Line intelligent planning and fault early warning method suitable for electric power engineering design

Abstract

The application discloses a line intelligent planning and fault early warning method suitable for power engineering design and relates to the technical field of power engineering. The method comprises the following steps: through spatial semantic segmentation and a cost surface model, quantifying geographic information, geological conditions and construction cost into a continuous decision surface, combining a dynamic clustering algorithm to identify a large-area low-cost area as a primary feasible region; based on spatial features, geological stability and facility correlation data, constructing multi-dimensional decision indicators, and through real-time correlation coefficients, identifying "fault regions" such as terrain mutations or facility conflicts, and triggering an abnormal processing mechanism; selecting a seed region with the optimal construction condition as an anchor point, constructing a joint probability model of spatial features and cost, combining an expectation maximization algorithm and graph optimization technology, and generating an optimal path that takes into account the feasibility probability and cost benefit. The application improves the scientificity and economy of cable planning under complex geological conditions and provides key technical support for the intelligentization of power engineering.

Description

Technical Field

[0001] This invention belongs to the field of power engineering technology, specifically relating to a method for intelligent line planning and fault early warning applicable to power engineering design. Background Technology

[0002] In the field of power engineering, the planning and fault early warning of underground cable lines have long faced the following technical bottlenecks: Traditional planning methods are extensive and inefficient: relying on manual experience or single GIS analysis makes it difficult to comprehensively consider multi-dimensional factors such as topography, geology, and conflicts with underground facilities, resulting in strong subjectivity in route selection and difficulty in quantifying and assessing construction risks and costs. Static zoning strategies cannot dynamically adapt to changes in engineering conditions. For example, when new underground facilities are added during construction or geological exploration data is updated, re-survey and planning are required, which is inefficient.

[0003] Insufficient multi-source data fusion capabilities: Geographic information, geological exploration, facility distribution, and other data are typically stored independently, lacking semantic association and in-depth data mining. For example, traditional methods may only use geological data to avoid undesirable areas without establishing a quantitative correlation with construction costs and maintenance difficulties. Existing semantic segmentation technologies (such as remote sensing image classification) can only achieve coarse regional division and cannot accurately identify mixed areas that are "partially suitable / partially unsuitable," leading to either conservative or aggressive feasible regional division.

[0004] The lack of a dynamic risk early warning mechanism: There is a disconnect between route planning and fault early warning. Existing technologies mostly focus on post-construction monitoring, making it difficult to predict construction risks in intermediate areas (such as areas with abrupt terrain changes or dense infrastructure) during the planning stage, resulting in high costs for subsequent route rerouting. Traditional decision-making models (such as cost-benefit analysis) only consider single indicators and ignore the coupling effects of multiple factors (such as the interaction between geological stability and facility distance), easily overlooking hidden risks.

[0005] The optimization algorithm suffers from poor adaptability: the path search algorithm lacks a real-time feedback mechanism, requiring recalculation of the global path when construction conditions change, making local optimization impossible. Probabilistic model parameters (such as feasibility weights) are typically preset to fixed values, making it difficult to adaptively adjust based on regional characteristics, leading to optimization results that deviate from actual needs.

[0006] This invention addresses the aforementioned pain points by proposing the following solutions: Multimodal data intelligent fusion: Through spatial semantic segmentation and cost surface models, discrete multi-source data (GIS, remote sensing, geological reports) are transformed into continuous decision surfaces, achieving a leap from "data silos" to "collaborative analysis." Dynamic clustering and adaptive weights: An iterative adjustment mechanism for clustering weights is introduced to automatically identify large-area, low-cost regions. Regional coordination is quantified through correlation coefficients, solving the problem of traditional clustering algorithms easily getting trapped in local optima. Real-time risk warning embedded in planning: The difference in decision indicators between adjacent sub-regions is transformed into correlation coefficients, identifying fault areas during the planning stage and triggering anomaly handling processes (such as data verification or model adjustment) to achieve risk feedforward control. Dual-probability model coupled with graph optimization: A joint probability model of spatial features and costs is constructed, and correlation coefficients are integrated into the graph search algorithm through edge weight design, ensuring global optimality of the path while improving local adaptability. This solution, through a data-driven intelligent decision-making system, significantly improves the scientific and economic efficiency of cable route planning under complex geological conditions, providing key technical support for the intelligent transformation of power engineering. Summary of the Invention

