A method and system for integrated distribution network planning in power engineering design

Abstract

This application provides a method and system for comprehensive planning of distribution networks in power engineering design. First, it acquires geospatial data, historical operational data, and cable vibration spectrum data monitored by a fiber optic sensor network within the planning area. Then, it uses a mixed-integer programming parallel acceleration algorithm to perform distributed storage and correlation analysis on the geospatial data and cable vibration spectrum data, generating a cable aging risk distribution map based on spatial topology constraints. Next, it uses a second-order cone programming model for multi-objective collaborative optimization, generating high-voltage distribution network line capacity expansion schemes and cable replacement priority sequences. Finally, it combines the partial discharge frequency domain characteristic variation trend in the cable vibration spectrum data to correct the line corridor expansion parameters, outputting a dynamic planning graph including cable life prediction nodes and capacity expansion paths. The technical solution provided in this application achieves accurate planning and optimization of high-voltage distribution networks, improving the accuracy and efficiency of cable life prediction and capacity expansion path planning.

Description

Technical Field

[0001] This application relates to the fields of power system planning and geographic information system technology, and in particular to a method and system for integrated planning of distribution networks in power engineering design. Background Technology

[0002] With the rapid development of new power systems, distribution network planning faces multiple challenges, including the high proportion of distributed energy integration, the surge in electric vehicle charging loads, and increased user demands for power supply reliability. Traditional distribution network planning methods are ill-equipped to effectively address the optimization needs of distributed power output fluctuations, load uncertainties, and complex grid operation scenarios. Therefore, there is an urgent need for a comprehensive distribution network planning method that can consider economy, reliability, environmental protection, and flexibility to achieve coordinated optimization of power sources, grid, load, and storage, and meet the future development goals of efficient, intelligent, and low-carbon power systems.

[0003] Currently, distribution network planning primarily employs traditional methods based on load forecasting, power flow calculation, and GIS zoning, combined with optimization algorithms such as linear programming and integer programming for network design and equipment selection. Some schemes incorporate adaptive analysis for distributed generation (DG) integration, improving grid economics by optimizing the capacity and location of DG. Furthermore, planning methods for flexible interconnection devices are increasingly being adopted to improve power flow distribution and supply reliability. In recent years, new concepts such as grid-based planning and capacity-to-load ratio adjustment have also been incorporated into the planning system to adapt to the complex and ever-changing grid operating environment.

[0004] However, existing methods are mostly based on deterministic models, which make it difficult to accurately reflect the stochasticity of distributed power generation output and load demand, resulting in significant deviations between planning results and actual operation. Traditional solutions often optimize economics or reliability independently, lacking a comprehensive balance among multiple objectives such as economy, reliability, and environmental protection, making it difficult to maximize overall benefits. Existing methods do not adequately consider the rapid recovery capabilities in disaster scenarios and lack support for intelligent interaction between distributed energy resources and the main grid, making it difficult to meet the future development needs of a highly resilient and intelligent power grid. Summary of the Invention

[0005] This application provides a method and system for comprehensive planning of power distribution networks in power engineering design, in order to solve the problems of limited accuracy and practicality in the prior art.

[0006] Firstly, this application provides a method for comprehensive distribution network planning in power engineering design, including:

[0007] Acquire geospatial data, historical operation data, and cable vibration spectrum data monitored by fiber optic sensor networks within the high-voltage distribution network planning area;

[0008] The geospatial data and the cable vibration spectrum data are distributed and analyzed in a parallel computing architecture using a mixed integer programming parallel acceleration algorithm to generate a cable aging risk distribution map based on spatial topology constraints. The spatial topology constraints are dynamically defined by the geospatial relationship between the substation coordinates and the line corridor in the geospatial data.

[0009] Based on the second-order cone programming model, multi-objective collaborative optimization is performed on the cable aging risk distribution map, the load peak-valley difference in the historical operation data, and the line corridor expansion parameters in the geospatial data to generate high-voltage distribution network line capacity expansion schemes and cable replacement priority sequences, respectively. The constraints of the multi-objective collaborative optimization include the real-time threshold of cable joint temperature fluctuation signal and the spatial accumulation effect of partial discharge frequency domain characteristics.

[0010] Based on the high-voltage distribution network line capacity expansion scheme and cable replacement priority sequence, and combined with the partial discharge frequency domain characteristic change trend in the cable vibration spectrum data, the line corridor expansion parameters in the high-voltage distribution network planning area are corrected, and a dynamic planning map including cable life prediction nodes and capacity expansion paths is output.

[0011] Optionally, a collaborative optimization model is constructed with the risk value gradient in the cable aging risk distribution map, the time series deviation of the load peak-valley difference, and the cost coefficient of the line corridor extension parameters as optimization variables.

[0012] Based on the real-time threshold definition of the temperature fluctuation signal of the cable joint, a set of thermodynamic constraint equations is defined, and the energy integral of the partial discharge frequency domain characteristics in the cable vibration spectrum data in the spatial dimension is coupled with the preset safety threshold to generate a spatial accumulation effect inequality constraint.

[0013] The thermodynamic constraint equations and spatial cumulative effect inequality constraints of the collaborative optimization model are relaxed. By introducing auxiliary variables, the spatial topological correlation of the cable aging risk distribution map is mapped to the boundary conditions of the second-order cone programming model. The optimization model containing the relaxed constraints is iteratively solved using a decomposition coordination mechanism to generate a feasible solution set.

[0014] Based on the optimal Pareto front of the feasible solution set, a high-voltage distribution network line capacity expansion scheme is generated. At the same time, based on the risk value gradient direction of the cable aging risk distribution map and the activity index of the spatial cumulative effect inequality constraint, the line impedance parameters in the feasible solution set are sorted by sensitivity to generate a cable replacement priority sequence.

[0015] Optionally, the inequality constraints related to spatial cumulative effects in the thermodynamic constraint equation set are relaxed by introducing auxiliary variables, and the auxiliary variables are quantified to determine the constraint deviation and construct a dynamic relaxation boundary.

[0016] The spatial topological correlation of the cable aging risk distribution map is mapped to the boundary conditions of a second-order cone programming model, and the coupling parameter matrix is ​​defined by the topological adjacency relationship.

[0017] Based on the constraint relationship between the coupling parameter matrix and the dynamic relaxation boundary, an alternating iterative update mechanism of main variables and auxiliary variables is adopted. In the main variable update stage, the thermodynamic state parameters are solved under the fixed relaxation boundary. In the auxiliary variable update stage, the deviation and penalty coefficient of the relaxation boundary are dynamically adjusted according to the current main variable residual and the topological connection strength until the residual converges to the preset threshold.

[0018] During the iteration process, intermediate solutions that satisfy the constraints of the dynamic relaxation boundary and the coupling parameter matrix are simultaneously screened. Solutions that conflict with spatial topology are eliminated by non-dominated sorting, generating a feasible set of solutions where the objective function value decreases and the relaxation boundary deviation converges with iteration.

[0019] Optionally, the substation coordinates and the line corridor in the geospatial data are spatially gridded based on the obtained segment length of the line corridor to generate a distributed storage cable topology partition dataset.

[0020] The partial discharge frequency domain features in the cable vibration spectrum data are mapped to the corresponding cable topology partition dataset according to spatial gridding, and the main mode features of vibration spectrum in each partition are extracted. Dynamic weighting coefficients of spatial topology constraints are defined based on the coordinate distance between the substation and the line corridor and the risk gradient direction.

