risk level (such as red, orange, and yellow warnings) to identify potential risk areas and issue early warnings.

[0170] Early warning notification module: Issues early warning notifications based on the risk level and suggests corresponding emergency measures (such as activating backup power). The system automatically notifies relevant emergency personnel for timely response to the risk situation.

[0171] Real-time monitoring and optimization module: Monitors the operation of the distribution network in real time and iteratively optimizes the model based on new data. It dynamically adjusts the weights of nodes and edges to adapt to weather changes and special events, improving the model's responsiveness and accuracy.

[0172] Historical data management module: Manages and utilizes historical fault data for model training and optimization to improve the model's generalization ability and prediction accuracy. Stores and analyzes historical data to help improve the system's responsiveness to unexpected situations.

[0173] In another embodiment of the present invention, a terminal device is provided, comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to achieve a corresponding method flow or corresponding function. The processor described in this embodiment of the present invention can be used in the operation of a refined early warning method for risks in overhead power distribution lines.

[0174] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory). This computer-readable storage medium is a memory device in a terminal device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and extended storage media supported by the terminal device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device.

[0175] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the method for refined early warning of risks of overhead lines in the distribution network in the above embodiments; one or more instructions in the computer-readable storage medium are loaded and executed by a processor.

[0176] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0177] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0178] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0179] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0180] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the implementation methods of the present invention, and should be understood that the scope of protection of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of the present invention.

Claims

1. A refined early warning method for risks in overhead lines of a power distribution network, characterized in that, Includes the following steps: S1. Collect meteorological data from different typical areas of the power distribution network. The meteorological data includes, but is not limited to: wind speed, maximum wind speed, temperature, humidity, wind direction, and rainfall. Collect topological overhead line parameter data for the corresponding area, preprocess the data, including: null value interpolation, averaging of repeated observations, and deleting duplicate data, and establish a dataset; S2. Construct a fault probability prediction model for overhead power lines in the distribution network based on time-series data; calculate the fault probability using a mathematical probability model based on overhead line parameter data and meteorological observation data; correct the mathematical fault probability model using actual power grid fault data; construct an overhead line fault probability network model by fusing static topology and dynamic meteorological data; train the overhead line fault probability network model based on the dataset constructed in step S1 to obtain a fault probability prediction model; calculate the probability of each overhead line failing under specific meteorological conditions and topology using the fault probability prediction model. S3. Based on the distribution network topology, model the importance of nodes and lines in the distribution network; construct an undirected graph between different nodes and lines in the distribution network, considering the key roles of node type and overhead lines in the topology, and calculate the importance of lines and nodes; where node importance includes: node type, load level, and the load loss ratio in the topology when a node fails; The importance of a line is determined by considering the connectivity of the network topology and the types of nodes connected to it. Node types include loads, main power sources, and distributed power sources. The importance of a node is then aggregated from its neighboring nodes and added to the importance of the edge. S4. Combine the fault probability of each line predicted in S2 with the importance and load loss ratio of the overhead line obtained in S3 to calculate the overhead line risk parameter index; divide the overhead line risk index into reasonable sections according to the set early warning threshold, and specify early warning strategies according to the segmentation of the risk index.

2. The method for refined early warning of risks in overhead lines of power distribution networks according to claim 1, characterized in that: S1 specifically includes: S11: Collect meteorological time-series datasets for different typical regions in the distribution network, including time, region, wind speed v, temperature t, wind direction angle θ, humidity, precipitation, and maximum wind speed; collect distribution network topology data and line parameter data for the region, and map these data to the topology diagram of IEEE 33 nodes; S12: Clean and preprocess the meteorological data; first, delete duplicate data, and replace the observations at the same location and within the same time step with the average value; fill in the missing values ​​using bilinear interpolation; group the data in the same area according to the continuity of time series to construct the dataset required in step S2.