[0007] To overcome the shortcomings and deficiencies of the existing technology, the present invention adopts the following technical solution: The process of a line intelligent planning and fault early warning method applicable to power engineering design is as follows: Step 1: Preliminary Delineation of Feasible Areas: S11. Obtain the underground pipeline network area and divide it into multiple adjacent sub-regions, collecting spatial feature data, geological feature data, and facility association data for each sub-region; S12. Construct a spatial semantic segmentation model and a cost surface model. The spatial semantic segmentation model divides the area into areas suitable for cable laying, areas unsuitable for cable laying, and areas with existing underground facilities, and encodes them. The cost surface model calculates cost weights based on terrain, geology, and facility distances; S13. Integrate the spatial semantic segmentation results, cost weights, and spatial location coordinates to form a comprehensive feature vector. Adjust the clustering using a Gaussian mixture model to obtain the initial feasible areas. The clustering adjustment includes calculating cluster... Class weighting: Clusters with cluster weights greater than or equal to a preset weight threshold are marked as large clusters, and clusters with weights less than the preset weight threshold are marked as small clusters. All sub-regions within a small cluster are traversed. For each sub-region belonging to a small cluster, the distance between the sub-region and all large clusters is calculated, and the sub-region is assigned to the large cluster with the smallest distance. Then, the recombined large clusters are iterated again. S14: Calculate the spatial characteristic decision index, geological characteristic decision index, and facility association decision index for each sub-region. S15: Calculate the real-time correlation coefficient based on the spatial characteristic decision index, geological characteristic decision index, and facility association decision index of adjacent sub-regions, generate fusion decision indexes, and perform anomaly handling. Step 2, refine the determination of feasible cable channels: S21, select seed areas with good construction conditions, low cost and low risk from the primary feasible areas; S31. Construct a feasibility probability model based on spatial characteristics and a cost probability model based on cost; S41-S42. Combine the two probability models to obtain a comprehensive probability value, optimize using the expectation-maximization algorithm, and determine the final feasible cable channel.

[0008] Preferably, the spatial semantic segmentation model in S12 adopts a deep learning model, encoding the semantic labels of sub-regions as: suitable for cable laying area 1, unsuitable for cable laying area 0, and existing underground facility area 2; the cost weight formula is: ;in, This is the normalized slope value; For geological stability scoring, ; Distance to facilities; Spatial location coordinates; , , Preset weights.

[0009] Preferably, the clustering adjustment step in S13 includes: Calculate the average cost weight of all sub-regions within the cluster. ;in, , This is an adjustment factor used to balance the importance of area and cost; Set a weight threshold to divide the clusters into large and small clusters, assign the small cluster sub-regions to the large clusters with the smallest distance, and iterate again.

[0010] Preferably, the calculation of decision indicators in S14 includes: Spatial Feature Decision Calculation: Based on spatial feature data and the spatial feature decision model, the first decision index corresponding to the sub-region is obtained. The spatial feature decision model is as follows: ;in, It is the primary decision-making indicator; The spatial characteristic value at time t; For the feature values ​​of the target space; Geological Feature Decision Indicators: Based on geological feature data and the geological feature decision model, the second decision indicators corresponding to the sub-regions are obtained. The geological feature decision model is as follows: ;in, As the second decision indicator; Let be the geological characteristic value at time t; The target geological characteristic value; Facility-related decision indicators: Based on facility-related data and the facility-related decision model, the third decision indicator corresponding to the sub-region is obtained. The facility-related decision model is as follows: ;in, As the third decision indicator; The facility association value at time t; This is the value associated with the target facility.