[0021] The dynamic weighting coefficients are matched with the main modal features of the vibration spectrum across nodes through a parallel computing architecture. The aging risk values ​​in the cable topology partition dataset are dynamically corrected, and the risk gradient direction is iteratively updated according to the boundary of the spatial gridded partition. The cable aging risk distribution map based on spatial topology constraints is then output.

[0022] Optionally, extract directly topologically adjacent node pairs from the cable aging risk distribution map, and calculate the spatial correlation strength based on the cable aging risk difference value and physical connection length between nodes;

[0023] Coupling weights are assigned to each pair of nodes based on the spatial association strength. The coupling weights are negatively correlated with the spatial association strength, and the coupling weights in high-density regions decrease exponentially with the increase in the number of adjacent nodes. Combined with the maximum allowable relaxation deviation dynamically set by the region topology density, asymmetric element values ​​of the coupling parameter matrix are generated.

[0024] The off-diagonal elements of the coupling parameter matrix are mapped to the second-order cone constraint boundary conditions of the relaxation optimization model, and the relaxation deviation vectors of adjacent nodes are constrained to satisfy that the magnitude of their difference vector does not exceed the product of the coupling weight and the maximum allowable relaxation deviation.

[0025] Based on the dynamic setting of regional topology density, the coupling weight is multiplied by the maximum allowable relaxation deviation of its region to generate matrix element values, and the coupling parameter matrix is ​​directly defined based on the matrix element values.

[0026] Optionally, the dynamic weight coefficients are hashed and partitioned according to the spatial gridded partition number, and key-value matching is performed with the vibration spectrum main modal features in the corresponding partition through parallel computing nodes to generate dynamic weight coefficients.

[0027] Based on the dynamic weighting coefficient, the aging risk value in the cable topology partition dataset is weighted and corrected zone by zone. The partition with the higher weighting coefficient and the greater the energy of the main mode characteristic of the vibration spectrum, the greater the correction of its aging risk value.

[0028] Based on the boundary changes of the spatial gridded partitions, the risk gradient direction of the cable aging risk distribution map is recalculated, and the updated risk gradient direction is fed back into the decay function of the dynamic weight coefficient to generate a new round of weight coefficient partitioning.

[0029] Until the rate of change of aging risk values ​​in all spatially gridded partitions is less than the convergence threshold, the cable aging risk distribution map based on spatial topology constraints is output in segments according to the weight coefficients.

[0030] Optionally, candidate solution sets that satisfy thermodynamic stability and cable aging risk tolerance are extracted from the optimal Pareto front of the feasible solution set. Based on the correlation rules between line impedance parameters and node voltage stability margin, a subset of solution sets that makes the voltage deviation rate of key load nodes lower than a preset threshold is selected to generate the line impedance adjustment amount of the high-voltage distribution network.

[0031] For each line in the solution subset, calculate the sensitivity index of its impedance parameter to the gradient direction of the risk value in the cable aging risk distribution map. The sensitivity index is jointly determined by the magnitude of the risk gradient in the area where the line is located and the activity index of the spatial cumulative effect inequality constraint.

[0032] The lines in the solution subset are sorted according to the sensitivity index. Lines whose risk gradient direction is opposite to the sensitivity index are marked as high priority. The cable replacement priority sequence is generated by combining the magnitude and direction of impedance adjustment in the line capacity expansion scheme.

[0033] Optionally, based on the boundary change of the spatial gridded partitions, the difference rate of cable aging risk values ​​between adjacent partitions is calculated to generate a new risk gradient direction vector;

[0034] The risk gradient direction vector is coupled with the decay function of the dynamic weight coefficient, and the ratio parameter of the distance decay factor and the risk correlation factor in the decay function is adjusted.

[0035] Based on the adjusted decay function's proportional parameter, the dynamic weight coefficient partitioning rules for each spatial gridded partition are redefined. The partition with the larger magnitude of the risk gradient direction vector and the higher the decay factor ratio has a smaller coverage area for its weight coefficient partitioning.

[0036] By using parallel computing nodes, the updated weight coefficient slices are matched with the main modal features of the vibration spectrum across nodes to verify the consistency between the slice rules and the risk gradient direction vector, and the final weight coefficient slices are output.

[0037] Optionally, based on the high-priority lines marked in the cable replacement priority sequence, the frequency domain characteristics of partial discharge in the corresponding cable vibration spectrum data are extracted to show the trend.

[0038] Based on the correlation rules between the frequency domain characteristic change trend and the line corridor expansion parameters, the high-priority line corridor expansion parameters within the high-voltage distribution network planning area are iteratively corrected.

[0039] The modified line corridor extension parameters are bound to the cable lifetime prediction model, which calculates the remaining lifetime nodes based on the cumulative effect of the change trend of partial discharge frequency domain characteristics, and associates the topology of the capacity expansion path with the risk suppression effect in the priority sequence.

[0040] By integrating the corrected extended parameters, cable life prediction nodes, and topological constraints of capacity expansion paths, a dynamic programming graph is generated that includes dynamic parameter update rules in the time dimension and risk hot zone markings in the spatial dimension.

[0041] Secondly, this application provides a power distribution network integrated planning system for power engineering design, comprising:

[0042] The acquisition module acquires geospatial data, historical operation data, and cable vibration spectrum data monitored by fiber optic sensor networks within the high-voltage distribution network planning area.

[0043] The analysis module performs distributed storage and correlation analysis on the geospatial data and the cable vibration spectrum data through a parallel computing architecture with a mixed integer programming parallel acceleration algorithm, generating a cable aging risk distribution map based on spatial topology constraints, wherein the spatial topology constraints are dynamically defined by the geospatial relationship between the substation coordinates and the line corridor in the geospatial data.

[0044] The generation module performs multi-objective collaborative optimization on the cable aging risk distribution map, the load peak-valley difference in the historical operation data, and the line corridor expansion parameters in the geospatial data based on the second-order cone programming model, and generates high-voltage distribution network line capacity expansion schemes and cable replacement priority sequences, respectively. The constraints of the multi-objective collaborative optimization include the real-time threshold of the cable joint temperature fluctuation signal and the spatial accumulation effect of the partial discharge frequency domain characteristics.

[0045] The correction module, based on the high-voltage distribution network line capacity expansion scheme and cable replacement priority sequence, and combined with the variation trend of partial discharge frequency domain characteristics in the cable vibration spectrum data, corrects the line corridor expansion parameters within the high-voltage distribution network planning area, and outputs a dynamic planning map including cable life prediction nodes and capacity expansion paths.

[0046] In this embodiment, geospatial data, historical operating data, and cable vibration spectrum data monitored by fiber optic sensor networks within the high-voltage distribution network planning area are acquired. The geospatial data and cable vibration spectrum data are then distributed and analyzed using a parallel computing architecture based on a mixed-integer programming parallel acceleration algorithm to generate a cable aging risk distribution map based on spatial topology constraints. These spatial topology constraints are dynamically defined by the geospatial relationship between substation coordinates and line corridors in the geospatial data. A second-order cone programming model is used to perform multi-objective collaborative optimization on the cable aging risk distribution map, the load peak-valley difference in the historical operating data, and the line corridor expansion parameters in the geospatial data. This generates high-voltage distribution network line capacity expansion schemes and cable replacement priority sequences, respectively. The constraints of this multi-objective collaborative optimization include the real-time threshold of cable joint temperature fluctuation signals and the spatial cumulative effect of partial discharge frequency domain characteristics. Based on the high-voltage distribution network line capacity expansion schemes and cable replacement priority sequences, and combined with the changing trend of partial discharge frequency domain characteristics in the cable vibration spectrum data, the line corridor expansion parameters within the high-voltage distribution network planning area are corrected, outputting a dynamic planning map containing cable life prediction nodes and capacity expansion paths.