3. The method for refined early warning of risks in overhead lines of power distribution networks according to claim 1, characterized in that: S2 includes: S21: Construct a mathematical model for the probability of faults in overhead power lines of the distribution network, and calculate the probability of faults in overhead lines under different meteorological data; based on the actual situation and mechanical performance characteristics of conductors in overhead transmission lines, and for the sake of simplifying the line components, the conductors are regarded as hinges, the resistance changes caused by the changes in conductor length are ignored, and it is assumed that the load of the conductor is uniformly distributed along its length, so a parabolic curve is used to calculate the relevant parameters; under the action of horizontal wind v, the wind load along the line direction is ignored, and the horizontal wind load W1 perpendicular to the conductor direction on each span of the conductor is: W1=0.625α1μ1N ′ LDv 2 sin 2 θ×10 -3 Where: α1 is the conductor wind pressure non-uniformity coefficient; μ1 is the conductor shape coefficient; N ′ W1 is the number of phase conductor splits; L is the length of the conductor under wind load; D is the outer diameter of the conductor; θ is the angle between the wind direction and the conductor direction; in plains or large-scale topographical studies, generally only W1 is considered; however, on hillsides, the vertical wind load F on the conductor should be considered. v for: In the formula: v v The upward wind speed along the vertical line; G=N ′ m0gL In the formula: m0 is the mass of the conductor per unit length; g is the acceleration due to gravity; Under wind load, the total load Q and total specific load γ of this span of conductor are respectively: Let the original length of the conductor at manufacturing temperature t0 be L0, and the length of the conductor under wind load be L. Where: σ av α is the average stress of the conductor; E is the elastic modulus of the conductor; α t t is the coefficient of thermal expansion of the conductor; m Temperature under wind load; Based on the principle that the original length of the conductor is equal under windless and wind-loaded conditions, the horizontal stress σ0 of the conductor under wind load is obtained; for conductors in continuous spans, the influence of insulator offset on the height difference angle between suspension points is ignored; the horizontal distance l from the lowest point of the conductor to the higher tower is calculated. m for: In the formula, l is the span of the line, and β is the elevation difference angle between the suspension points of the line; The maximum stress σ of this conductor m The exponential function was used to fit the fault rate P of transmission lines under wind load overload. i With the maximum stress σ of the conductor m Relationship between wind speed v and typhoon speed: In the formula P i The probability of a utility pole failing is σ. s The ultimate stress that the conductor can withstand is σ0, which is related to the design stress of the conductor and the safety factor μ2 under wind load. K1 and T1 are constants related to the line parameters, and C1 is a constant related to the wind speed parameter. S22: Construct a deep neural network model for fault probability prediction. This network model comprehensively considers overhead line parameter data and time-series meteorological data. Feature extraction method: Time-series data is extracted through a bidirectional LSTM network, and overhead line parameter data is processed by embedding. Finally, the two feature vectors are concatenated into one feature vector. The fault probability in S21 is used as the ground truth for training the network model. The fault probability prediction model is trained.

4. The method for refined early warning of risks in overhead lines of power distribution networks according to claim 1, characterized in that: The S3 section describes the indicators for the importance of nodes and lines in the distribution network based on the static topology. S31: Analyze node type weights, including power supply, primary load, secondary load, tertiary load, and important factor vectors; S32: Analyze the importance indicators of nodes in the topology, specifically including degree centrality, proximity centrality, betweenness centrality, and eigenvector centrality; Degree centrality: In the formula, deg(v) is the degree of node v, that is, the number of edges connected to node v, and n is the total number of nodes in the network. Proximity centrality: Where d(v,u) is the shortest path length from node v to node u, and the summation is performed on all nodes u that are different from v; Betweenness centrality: Where σ st σ is the number of shortest paths from node s to node t; st (v) is the number of shortest paths through node v; the summation is performed on all pairs of nodes (s,t), where s and t are not equal to v to avoid self-loops; Eigenvector centrality; Let A be the adjacency matrix of the network, and w be the eigenvector, where w i Let represent the eigenvector centrality score of node i; the eigenvector centrality is calculated as follows: w=λA w Where A is the adjacency matrix of the network, and w is the eigenvector, where w i Let λ represent the eigenvector centrality score of node i, and λ be the largest eigenvalue of matrix A. S33: Considering the importance of lines, line types include lines connected to the power source, important shadow vectors, and line lengths; for the importance index of edges in static topology, the analysis includes edge betweenness centrality, edge compact centrality, and eigenvector centrality; Edge betweenness centrality: Where σ st σ is the total number of shortest paths from node s to node t; st (u,v) is the number of shortest paths from s to t that pass through edge (u,v); the summation is performed on all pairs of nodes (s,t), where s and t are not equal to v to exclude self-looping and repeating edges. Edge-tight centrality: Among them, C C (u) and C C (v) represents the proximity centrality of the two endpoints u and u' at the two ends of the edge, respectively; Eigenvector centrality: Calculate the eigenvector corresponding to its largest eigenvalue, and then normalize it so that its sum is 1; the normalization method is Min-Max normalization, which maps the data to the interval [0,1]; Where x is the original data, Min is the minimum value of the data, Max is the maximum value of the data, and x′ is the transformed data.