[0011] Preferably, the real-time correlation coefficient in S15 is calculated as follows: The first difference is obtained based on the first decision index of adjacent sub-regions. The second difference is obtained based on the second decision index of the adjacent sub-regions. The third difference is obtained based on the third decision index of the adjacent sub-regions. First difference Second difference and the third difference All have been normalized. , , ; , include , and ; Based on the association model Get real-time correlation coefficient ,in, , , For adjustment coefficients, correlation coefficients The closer it is to 1, the more coordinated the adjacent sub-regions are in terms of engineering attributes.

[0012] Preferably, the seed region selection step in S21 includes: screening sub-regions with semantic label 1, performing K-means clustering based on the comprehensive score of spatial features, geological features, cost weights and facility associations, extracting cluster centroids as seed regions, and considering normalized decision indicators and cost weights in the comprehensive score.

[0013] Preferably, the feasibility probability model based on spatial characteristics considers terrain slope, geological stability and facility distance, while the cost probability model based on cost considers construction cost and maintenance cost, and the model parameters are determined by historical data or expert weights.

[0014] Preferably, the comprehensive probability value is calculated as follows: ;in, Let be the feasibility probability of subregion i; Let be the cost probability of subregion i. Weights for spatial features, The weights for cost characteristics, and .

[0015] Preferably, a graph model is constructed based on the comprehensive probability value and real-time correlation coefficient, and optimized using the expectation-maximization algorithm to determine the final feasible cable channel, including: A graph model is constructed using sub-regions as nodes and the connections between adjacent sub-regions as edges. Edge weights are determined based on comprehensive probability values ​​and real-time correlation coefficients, employing either Dijkstra's algorithm or A*. The algorithm finds the path with the minimum total weight, iterates and adjusts the weights until convergence, and outputs the final feasible cable channel.

[0016] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: 1. This invention introduces a deep learning model for spatial semantic segmentation and combines it with a cost surface model to quantitatively evaluate geographical, geological, and construction cost factors, thus achieving a shift from traditional experience-driven to data-driven approaches. This method can automatically identify suitable areas for cable laying, avoiding the subjectivity and inefficiency of human judgment, and significantly improving the intelligence level and scientific decision-making in power engineering design.

[0017] 2. After initially delineating feasible areas, this invention further integrates and analyzes sub-regions using multi-dimensional decision indicators (including spatial characteristics, geological characteristics, and facility association characteristics), and introduces real-time correlation coefficients to assess the coordination between adjacent regions, thereby effectively identifying potential "fault areas" or high-risk connection points. This mechanism not only improves the continuity and consistency of the route but also enhances the safety and stability of the construction process and reduces the risks of later maintenance.

[0018] 3. This invention constructs a spatial feature-based probabilistic model and a cost-based probabilistic model, and employs the expectation-maximization algorithm and graph model optimization technique to search for the optimal connectivity path starting from the seed region. This ensures that the selected cable channel meets both engineering feasibility requirements and construction and maintenance costs. This method can efficiently find the globally optimal path in complex geographical environments, achieving an organic balance between safety, feasibility, and economy in power engineering design. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. 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.

[0020] Figure 1 A flowchart of the intelligent line planning and fault early warning method applicable to power engineering design according to the present invention is shown; Figure 2This diagram illustrates a flowchart of the preliminary division of feasible regions according to the present invention; Figure 3 A flowchart illustrating the present invention for determining the final feasible cable path is shown. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more exemplary embodiments. Numerous specific details are provided in the following description to give a full understanding of exemplary embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced with one or more of the specific details omitted, or other methods, components, steps, etc., can be employed. In other instances, well-known structures, methods, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0023] Example 1: See Figure 1 As shown in the figure, the intelligent line planning and fault early warning method applicable to power engineering design in this embodiment has the following process: Step 1: Initially delineate the feasible area.

[0024] See Figure 2 As shown, the specific process for initially delineating feasible areas is as follows: S11. Data collection and feature extraction.