[0047] The technical solution of this application has the following beneficial effects:

[0048] This application acquires geospatial data, historical operational data, and cable vibration spectrum data monitored by fiber optic sensor networks. It then utilizes a mixed-integer programming parallel acceleration algorithm for distributed storage and correlation analysis to generate a cable aging risk distribution map based on spatial topology constraints. Using a second-order cone programming model, it performs multi-objective collaborative optimization of cable aging risk, load peak-valley difference, and line corridor expansion parameters to generate line capacity expansion schemes and cable replacement priority sequences. Finally, it corrects the line corridor expansion parameters by incorporating the changing trends of partial discharge frequency domain characteristics, outputting a dynamic programming graph that includes cable life prediction nodes and capacity expansion paths. This achieves scientific, real-time, and intelligent distribution network planning, significantly improving the safety and reliability of the power system.

[0049] Furthermore, a collaborative optimization model is constructed using the risk value gradient in the cable aging risk distribution map, the time series deviation of the load peak-valley difference, and the cost coefficient of the line corridor expansion parameters as optimization variables. By defining thermodynamic constraint equations and spatial cumulative effect inequality constraints, the model is relaxed and iteratively solved using a decomposition and coordination mechanism to generate a feasible solution set. Finally, a high-voltage distribution network line capacity expansion scheme is generated based on the optimal Pareto front. Simultaneously, based on the activity index of the risk value gradient direction and spatial cumulative effect inequality constraints, the line impedance parameters are sensitively ranked to generate a cable replacement priority sequence. Moreover, this method, through multi-objective collaborative optimization and refined constraint processing, achieves the scientific generation of high-voltage distribution network line capacity expansion schemes and cable replacement priority sequences, significantly improving the feasibility and practicality of the planning scheme. At the same time, the iterative solution based on relaxation processing and decomposition and coordination mechanisms ensures the efficiency and stability of the optimization model, providing reliable technical support for the safe operation and long-term planning of the power system.

[0050] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of this application 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 A flowchart of a power distribution network integrated planning method in power engineering design provided in this application is shown;

[0053] Figure 2 This application provides a schematic diagram of the structure of a power distribution network integrated planning system in power engineering design. Detailed Implementation

[0054] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0055] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0056] This project aims to develop a comprehensive planning method for power distribution networks in power engineering design. By integrating geospatial data, historical operational data, and cable vibration spectrum data monitored by fiber optic sensor networks within the high-voltage distribution network planning area, a mixed-integer programming parallel acceleration algorithm is used for distributed storage and correlation analysis to generate a cable aging risk distribution map based on spatial topology constraints. A second-order cone programming model is used to perform multi-objective collaborative optimization of cable aging risk, load peak-valley difference, and line corridor expansion parameters, generating high-voltage distribution network line capacity expansion schemes and cable replacement priority sequences. Furthermore, by combining the partial discharge frequency domain characteristic change trend in the cable vibration spectrum data, the line corridor expansion parameters are dynamically corrected. Finally, a dynamic planning graph containing cable life prediction nodes and capacity expansion paths is output, realizing intelligent, scientific, and dynamic distribution network planning, and providing reliable technical support for the safe operation and long-term development of the power system.

[0057] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0058] Figure 1 This application provides a flowchart of a power engineering design method for integrated distribution network planning, as shown in the embodiments of this application. Figure 1 As shown, the method includes:

[0059] 101. Obtain geospatial data, historical operation data, and cable vibration spectrum data monitored by fiber optic sensor networks within the high-voltage distribution network planning area;

[0060] In this step, the high-voltage distribution network planning area refers to the specific geographical area requiring power grid planning and cable aging risk management. Geospatial data includes the precise coordinates of substations, the geographical location of line corridors, topographic information, and surrounding environment, used to describe the physical layout and spatial relationships of the high-voltage distribution network. Historical operational data includes operational records such as load peak-valley differences, cable joint temperature fluctuation signals, and partial discharge frequency domain characteristics. Cable vibration spectrum data monitored by the fiber optic sensor network is cable vibration signal acquired through fiber optic sensing technology, used to analyze cable aging status and partial discharge characteristics. Cable vibration spectrum data refers to the real-time monitoring of cable vibration frequency characteristics through the fiber optic sensor network, including partial discharge frequency domain characteristics and vibration amplitude, reflecting the cable's health status and aging trend.

[0061] In this embodiment, firstly, the precise coordinates of substations, the geographical location of the transmission line corridor, and topographic information within the planning area are extracted using Geographic Information System (GIS) technology, and the geological conditions of the transmission line corridor are analyzed using a digital elevation model (DEM). Secondly, load peak-valley differences, fault records, and equipment status data from the past 5-10 years are extracted from historical databases, and the data is cleaned and standardized. Simultaneously, vibration spectrum data is collected in real time using a fiber optic sensor network deployed on the cable, and noise interference is removed and effective signals are extracted using a filtering algorithm. Finally, the geospatial data, historical operating data, and vibration spectrum data are integrated into a unified dataset, providing multi-dimensional basic support for subsequent analysis.

[0062] 102. The geospatial data and the cable vibration spectrum data are distributed and analyzed in a parallel computing architecture using a mixed integer programming parallel acceleration algorithm to generate a cable aging risk distribution map based on spatial topology constraints, wherein the spatial topology constraints are dynamically defined by the geospatial relationship between the substation coordinates and the line corridor in the geospatial data.

[0063] In this step, the mixed-integer programming parallel acceleration algorithm refers to an optimization algorithm used to process large-scale data, improving efficiency through parallel computing, and suitable for the analysis and optimization of complex networks. Spatial topology constraints are dynamically defined by the geographic spatial relationship between substation coordinates and line corridors, used to describe the physical connectivity and spatial layout of the cable network. The cable aging risk distribution map is a risk distribution visualization map generated based on cable vibration spectrum data and spatial topology relationships, used to identify high-risk areas.

[0064] In this embodiment, firstly, a distributed storage system is used to partition and store geospatial data and cable vibration spectrum data to improve data access efficiency; then, a mixed integer programming parallel acceleration algorithm is used to perform correlation analysis on the two types of data, and combined with dynamically defined spatial topology constraints, the aging risk index of each cable is calculated; finally, a cable aging risk distribution map is generated through visualization tools, and high-risk areas are marked in the form of a heat map to intuitively display the distribution characteristics of power grid aging risk.

[0065] 103. Based on the second-order cone programming model, multi-objective collaborative optimization is performed on the cable aging risk distribution map, the load peak-valley difference in the historical operation data, and the line corridor expansion parameters in the geospatial data to generate high-voltage distribution network line capacity expansion schemes and cable replacement priority sequences, respectively. The constraints of the multi-objective collaborative optimization include the real-time threshold of the cable joint temperature fluctuation signal and the spatial accumulation effect of the partial discharge frequency domain characteristics.