5. The method for refined early warning of risks in overhead lines of power distribution networks according to claim 4, characterized in that: S3 further includes: constructing an undirected graph of different nodes and different lines in the distribution network, analyzing the node type weights, and then calculating the importance and correlation of lines and nodes. The probability of failure predicted by meteorological data is denoted as P, which itself ranges from 0 to 1; Analyze a branch in a distribution network and calculate its edge importance and node importance. Node importance C1: is composed of the weighted sum of degree centrality, proximity centrality, betweenness centrality, and eigenvector centrality; The calculated C D (v),C C (v),C B (v) and w are normalized, and then summed to obtain the node importance C1, as shown in the following formula: C1=C D (v)+C C (v)+C B (v)+w Edge importance C2: consists of the weighted sum of edge betweenness centrality, edge compactness centrality, and eigenvector centrality; Calculate C B (u,v),C E (u,v)w1 is normalized, and then summed to obtain the importance C2 of the edge, as shown in the following formula: C2=C B (u,v)+C E (u,v)+w1 Node load loss C3: This represents the proportion of load lost in the circuit relative to the total load of the distribution network when a node fails, as shown in the following formula: In the formula L i The load occupied by node i, L s Refers to the total load of the distribution network; Therefore, the summation of the above weight indicators after normalization gives the importance C of the branch and node: C=P+C1 ′ +C2 ′ +C3 ′ Based on the aforementioned importance indicators, refined threshold warnings can be implemented.

6. The method for refined early warning of risks in overhead lines of power distribution networks according to claim 5, characterized in that: In S4: S41: Adjust the importance C using the following formula: C=P+ω1C1 ′ +ω2C2 ′ +ω3C3 ′ +C0 Where ω1, ω2, and ω3 are weight coefficients, and C0 is the subjective weight; S42: Design a graded early warning mechanism, set different levels of early warning signals according to the risk level, and take corresponding countermeasures; when a high-level early warning occurs, the system can automatically notify relevant emergency personnel and suggest starting backup power or taking emergency measures. S43: Use historical fault data for model training and optimization to improve the model's ability to respond to emergencies; monitor the operation of the distribution network in real time and feed new data back into the model for iterative optimization to continuously improve model performance; S44: Dynamically adjust the weight of each node in the distribution network based on real-time data; at the same time, dynamically adjust the early warning level of the distribution network, taking into account the special needs of special events and peak holiday periods.

7. A refined early warning system for risks in overhead power distribution lines, characterized in that: This system can be used to implement the refined early warning method for risks of overhead lines in power distribution networks as described in any one of claims 1 to 6, specifically including: Data collection module: Collects meteorological data and electrical parameters of the power distribution network, and records the geographical location of the power distribution network lines to obtain complete data input; Data preprocessing module: Cleans and preprocesses the collected data, including removing duplicate data, interpolating to handle null values, and standardizing data; Topology Analysis Module: Based on the geographical information and line parameters of the distribution network, construct the static topology structure of the overhead lines of the distribution network and generate a static topology map of the distribution network; Model building module: Based on distribution network parameters and meteorological time series data, a refined risk early warning model is built; this includes establishing a fault probability prediction model and an importance analysis model for nodes and edges to assess the risks of the distribution network; The calculation and analysis module performs weight normalization based on the fault prediction probability, node load loss, edge importance, and node importance; it obtains the importance of branches through weighted summation and provides refined risk warnings based on set thresholds. Risk assessment module: Based on the calculation results, assess the risk level of the distribution network, set early warning thresholds, classify risk levels, identify potential risk areas, and issue early warnings; Early warning notification module: Issues early warning notifications based on risk levels and suggests corresponding emergency measures; the system automatically notifies relevant emergency personnel for timely response to risk situations; Real-time monitoring and optimization module: Monitors the operation status of the distribution network in real time, iteratively optimizes the model based on new data, and dynamically adjusts the weights of nodes and edges to adapt to weather changes and special events; Historical data management module: manages and utilizes historical fault data for model training and optimization to improve the model's generalization ability and prediction accuracy; stores and analyzes historical data to help improve the system's response to emergencies.

8. A computer device, characterized in that: The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the refined early warning method for risk of overhead lines in a power distribution network as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: It stores a computer program that, when executed by a processor, implements the refined early warning method for risks of overhead power distribution lines as described in any one of claims 1 to 6.

Images