[0025] The underground pipeline network area is acquired and divided into multiple sequentially adjacent sub-regions. Data is then collected for each sub-region. These data sources include, but are not limited to, Geographic Information Systems (GIS), satellite remote sensing images, geological exploration reports, topographic data, geological condition data, and existing underground facility distribution data.

[0026] S12. Construct a spatial semantic segmentation and cost surface model.

[0027] Spatial semantic segmentation: A deep learning model is used to perform spatial semantic segmentation on the underground pipeline network area. By training on a large amount of labeled geospatial data, the model can automatically divide the area into different semantic categories, including areas suitable for laying cables, areas unsuitable for laying cables, and areas with existing underground facilities. Each sub-region corresponds to a semantic label: areas suitable for laying cables are coded as 1, areas unsuitable for laying cables are coded as 0, and areas with existing underground facilities are coded as 2.

[0028] Cost surface model: A cost surface model is constructed based on a geographic information system. It comprehensively considers factors such as terrain undulation, geological conditions, and construction difficulty, assigning different cost weights to different locations. For example, areas with steep terrain and complex geology have higher construction costs and correspondingly larger cost weights; while areas with flat terrain and stable geology have lower construction costs and smaller cost weights.

[0029] Define the cost weight formula: ; in, This is the normalized slope value; For geological stability scoring, ; Distance to facilities; Spatial location coordinates; , , Preset weights.

[0030] Slope refers to the degree of inclination of terrain, usually expressed as an angle or percentage. In power engineering design, steeper terrain increases construction difficulty and cost because more civil engineering work is required to ensure the safety and stability of cables. Therefore, in the cost weighting formula, the greater the slope value, the higher the corresponding cost weight.

[0031] A geological stability score is a quantitative evaluation of the geological conditions of a specific area, taking into account factors such as soil type, rock structure, and groundwater conditions. A high score indicates that the area has favorable geological conditions, suitable for infrastructure construction such as cable laying; while a low score may indicate the risk of geological hazards such as landslides and subsidence, making it unsuitable for related construction activities. Therefore, the lower the geological stability score, the higher its corresponding cost weight, reflecting higher construction costs and risks.

[0032] Facility distance refers to the distance from the planned cable laying location to the nearest existing underground infrastructure. This distance affects the number and complexity of obstacles that may be encountered during construction. A shorter distance may increase construction difficulty because it requires more precise operations to avoid damaging existing underground infrastructure, such as water pipes, gas pipelines, or other cables. Therefore, the shorter the facility distance, the higher its corresponding cost weight.

[0033] S13, Clustering of primary feasible regions.

[0034] Feature fusion: The spatial semantic segmentation results and the cost surface model results are fused to obtain a comprehensive feature vector. This feature vector contains spatial location, semantic category, and cost information. Specifically, it integrates the spatial coordinates of each sub-region. Semantic category encoding (0 / 1 / 2), cost weight They are concatenated into a single comprehensive feature vector.

[0035] Clustering algorithms: These algorithms cluster composite feature vectors, calculating the weight of each cluster at regular intervals during the clustering process. The weights are calculated based on the area of ​​the clustered regions and construction costs. For example, clusters with larger areas and lower construction costs tend to have higher weights.

[0036] Specifically, we set the initial number of clusters K, each cluster follows a multidimensional Gaussian distribution, and randomly initialize the mean of each Gaussian distribution. Covariance matrix and prior probability ; Calculate the posterior probability of each sub-region belonging to each cluster. , The mean, covariance, and prior probability are updated based on the posterior probability to minimize the distribution difference between the data and the model.

[0037] Weight Calculation: At fixed iteration intervals, pause clustering and calculate the weights of all current clusters. Calculate the total area of ​​the sub-regions within each cluster (geometric area can be calculated via GIS overlay), and then calculate the average cost weight of all sub-regions within the cluster. ;in, , This is an adjustment factor used to balance the importance of area and cost, such as prioritizing large, low-cost areas.