[0066] In this step, Second-Order Cone Programming (SOCPR) is a mathematical optimization model suitable for handling multi-objective optimization problems, capable of finding optimal solutions under complex constraints. Multi-objective collaborative optimization refers to finding the optimal balance among multiple objectives (such as capacity expansion, risk reduction, and cost control). Cable joint temperature fluctuation signals are used to monitor temperature changes at cable joints in real time, serving as optimization constraints to ensure safe cable operation. The spatial cumulative effect of partial discharge frequency domain characteristics describes the cumulative impact of partial discharge in space, serving as an optimization constraint to assess cable aging risk.

[0067] In this embodiment, firstly, a second-order cone programming model is constructed with the objective functions of maximizing capacity expansion benefits, minimizing aging risks, and minimizing retrofit costs. The cable aging risk distribution map, the load peak-valley difference in the historical operating data, the line corridor expansion parameters, and the cable aging risk index are used as input parameters for multi-objective collaborative optimization. Secondly, the real-time threshold of cable joint temperature fluctuation signals and the spatial accumulation effect of partial discharge frequency domain characteristics are introduced as hard constraints. The model is solved through a multi-objective optimization algorithm to generate line capacity expansion schemes and cable replacement priority sequences. Finally, sensitivity analysis is used to verify the robustness of the scheme and ensure its feasibility in complex scenarios.

[0068] For the aforementioned high-voltage distribution network, a second-order cone programming model was constructed based on the cable aging risk distribution map and load peak-valley difference. Through multi-objective collaborative optimization, a line capacity expansion scheme was generated, suggesting capacity expansion for a high-risk line, such as increasing cable capacity, replacing equipment, and prioritizing the replacement of severely aging cables. The optimization results effectively reduced the grid operation risk while controlling the transformation cost. Verification results show that the scheme achieves optimal balance among multiple objectives while satisfying the spatial accumulation effect constraints of cable joint temperature fluctuation signals and partial discharge frequency domain characteristics.

[0069] 104. Based on the high-voltage distribution network line capacity expansion scheme and cable replacement priority sequence, and combined with the partial discharge frequency domain characteristic change trend in the cable vibration spectrum data, correct the line corridor expansion parameters in the high-voltage distribution network planning area, and output a dynamic planning map including cable life prediction nodes and capacity expansion paths.

[0070] In this step, the frequency domain characteristic change trend of partial discharge refers to the change law of partial discharge characteristics extracted through time series analysis, which is used to predict the aging process of cables. Cable life prediction nodes refer to the key time points of the remaining cable life predicted based on partial discharge characteristics and machine learning models. Capacity expansion path refers to the implementation path of line renovation planned according to the capacity expansion scheme and geospatial parameters, including the priority and time nodes of capacity expansion measures. Dynamic planning map refers to a visual map integrating cable life prediction, capacity expansion path, and correction parameters, supporting interactive adjustment of power grid planning schemes through a GIS platform.

[0071] In this embodiment, firstly, based on the capacity expansion scheme and cable replacement priority sequence, and combined with the time series change trend analysis of partial discharge frequency domain characteristics, the line corridor expansion parameters within the high-voltage distribution network planning area are corrected; secondly, a machine learning model is used to predict the remaining cable life and generate life prediction nodes; subsequently, combined with geospatial data and power grid topology, a capacity expansion path is planned, and the reliability of the path is evaluated through Monte Carlo simulation; finally, the corrected parameters, life prediction nodes, and capacity expansion paths are integrated into a dynamic planning map, which is then visualized interactively through a GIS platform, supporting power grid managers to adjust planning and resource allocation in real time.

[0072] In the aforementioned high-voltage distribution network, the line corridor expansion parameters were adjusted based on the capacity expansion plan, cable replacement priority sequence, and the changing trend of partial discharge frequency domain characteristics. Based on the changing trend of partial discharge frequency domain characteristics, a machine learning model predicted that the remaining lifespan of a certain cable was less than 5 years, and recommended prioritizing its replacement in the capacity expansion path. Finally, a dynamic programming graph was generated using visualization tools, marking the cable lifespan prediction nodes and capacity expansion paths, providing a scientific basis for power grid planning.

[0073] In summary, steps 101 to 104 achieve accurate visualization and early warning of cable aging risks; based on the parallel acceleration algorithm of mixed integer programming and the second-order cone programming model, the capacity expansion benefits, risk control and transformation costs are synergistically optimized to generate scientific high-voltage distribution network line capacity expansion schemes and cable replacement priority sequences; combined with machine learning to predict the changing trend of partial discharge frequency domain characteristics and dynamically correct the line corridor expansion parameters, a dynamic planning map integrating cable life prediction nodes and capacity expansion paths is output, ultimately realizing accurate management of the entire life cycle of the high-voltage distribution network, significantly improving the reliability of power grid operation, optimizing resource allocation efficiency, and providing data-driven decision support for power grid planning.

[0074] In some embodiments, step 103, which involves multi-objective collaborative optimization based on a second-order cone programming model of the cable aging risk distribution map, the load peak-valley difference in the historical operating data, and the line corridor expansion parameters in the geospatial data, generates high-voltage distribution network line capacity expansion schemes and cable replacement priority sequences, including:

[0075] 201. Construct a collaborative optimization model with the risk value gradient in the cable aging risk distribution map, the time series deviation of the load peak-valley difference, and the cost coefficient of the line corridor expansion parameter as optimization variables;

[0076] In step 201, the cable aging risk distribution map is a risk assessment result generated based on cable vibration spectrum data and spatial topological constraints, used to identify high-risk areas. The risk value gradient refers to the rate of change of risk values ​​in the cable aging risk distribution map, typically used to assess the spatial distribution characteristics of risk. The time series deviation of the load peak-valley difference refers to the change in the difference between peak and valley loads over time, typically used to assess the fluctuation characteristics of grid load. The cost coefficient of the line corridor extension parameters refers to the economic cost of line corridor extension planning, typically determined based on factors such as line length, terrain complexity, and construction difficulty. The collaborative optimization model is a multi-objective optimization model used to comprehensively consider factors such as cable aging risk, load fluctuation, and economic cost to generate the optimal planning scheme.

[0077] In this embodiment, the system first reads the risk value gradient, the time series deviation of the load peak-valley difference, and the cost coefficient of the line corridor extension parameters from the cable aging risk distribution map, and uses these as optimization variables. Next, the system constructs a collaborative optimization model, comprehensively considering factors such as cable aging risk, load fluctuations, and economic costs, to generate a multi-objective optimization problem. Finally, the system stores the collaborative optimization model as structured data for subsequent optimization solutions.

[0078] 202. Based on the real-time threshold definition of the temperature fluctuation signal of the cable joint, a set of thermodynamic constraint equations is defined, and the energy integral of the partial discharge frequency domain characteristics in the cable vibration spectrum data in the spatial dimension is coupled with the preset safety threshold to generate a spatial accumulation effect inequality constraint.

[0079] In step 202, the thermodynamic constraint equations are constraints defined based on the real-time threshold of the cable joint temperature fluctuation signal, used to ensure that the temperature fluctuation of the cable during operation does not exceed the safe range. Partial discharge frequency domain characteristics refer to the distribution characteristics of partial discharge signals in the frequency domain of the cable vibration spectrum data, typically used to assess the aging state of the cable. Spatial accumulation effect refers to the energy integral of the partial discharge frequency domain characteristics in the spatial dimension, used to assess the spatial impact of cable aging. The preset safety threshold is the allowable upper limit of the energy integral of the partial discharge frequency domain characteristics, typically determined based on the cable's rated operating conditions and aging assessment standards. The spatial accumulation effect inequality constraint is a constraint generated by coupling the energy integral of the partial discharge frequency domain characteristics with the preset safety threshold, used to ensure the controllability of cable aging risk in the spatial dimension.

[0080] In this embodiment, the system first reads the real-time threshold of the cable joint temperature fluctuation signal and defines a set of thermodynamic constraint equations. Next, the system couples the energy integral of the partial discharge frequency domain characteristics in the spatial dimension with a preset safety threshold to generate a spatial accumulation effect inequality constraint. Finally, the system stores the thermodynamic constraint equations and the spatial accumulation effect inequality constraint as structured data for subsequent optimization solutions.

[0081] 203. Relax the thermodynamic constraint equations and spatial cumulative effect inequality constraints of the collaborative optimization model. By introducing auxiliary variables, the spatial topological correlation of the cable aging risk distribution map is mapped to the boundary conditions of the second-order cone programming model. The decomposition coordination mechanism is used to iteratively solve the optimization model containing the relaxation constraints to generate a feasible solution set.

[0082] In step 203, relaxation is the process of transforming nonlinear constraints into linear constraints by introducing auxiliary variables, typically used to simplify the solution of optimization models. Auxiliary variables are intermediate variables used in relaxation, usually defined based on the structure and constraints of the optimization model. The spatial topological correlation of the cable aging risk distribution map refers to the distribution relationship of risk values ​​in the spatial dimension, typically used to define the boundary conditions of the optimization model. The decomposition and coordination mechanism is a solution method for handling large-scale optimization problems. It generates a feasible solution set by decomposing the optimization problem into multiple sub-problems and coordinating their solutions. The feasible solution set is the set of optimal solutions that satisfy all constraints, typically used to generate the final planning scheme.

[0083] In this embodiment, firstly, the system relaxes the thermodynamic constraint equations and the spatial cumulative effect inequality constraints, introducing auxiliary variables to transform nonlinear constraints into linear constraints. Next, the system maps the spatial topological correlation of the cable aging risk distribution map to the boundary conditions of a second-order cone programming model, ensuring that the optimization model accurately reflects the spatial distribution characteristics of the risk values. Then, the system uses a decomposition and coordination mechanism to iteratively solve the optimization model, decomposing the large-scale optimization problem into multiple sub-problems and coordinating their solutions to generate a feasible solution set. Finally, the system stores the feasible solution set as structured data for subsequent planning scheme generation.

[0084] 204. Based on the optimal Pareto front of the feasible solution set, generate a high-voltage distribution network line capacity expansion scheme. At the same time, based on the risk value gradient direction of the cable aging risk distribution map and the activity index of the spatial cumulative effect inequality constraint, sort the line impedance parameters in the feasible solution set by sensitivity and generate a cable replacement priority sequence.

[0085] In step 204, the optimal Pareto front is the set of optimal solutions in the feasible solution set, typically used to generate the final planning scheme. The risk value gradient direction refers to the direction of change of risk values ​​in the cable aging risk distribution map, typically used to assess the spatial distribution characteristics of cable aging risk. The activity index of the spatial cumulative effect inequality constraint refers to the activity level of the constraint during the optimization process, typically used to assess the impact of the constraint on the optimization results. Line impedance parameters refer to the impedance characteristics of high-voltage distribution network lines, typically used to assess the electrical performance of the lines. Sensitivity ranking is the process of ranking line impedance parameters according to the activity index, typically used to generate a cable replacement priority sequence.

[0086] In this embodiment, firstly, the system reads the optimal Pareto front of the feasible solution set and generates a high-voltage distribution network line capacity expansion scheme, ensuring that the planning scheme can take into account both cable aging risk management and grid expansion needs. Next, the system evaluates the spatial distribution characteristics of cable aging risk based on the risk value gradient direction of the cable aging risk distribution map. Then, the system extracts the activity index of the spatial cumulative effect inequality constraint to evaluate the impact of the constraint on the optimization results. Finally, the system performs sensitivity ranking of the line impedance parameters, generates a cable replacement priority sequence, and stores it as structured data.

[0087] Here is a specific example:

[0088] In a high-voltage distribution network renovation project in a certain city, power engineer Wang Gong employed a multi-objective collaborative optimization method based on a second-order cone programming model. He constructed a collaborative optimization model with cable aging risk gradient, load peak-valley difference deviation, and line corridor expansion cost as optimization variables, defining thermodynamic constraints and spatial cumulative effect inequality constraints. He relaxed the constraints by introducing auxiliary variables and used a decomposition and coordination mechanism for iterative solution, generating a feasible solution set. Based on the optimal Pareto front, Wang Gong formulated a line capacity expansion plan, for example, increasing the capacity of a certain line to 150MVA. Simultaneously, based on the risk gradient and activity index, he ranked the line impedance parameters by sensitivity, generating a cable replacement priority sequence, prioritizing the replacement of high-risk lines. After implementation, the regional power supply reliability significantly improved, and the power company highly recognized the solution, demonstrating the scientific validity and practicality of this method in distribution network optimization.

[0089] In summary, steps 201 to 204 achieved synergistic optimization of cable aging risk management and power grid expansion planning, improving the accuracy and reliability of the planning results and ensuring the safe operation and sustainable development of the high-voltage distribution network. This method comprehensively considers multiple factors such as cable aging risk, load fluctuations, and economic costs through the construction of a collaborative optimization model; it ensures the feasibility and reliability of the optimization model through the definition of thermodynamic constraint equations and spatial cumulative effect inequalities; it achieves efficient solution of the optimization model through relaxation processing and decomposition coordination mechanisms; and it generates a high-voltage distribution network line capacity expansion scheme and cable replacement priority sequence through optimal Pareto front and sensitivity ranking.

[0090] In some embodiments, step 203 involves relaxing the thermodynamic constraint equations and spatial cumulative effect inequality constraints of the collaborative optimization model. This is achieved by introducing auxiliary variables to map the spatial topological correlation of the cable aging risk distribution map to the boundary conditions of a second-order cone programming model. A decomposition and coordination mechanism is then used to iteratively solve the optimization model containing the relaxed constraints, generating a feasible solution set, including:

[0091] 301. The inequality constraints related to spatial cumulative effects in the thermodynamic constraint equation set are relaxed by introducing auxiliary variables, and the auxiliary variables are quantified to determine the constraint deviation and a dynamic relaxation boundary is constructed.

[0092] In step 301, the thermodynamic constraint equations are constraints defined based on the real-time threshold of the cable joint temperature fluctuation signal, used to ensure that the temperature fluctuation of the cable during operation does not exceed the safe range. The spatial accumulation effect inequality constraint is a constraint generated by coupling the energy integral of the partial discharge frequency domain characteristics in the spatial dimension with a preset safety threshold, used to ensure the controllability of cable aging risk in the spatial dimension. Auxiliary variables are intermediate variables used for relaxation processing, usually defined based on the structure and constraints of the optimization model. The dynamic relaxation boundary is a boundary condition constructed by introducing auxiliary variables, used to quantify the constraint deviation and dynamically adjust the effect of relaxation processing.