[0038] The area of ​​cluster k refers to the total area of ​​all sub-regions contained in the k-th cluster in cluster analysis. It can be obtained by overlaying geometric areas using GIS (Geographic Information System). It reflects the spatial size of the cluster and is an important indicator for measuring cluster size. In weight calculation, it is used to reflect the impact of cluster size on weights.

[0039] The k-average cost weight of clustering is the cost weight of all sub-regions within the k-th cluster. The average value. Cost weight. Related to factors such as construction costs and resource consumption in sub-regions, this average value reflects the average cost level of sub-regions within the k-th cluster, embodying the cost characteristics of the cluster. In weight calculation, it is used to measure the impact of cluster cost on weights.

[0040] The total area represents the area of ​​the entire study area (i.e., the sum of the areas covered by all clusters). As a reference, it is compared with the area of ​​cluster k when calculating cluster weights to determine the proportion of the area of ​​cluster k in the total area, and thus participates in the weight calculation to measure the size proportion of the cluster in the overall area.

[0041] Clustering Adjustment: Based on a weight threshold, clusters with weights greater than or equal to the threshold are labeled as large clusters, and those with weights less than the threshold are labeled as small clusters. All sub-regions within a small cluster are traversed. For each sub-region belonging to a small cluster, its distance to all large clusters is calculated, and the sub-region is assigned to the large cluster with the smallest distance. The recombined large clusters are then iterated again until the clustering results converge. This clustering adjustment eliminates unreasonable small-scale region partitioning.

[0042] S14. Calculate decision indicators for each sub-region after clustering.

[0043] For each sub-region, the following three types of data are obtained (after clustering, the data has been divided into primary feasible regions): Spatial feature data: terrain slope, elevation, standard deviation of terrain relief, etc.

[0044] Geological characteristic data: soil type (such as clay, sand, which can be coded as numerical value), geological stability score (1-10 points, with 10 points being the most stable), groundwater level depth, etc.

[0045] Facility association data: horizontal distance to the nearest underground facility, intersection angle, existing facility type (such as gas pipeline, drainage pipeline, coded as conflict risk level), etc.

[0046] Spatial Feature Decision Calculation: Based on spatial feature data and the spatial feature decision model, the first decision index corresponding to the sub-region is obtained. The spatial feature decision model is assumed to be: ;in, It is the primary decision-making indicator; The spatial characteristic value at time t (such as terrain slope); For target spatial characteristic values ​​(such as the slope suitable for laying cables).

[0047] Geological Feature Decision Indicators: Secondary decision indicators are obtained for each sub-region based on geological feature data and the geological feature decision model. The geological feature decision model is assumed to be: ;in, As the second decision indicator; The geological characteristic value at time t (such as the stability index of soil type); The target geological characteristic value.

[0048] Facility-related decision indicators: Third decision indicators corresponding to sub-regions are obtained based on facility-related data and the facility-related decision model. The facility-related decision model is assumed to be: ;in, As the third decision indicator; The facility association value at time t (such as the distance to other underground facilities); This is the value associated with the target facility.

[0049] S15, Judgment and Integration.

[0050] Real-time correlation coefficient calculation: The real-time correlation coefficient of adjacent sub-regions is obtained based on the first, second, and third decision indicators of the adjacent sub-regions. Specifically: The first difference is obtained based on the first decision index of adjacent sub-regions. The second difference is obtained based on the second decision index of the adjacent sub-regions. The third difference is obtained based on the third decision index of the adjacent sub-regions. .

[0051] First difference Second difference and the third difference All have been normalized. , include , and ; , , Based on the association model Obtain real-time correlation coefficient ,in, , , It is the adjustment coefficient, and Correlation coefficient The closer the correlation coefficient is to 1, the more coordinated the adjacent sub-regions are in terms of engineering attributes (e.g., consistent slope, similar geological stability, and compliant distance from existing facilities); Used to identify "fault areas": If the correlation coefficient between adjacent sub-regions is too low (such as abrupt changes in terrain or large differences in geological conditions), it indicates that there may be construction risks at the junction of the two (such as uneven settlement or sudden changes in excavation difficulty), which need to be checked carefully.

[0052] Judgment and fusion decision indicator generation: Determine whether all multiple real-time correlation coefficients are greater than the preset threshold. If so, obtain the fusion decision indicator corresponding to the sub-region based on the fusion model, the first decision indicator, the second decision indicator, and the third decision indicator.

[0053] The fusion model operates as follows: The first, second, and third decision indicators are normalized to generate a first string, a second string, and a third string, respectively; the feature characters of the first string are extracted as the first character, the feature characters of the second string as the second character, and the feature characters of the third string as the third character; the first, second, and third characters are then concatenated sequentially to form the fused decision indicator. This integrates the spatial, geological, and facility-related decision indicators (…). , , After normalization, a fusion index is generated by extracting and concatenating string features (for example, rounding the normalized value to an integer and concatenating it into the form of "8-6-9").

[0054] Anomaly Handling: If not all real-time correlation coefficients are greater than the preset threshold, a fusion anomaly strategy is triggered, such as re-checking the accuracy of data collection and adjusting the decision model. By determining whether the real-time correlation coefficients of adjacent sub-regions are all greater than the threshold, the existence of "fault areas" can be identified. For example, if the geological stability scores of two adjacent sub-regions differ significantly, their real-time correlation coefficients may be lower than the threshold, indicating a risk of uneven settlement at the junction. If such faults are not eliminated, they may cause the cable to break due to abrupt changes in geological conditions after laying. Fusion decision indicators are only generated when all adjacent sub-regions meet the condition that their real-time correlation coefficients are greater than the preset threshold. This ensures that the final path consists of continuous, low-risk areas, avoiding the failure of the overall scheme due to local inconsistencies.

[0055] Step 2: Determine the feasible cable route.

[0056] Seed Region Selection: Several seed regions are selected from each primary feasible region. These seed regions are typically locations within the region with relatively good construction conditions and lower costs. Representative locations can be selected as seed regions by evaluating the comprehensive characteristics of the primary feasible regions.

[0057] From the initial feasible areas, select the core areas with the best construction conditions, lowest cost, and least risk, and use them as the starting point for subsequent detailed planning (similar to "anchor points" to reduce the search space).

[0058] Step 3: Construct a probability model.

[0059] A probabilistic model based on spatial characteristics is constructed to calculate the feasibility probability of laying cables in different locations, taking into account factors such as terrain, geology, and distance from other underground facilities. For example, locations with flat terrain, stable geology, and distance from other underground facilities have a higher feasibility probability. The model output is a probability value between [0,1], representing the confidence level that the sub-region meets the spatial feasibility requirements. .

[0060] Cost-based probabilistic model: Considering factors such as construction and maintenance costs, a probabilistic model is constructed to calculate the cost probability of laying cables at different locations. Locations with low construction and maintenance costs have a higher cost probability. The model output is a probability value between [0,1], representing the feasibility probability of a sub-region in terms of cost. .

[0061] Step 4: Determine the final feasible cable route by combining the probability model.

[0062] See Figure 3 As shown, the specific process for determining the final feasible cable channel using a probabilistic model is as follows: S41. The feasibility probability of the seed region is obtained using a probability model based on spatial features. Cost probabilities obtained from cost-based probabilistic models By combining these factors, a comprehensive probability value is obtained. , ;in, Let be the feasibility probability of subregion i; Let be the cost probability of subregion i. Weights for spatial features, The weights for cost characteristics, and .

[0063] S42. Optimize the overall probability value using the expectation-maximization algorithm to determine the final feasible cable route.

[0064] Specifically, setting initial weights , Treating sub-regions as nodes, adjacent sub-regions are connected by edges, with the edge weights designed as follows: ;in This refers to the correlation coefficient between adjacent sub-regions calculated in step one (the larger the value, the better the coordination).