[0093] In this embodiment, the system first reads the inequality constraints related to spatial cumulative effects in the thermodynamic constraint equations and analyzes their nonlinear characteristics. Next, the system introduces auxiliary variables to transform the nonlinear constraints into linear constraints and quantifies the constraint deviation. Then, the system constructs dynamic relaxation boundaries to ensure that the relaxation effect dynamically adapts to the needs of the optimization model. Finally, the system stores the dynamic relaxation boundaries as structured data for subsequent iterative optimization.

[0094] 302. Map the spatial topological correlation of the cable aging risk distribution map to the boundary conditions of the second-order cone programming model, and define the coupling parameter matrix through topological adjacency relationships;

[0095] In step 302, the spatial topological correlation of the cable aging risk distribution map refers to the distribution relationship of risk values ​​in the spatial dimension, and is usually used to define the boundary conditions of the optimization model. The second-order cone programming model is an optimization model generated through relaxation processing, and is usually used to simplify the solution process. Topological adjacency refers to the connection relationship between nodes in the cable aging risk distribution map, and is usually used to define the coupling parameter matrix. The coupling parameter matrix is ​​a matrix generated based on the topological adjacency and is used to describe the coupling strength between nodes.

[0096] In this embodiment, the system first reads the spatial topological correlation of the cable aging risk distribution map and analyzes the spatial distribution of risk values. Next, the system maps the spatial topological correlation to the boundary conditions of a second-order cone programming model, ensuring that the optimization model accurately reflects the spatial distribution characteristics of the risk values. Then, the system defines a coupling parameter matrix based on topological adjacency relationships to describe the coupling strength between nodes. Finally, the system stores the coupling parameter matrix as structured data for subsequent iterative optimization.

[0097] 303. Based on the constraint relationship between the coupling parameter matrix and the dynamic relaxation boundary, an alternating iterative update mechanism of main variables and auxiliary variables is adopted. In the main variable update stage, the thermodynamic state parameters are solved under the fixed relaxation boundary. In the auxiliary variable update stage, the deviation and penalty coefficient of the relaxation boundary are dynamically adjusted according to the current main variable residual and the topological connection strength until the residual converges to the preset threshold.

[0098] In step 303, the main variables are the primary variables in the optimization model, typically used to describe the core parameters of the optimization problem. Auxiliary variables are intermediate variables used for relaxation processing, usually defined based on the structure and constraints of the optimization model. The alternating iterative update mechanism is an iterative method for solving optimization problems, gradually approaching the optimal solution by alternately updating the main and auxiliary variables. The main variable update stage is the process of solving the thermodynamic state parameters under fixed relaxation boundaries, typically used to optimize core parameters. The auxiliary variable update stage is the process of dynamically adjusting the deviation and penalty coefficient of the relaxation boundaries based on the current main variable residuals and topological connectivity strength, typically used to optimize the effect of relaxation processing. The residual is the difference between the objective function value of the optimization model and the actual value, typically used to evaluate the convergence of the optimization results. The preset threshold is the upper limit of the allowable residual, typically determined based on the accuracy requirements of the optimization model.

[0099] In this embodiment, firstly, the system reads the coupling parameter matrix and the constraint relationship of the dynamic relaxation boundary, and initializes the main variables and auxiliary variables. Next, during the main variable update phase, the system fixes the relaxation boundary and solves for the thermodynamic state parameters. Then, during the auxiliary variable update phase, the system dynamically adjusts the deviation and penalty coefficient of the relaxation boundary based on the current main variable residuals and topological connectivity strength. Finally, the system repeats the above process until the residuals converge to a preset threshold.

[0100] 304. During the iteration process, intermediate solutions that satisfy the constraints of the dynamic relaxation boundary and the coupling parameter matrix are simultaneously screened. Solutions that conflict with spatial topology are eliminated by non-dominated sorting, and a feasible solution set is generated where the objective function value decreases and the relaxation boundary deviation converges with iteration.

[0101] In step 304, intermediate solutions are optimal solutions generated during the iteration process that satisfy the constraints of dynamic relaxation boundaries and coupling parameter matrices. These are typically used to generate the final feasible solution set. Non-dominated sorting is a method for filtering optimal solutions. By eliminating solutions that conflict with spatial topological correlations, it generates a feasible solution set where the objective function value decreases and the relaxation boundary deviation converges with iteration. Spatial topological correlations refer to the distribution relationship of risk values ​​in the spatial dimension and are typically used to define the boundary conditions of the optimization model. The feasible solution set is the set of optimal solutions that satisfy all constraints and is typically used to generate the final planning scheme.

[0102] In this embodiment, firstly, the system synchronously filters intermediate solutions that satisfy the constraints of dynamic relaxation boundaries and coupling parameter matrices during the iteration process, ensuring that the optimized solution satisfies all constraints. Next, the system eliminates solutions that conflict with spatial topological relationships through non-dominated sorting, generating a feasible solution set where the objective function value decreases and the relaxation boundary deviation converges with iteration. Finally, the system stores the feasible solution set as structured data for subsequent planning scheme generation.

[0103] Here is a specific example:

[0104] In a high-voltage power distribution network optimization project in a certain city, power engineer Liu was responsible for handling the thermodynamic constraints and spatial cumulative effect constraints of the collaborative optimization model. He first introduced auxiliary variables to relax the inequality constraints related to spatial cumulative effects in the thermodynamic constraint equations. For example, the deviation of the thermodynamic constraints of a certain line was quantified as an auxiliary variable, and a dynamic relaxation boundary was constructed. Next, Liu mapped the spatial topological correlation of the cable aging risk distribution map to the boundary conditions of a second-order cone programming model, defining a coupling parameter matrix through topological adjacency relationships. For example, the coupling parameter between substation A and distribution cabinet B was set to 0.8, reflecting their close connection. Based on the constraints of the coupling parameter matrix and the dynamic relaxation boundary, Liu adopted an alternating iterative update mechanism of main variables and auxiliary variables: in the main variable update stage, he solved for thermodynamic state parameters under a fixed relaxation boundary, such as the temperature distribution of a certain line; in the auxiliary variable update stage, he dynamically adjusted the deviation and penalty coefficient of the relaxation boundary according to the current main variable residuals and topological connection strength, for example, adjusting the relaxation deviation of a certain area from 0.1 to 0.05. Through multiple iterations, the residuals gradually converged to a preset threshold. During the iteration process, Liu simultaneously screened intermediate solutions that satisfied the constraints of the dynamic relaxation boundary and the coupling parameter matrix, and eliminated solutions that conflicted with spatial topology through non-dominated sorting. For example, a solution for a certain route was eliminated due to a conflict with topology. Finally, he generated a feasible set of solutions where the objective function value decreased and the relaxation boundary deviation converged with iteration.

[0105] In summary, steps 301 to 304 achieve efficient solution of the optimization model, improve the accuracy and reliability of the planning results, and ensure the safe operation and sustainable development of the high-voltage distribution network. Specifically, this method simplifies the constraints of the optimization model through relaxation processing, gradually approximates the optimal solution through an alternating iterative update mechanism, and generates a feasible solution set that satisfies all constraints through non-dominated sorting.

[0106] In some embodiments, step 102, which involves distributing and performing correlation analysis on the geospatial data and the cable vibration spectrum data using a parallel computing architecture based on a mixed-integer programming parallel acceleration algorithm to generate a cable aging risk distribution map based on spatial topology constraints, includes:

[0107] 401. Based on the obtained segment length of the line corridor, the substation coordinates and line corridor in the geospatial data are spatially gridded and partitioned to generate a distributed storage cable topology partition dataset.

[0108] In step 401, the segment length of the line corridor refers to the division length of cable lines in the high-voltage distribution network, which is usually determined based on the physical characteristics of the lines and management requirements. Geospatial data includes spatial information such as substation coordinates, line corridor locations, and terrain features. Spatial gridding is the process of dividing geospatial data into multiple grid regions according to the segment length of the line corridor, typically used to simplify data processing and optimize calculations. Distributed storage of the cable topology partitioning dataset disperses the spatially gridded data across multiple computing nodes to improve the efficiency and reliability of data processing.

[0109] In this embodiment, the system first reads the segment lengths of the transmission line corridor to determine the size and range of the grid division. Next, the system divides the substation coordinates and transmission line corridor locations into multiple grid regions according to the segment lengths, generating spatial gridded partitions. Then, the system distributes the partitioned data across multiple computing nodes, generating a distributed cable topology partition dataset. Finally, the system stores the cable topology partition dataset as structured data for subsequent analysis.

[0110] 402. Map the partial discharge frequency domain features in the cable vibration spectrum data to the corresponding cable topology partition dataset according to spatial gridding, and extract the main mode features of the vibration spectrum in each partition. Define the dynamic weight coefficients of the spatial topology constraints based on the coordinate distance between the substation and the line corridor and the risk gradient direction;

[0111] In step 402, the partial discharge frequency domain characteristics in the cable vibration spectrum data refer to the distribution characteristics of the cable vibration signal in the frequency domain, which are typically used to assess the aging state of the cable. The dominant mode characteristics of the vibration spectrum refer to the main frequency components in the cable vibration spectrum data, which are typically used to analyze the source and mode of vibration. The coordinate distance between the substation and the transmission line corridor refers to the spatial distance between the substation and the transmission line corridor, which is typically used to assess the spatial impact of cable aging. The risk gradient direction refers to the direction of change of risk values ​​in the cable aging risk distribution map, which is typically used to assess the spatial distribution characteristics of risk. The dynamic weighting coefficient of the spatial topology constraint is a weighting parameter generated based on the coordinate distance between the substation and the transmission line corridor and the risk gradient direction, used to dynamically adjust the results of the cable aging risk assessment.

[0112] In this embodiment, firstly, the system reads the partial discharge frequency domain characteristics from the cable vibration spectrum data and maps them to the corresponding cable topology partition dataset according to spatial gridding. Next, the system extracts the main modal characteristics of the vibration spectrum within each partition and analyzes the main frequency components of the vibration. Then, based on the coordinate distance between the substation and the line corridor and the risk gradient direction, the system defines dynamic weighting coefficients for the spatial topology constraints. Finally, the system stores the dynamic weighting coefficients as structured data for subsequent analysis.

[0113] 403. By using a parallel computing architecture, the dynamic weight coefficients are matched with the main modal features of the vibration spectrum across nodes, the aging risk values ​​in the cable topology partition dataset are dynamically corrected, and the risk gradient direction is iteratively updated according to the boundary of the spatial gridded partition, outputting a cable aging risk distribution map based on spatial topology constraints.

[0114] In step 403, the parallel computing architecture is a computing architecture used for efficiently processing large-scale data, typically implemented using distributed computing technology. Cross-node matching is the process of matching dynamic weighting coefficients with the main modal characteristics of the vibration spectrum across different computing nodes, usually used to optimize computational efficiency. The aging risk value refers to the result of cable aging risk assessment, typically used to identify high-risk areas. The boundary of the spatial gridded partition refers to the boundary of the grid region, typically used to define the calculation range of the risk gradient direction. The risk gradient direction refers to the direction of change of risk values ​​in the cable aging risk distribution map, typically used to assess the spatial distribution characteristics of risk. The cable aging risk distribution map based on spatial topology constraints is a risk assessment result generated based on dynamic weighting coefficients and the main modal characteristics of the vibration spectrum, used to identify high-risk areas.

[0115] In this embodiment, firstly, the system optimizes computational efficiency by performing cross-node matching of dynamic weighting coefficients with the main modal features of the vibration spectrum using a parallel computing architecture. Next, the system dynamically corrects the aging risk values ​​in the cable topology partitioning dataset to identify high-risk areas. Then, the system iteratively updates the risk gradient direction based on the boundaries of the spatially gridded partitions to assess the spatial distribution characteristics of the risk. Finally, the system generates a cable aging risk distribution map based on spatial topology constraints and stores it as visualization data.

[0116] Here is a specific example:

[0117] In a high-voltage power distribution network optimization project in a large industrial park, power engineer Chen was responsible for generating a cable aging risk distribution map. He first spatially gridded the substation coordinates and power corridors in the geospatial data based on the obtained line corridor segment lengths. For example, he divided the park into 20 partitions, generating a distributed cable topology partition dataset. Next, he mapped the partial discharge frequency domain features from the cable vibration spectrum data to the corresponding cable topology partition datasets, extracting the dominant mode features of the vibration spectrum within each partition. For example, the dominant mode feature frequency of partition A is 50Hz, indicating a high degree of cable aging in this area. Based on the coordinate distance between the substation and the line corridor and the risk gradient direction, Chen defined dynamic weight coefficients for the spatial topology constraints. For example, partition B has a higher weight coefficient due to its proximity to the substation and significant risk gradient direction. Subsequently, he used a parallel computing architecture to perform cross-node matching between the dynamic weight coefficients and the dominant mode features of the vibration spectrum, dynamically correcting the aging risk values ​​in the cable topology partition datasets. For example, the aging risk value of partition C was corrected from 70% to 85%. Based on the boundary changes of the spatial gridded partitions, Engineer Chen iteratively updated the risk gradient direction, ultimately outputting a cable aging risk distribution map based on spatial topology constraints. This distribution map provided a scientific basis for the transformation of the park's power distribution network, significantly improved power supply reliability, and received high recognition from the park management.

[0118] In summary, steps 401 to 403 achieved spatial optimization of cable aging risk assessment, improving the accuracy and reliability of risk assessment and ensuring the safe operation and sustainable development of the high-voltage distribution network. Specifically, this method simplifies data processing complexity through spatial grid partitioning, dynamically adjusts risk assessment through dynamic weighting coefficients, optimizes computational efficiency through parallel computing architecture, and provides accurate risk distribution information through iterative updates of the risk gradient direction.

[0119] In some embodiments, step 302, which maps the spatial topological correlation of the cable aging risk distribution map to the boundary conditions of a second-order cone programming model and defines a coupling parameter matrix through topological adjacency relationships, includes:

[0120] 501. Extract directly topologically adjacent node pairs from the cable aging risk distribution map, and calculate the spatial correlation strength based on the cable aging risk difference value and physical connection length between nodes.