[0065] Based on the current weight and Calculate the overall probability value for each sub-region. ; Constructing a graph model: If the correlation coefficient between adjacent sub-regions is... If the edge weight is less than a set threshold, set the edge weight to infinity. Otherwise, the edge weight is... This ensures that sub-regions with high probability and high coordination are more likely to be selected.

[0066] Use Dijkstra's algorithm or A The algorithm starts from the seed region and finds the path with the minimum total weight that connects all target regions. If there are multiple seed regions (such as start point, end point, and transfer point), use the multi-source shortest path or Steiner tree algorithm to generate connected paths. Use the M-step method to optimize parameters, determine convergence, and output the final path.

[0067] The beneficial effects of this embodiment are as follows: the fusion of spatial semantic segmentation and cost model enables intelligent initial screening of construction areas, accurately eliminating high-risk areas and reducing preliminary survey costs; dynamic clustering algorithm combined with weight adaptive adjustment mechanism automatically identifies large-area low-cost areas and optimizes path topology; multi-decision index fusion model quantifies regional coordination through real-time correlation coefficients, effectively avoiding the risks of terrain abrupt changes and facility conflicts; dual-probability model coupled with expectation-maximization algorithm establishes a quantitative balance between construction feasibility and cost-effectiveness, and generates the globally optimal path by combining with graph optimization algorithm. This solution improves the efficiency of cable channel planning and reduces construction risks through a data-driven intelligent decision-making system, providing a reliable solution for power engineering under complex geological conditions.

[0068] The weights in this invention are used to measure the degree of influence of different factors or variables on a particular outcome or decision. Weights are defined as numerical values ​​assigned to each factor when comparing and evaluating multiple factors to reflect their importance or priority. These weights can be determined based on specific circumstances and needs, and are typically developed and confirmed jointly by professionals or relevant stakeholders. By setting appropriate weights, programs or systems can make more accurate decisions or predictions.

[0069] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

[0070] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to specific implementations. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for intelligent line planning and fault early warning applicable to power engineering design, characterized in that, The method is as follows: Step 1: Preliminary Delineation of Feasible Areas: S11. Obtain the underground pipeline network area and divide it into multiple adjacent sub-regions, collecting spatial feature data, geological feature data, and facility association data for each sub-region; S12. Construct a spatial semantic segmentation model and a cost surface model. The spatial semantic segmentation model divides the area into areas suitable for cable laying, areas unsuitable for cable laying, and areas with existing underground facilities, and encodes them. The cost surface model calculates cost weights based on terrain, geology, and facility distances; S13. Integrate the spatial semantic segmentation results, cost weights, and spatial location coordinates to form a comprehensive feature vector. Adjust the clustering using a Gaussian mixture model to obtain the initial feasible areas. The clustering adjustment includes calculating cluster... Class weighting: Clusters with cluster weights greater than or equal to a preset weight threshold are marked as large clusters, and clusters with weights less than the preset weight threshold are marked as small clusters. All sub-regions within a small cluster are traversed. For each sub-region belonging to a small cluster, the distance between the sub-region and all large clusters is calculated, and the sub-region is assigned to the large cluster with the smallest distance. Then, the recombined large clusters are iterated again. S14: Calculate the spatial characteristic decision index, geological characteristic decision index, and facility association decision index for each sub-region. S15: Calculate the real-time correlation coefficient based on the spatial characteristic decision index, geological characteristic decision index, and facility association decision index of adjacent sub-regions, generate fusion decision indexes, and perform anomaly handling. Step 2, refine the determination of feasible cable channels: S21, select seed areas with good construction conditions, low cost and low risk from the primary feasible areas; Step 3: Construct a feasibility probability model based on spatial features and a cost probability model based on cost. Step 4: Combine the two probability models to obtain a comprehensive probability value. Based on the comprehensive probability value and the real-time correlation coefficient, construct a graph model and optimize it using the expectation-maximization algorithm to determine the final feasible cable channel.

2. The intelligent line planning and fault early warning method applicable to power engineering design according to claim 1, characterized in that, The spatial semantic segmentation model in S12 uses a deep learning model to encode the semantic labels of sub-regions as: suitable for cable laying area 1, unsuitable for cable laying area 0, and existing underground facility area 2; the cost weight formula is: ;in, This is the normalized slope value; For geological stability scoring, ; Distance to facilities; Spatial location coordinates; , , Preset weights.

3. The intelligent line planning and fault early warning method applicable to power engineering design according to claim 1, characterized in that, The clustering adjustment steps include: Calculate the average cost weight of all sub-regions within the cluster. ;in, , This is an adjustment factor used to balance the importance of area and cost; Set a weight threshold to divide the clusters into large and small clusters, assign the small cluster sub-regions to the large clusters with the smallest distance, and iterate again.

4. The intelligent line planning and fault early warning method applicable to power engineering design according to claim 1, characterized in that, The calculation of decision indicators in S14 includes: Spatial Feature Decision Calculation: Based on spatial feature data and the spatial feature decision model, the first decision index corresponding to the sub-region is obtained. The spatial feature decision model is as follows: ;in, It is the primary decision-making indicator; The spatial characteristic value at time t; For the feature values ​​of the target space; Geological Feature Decision Indicators: Based on geological feature data and the geological feature decision model, the second decision indicators corresponding to the sub-regions are obtained. The geological feature decision model is as follows: ;in, As the second decision indicator; Let be the geological characteristic value at time t; The target geological characteristic value; Facility-related decision indicators: Based on facility-related data and the facility-related decision model, the third decision indicator corresponding to the sub-region is obtained. The facility-related decision model is as follows: ;in, As the third decision indicator; The facility association value at time t; This is the value associated with the target facility.

5. The intelligent line planning and fault early warning method applicable to power engineering design according to claim 1, characterized in that, The real-time correlation coefficient in S15 is calculated as follows: The first difference is obtained based on the first decision index of adjacent sub-regions. The second difference is obtained based on the second decision index of the adjacent sub-regions. The third difference is obtained based on the third decision index of the adjacent sub-regions. First difference Second difference and the third difference All have been normalized. , include , and ; , , Based on the association model Get real-time correlation coefficient ,in, , , For adjustment coefficients, correlation coefficients The closer it is to 1, the more coordinated the adjacent sub-regions are in terms of engineering attributes.

6. The intelligent line planning and fault early warning method applicable to power engineering design according to claim 1, characterized in that, The seed region selection step in S21 includes: filtering sub-regions with semantic label 1, performing K-means clustering based on the comprehensive score of spatial features, geological features, cost weights and facility associations, and extracting the cluster centroids as seed regions. The comprehensive score takes into account normalized decision indicators and cost weights.

7. The intelligent line planning and fault early warning method applicable to power engineering design according to claim 1, characterized in that, The feasibility probability model based on spatial features considers terrain slope, geological stability, and facility distance, while the cost probability model based on cost considers construction and maintenance costs. The model parameters are determined through historical data or expert weights.

8. The intelligent line planning and fault early warning method applicable to power engineering design according to claim 1, characterized in that, The comprehensive probability value is calculated as follows: ;in, Let be the feasibility probability of subregion i; Let i be the cost probability of subregion i. Weights for spatial features, The weights for cost characteristics, and .

9. The intelligent line planning and fault early warning method applicable to power engineering design according to claim 1, characterized in that, The process of constructing a graph model based on comprehensive probability values ​​and real-time correlation coefficients, and optimizing it using the expectation-maximization algorithm to determine the final feasible cable channel, includes: A graph model is constructed using sub-regions as nodes and the connections between adjacent sub-regions as edges. Edge weights are determined based on comprehensive probability values ​​and real-time correlation coefficients, employing either Dijkstra's algorithm or A*. The algorithm finds the path with the minimum total weight, iterates and adjusts the weights until convergence, and outputs the final feasible cable channel.

Images