[0121] In step 501, the cable aging risk distribution map is a risk assessment result generated based on cable vibration spectrum data and spatial topological constraints, used to identify high-risk areas. Directly topologically adjacent node pairs refer to directly connected node pairs in the cable aging risk distribution map, typically used to assess the spatial correlation between nodes. The cable aging risk difference value refers to the difference in risk values ​​between node pairs, typically used to assess the risk changes between nodes. The physical connection length refers to the physical distance between node pairs, typically used to assess the connection strength between nodes. The spatial correlation strength is the correlation strength between node pairs calculated based on the cable aging risk difference value and the physical connection length, typically used to assess the spatial correlation between nodes.

[0122] In this embodiment, the system first reads the cable aging risk distribution map and extracts directly topologically adjacent node pairs. Next, the system calculates the cable aging risk difference value between node pairs to assess the risk changes between nodes. Then, the system reads the physical connection length between node pairs to assess the connection strength between nodes. Finally, the system calculates the spatial association strength based on the cable aging risk difference value and the physical connection length, and stores it as structured data.

[0123] 502. Assign coupling weights to each pair of nodes according to the spatial association strength. The coupling weights are negatively correlated with the spatial association strength, and the coupling weights in high-density regions decrease exponentially with the increase in the number of adjacent nodes. Combined with the maximum allowable relaxation deviation dynamically set by the region topology density, generate asymmetric element values ​​of the coupling parameter matrix.

[0124] In step 502, the coupling weight is a weight parameter assigned to each pair of nodes based on the spatial association strength, typically used to evaluate the coupling strength between nodes. Spatial association strength is negatively correlated with coupling weight; that is, the higher the spatial association strength, the lower the coupling weight. High-density areas refer to regions with high node density in the cable aging risk distribution map, typically used to evaluate the coupling strength between nodes. The number of adjacent nodes refers to the number of directly connected nodes in a high-density area, typically used to evaluate the coupling strength between nodes. Exponential decay refers to the exponential decrease in coupling weight as the number of adjacent nodes increases, typically used to evaluate the coupling strength between nodes. The maximum allowable relaxation deviation refers to the maximum value that a node's state is allowed to deviate from its ideal state during the optimization process. The coupling parameter matrix is ​​a matrix describing the coupling relationship between nodes. Its asymmetric element values ​​are determined by the coupling weight and the maximum allowable relaxation deviation.

[0125] In this embodiment, firstly, the system reads the spatial association strength and assigns coupling weights to each pair of nodes, ensuring that the coupling weights are negatively correlated with the spatial association strength. Next, the system identifies high-density regions and evaluates regions with high node density. Then, the system calculates the number of adjacent nodes in the high-density regions and evaluates the coupling strength between nodes. The system adjusts the coupling weights based on the number of adjacent nodes, ensuring that the coupling weights decrease exponentially with the increase in the number of adjacent nodes. Then, the maximum allowable relaxation deviation is dynamically set based on the region's topology density; for example, a smaller relaxation deviation is set in high-density regions to ensure precise control of node states, while a larger relaxation deviation is set in low-density regions to reduce optimization complexity. Finally, asymmetric element values ​​of the coupling parameter matrix are generated based on the coupling weights and the maximum allowable relaxation deviation, forming a matrix describing the coupling relationships between nodes. This matrix provides important input for the subsequent relaxation optimization model.

[0126] 503. Map the off-diagonal elements of the coupling parameter matrix to the second-order cone constraint boundary conditions of the relaxation optimization model, and constrain the relaxation deviation vectors of adjacent nodes to satisfy that the magnitude of their difference vectors does not exceed the product of the coupling weight and the maximum allowable relaxation deviation.

[0127] In step 503, the relaxation optimization model refers to a mathematical model used to optimize node states. It allows node states to deviate from their ideal state within a certain range, improving the flexibility and practicality of the optimization. A second-order cone constraint is a mathematical constraint used to limit the magnitude of a vector. It is commonly used in optimization models to describe the range of differences between node states. The relaxation deviation vector describes the deviation of a node state from its ideal state. The difference vector is the difference between the relaxation deviation vectors of adjacent nodes. It describes the degree of difference between adjacent node states.

[0128] In this embodiment, firstly, the off-diagonal elements of the coupling parameter matrix are mapped to second-order cone constraint boundary conditions of the relaxation optimization model. Specifically, the relaxation deviation vectors of adjacent nodes are constrained to ensure that the magnitude of their difference vector does not exceed the product of the coupling weight and the maximum allowable relaxation deviation. For example, in a power network, the voltage state difference between adjacent substations is limited to a certain range to ensure the stability of the power system; in a traffic network, the signal timing difference between adjacent intersections is limited to a certain range to ensure the smoothness of traffic flow. Through this constraint condition, the relaxation deviation of adjacent nodes is ensured to be within the allowable range, and the constraint strength is dynamically adjusted in conjunction with the coupling weight, thereby achieving global optimization of the node state. This process provides precise constraints for the optimization model, ensuring the rationality and practicality of the optimization results.

[0129] 504. Based on the dynamic setting of regional topology density, the coupling weight is multiplied by the maximum allowable relaxation deviation of its region to generate matrix element values, and the coupling parameter matrix is ​​directly defined according to the matrix element values.

[0130] In step 504, the region topology density refers to the distribution of node density in the cable aging risk distribution map, typically used to assess the coupling strength between nodes. The maximum allowable relaxation deviation refers to the maximum allowable deviation in the optimization model, usually determined based on the accuracy requirements of the optimization model. Matrix element values ​​are matrix elements generated based on the coupling weights and the maximum allowable relaxation deviation, typically used to define the coupling parameter matrix. The coupling parameter matrix is ​​a matrix generated based on the matrix element values ​​and is used to describe the coupling strength between nodes.

[0131] In this embodiment, the system first reads the region's topology density and dynamically sets the maximum allowable relaxation bias. Next, the system multiplies the coupling weights by the maximum allowable relaxation bias of their respective regions to generate matrix element values. Then, the system directly defines a coupling parameter matrix based on these matrix element values ​​to describe the coupling strength between nodes. Finally, the system stores the coupling parameter matrix as structured data for subsequent optimization.

[0132] Here is a specific example:

[0133] In a coastal industrial zone, the high-voltage power distribution network suffered from severe cable aging due to long-term salt spray corrosion and load fluctuations, leading to multiple localized power outages. Upon receiving the task, power engineer Zhao first conducted a comprehensive investigation of the industrial zone's power distribution network, extracting directly topologically adjacent node pairs from the cable aging risk distribution map, such as the connection line between substation A and distribution cabinet B. Based on the aging risk difference values ​​and physical connection lengths between nodes, Zhao calculated the spatial correlation strength, finding that lines near the coast had significantly higher risk difference values ​​than inland areas due to severe corrosion. Next, he assigned coupling weights to each node pair; for example, the weight between substation A and distribution cabinet B was lower due to their higher spatial correlation strength, while the weight between distribution cabinet B and distribution cabinet C was higher due to their shorter connection line and smaller aging risk difference. In high-density areas, such as the core power supply hub of the industrial zone, the coupling weight decreased exponentially with the number of adjacent nodes to avoid excessive relaxation affecting optimization accuracy. Finally, Zhao dynamically set the maximum allowable relaxation deviation based on the regional topology density, multiplied the coupling weights by the relaxation deviation to generate matrix element values, defined the coupling parameter matrix, and applied it to a second-order cone programming model.

[0134] In summary, steps 501 to 503 effectively calculated the spatial correlation strength and generated the coupling parameter matrix, improving the accuracy and reliability of the optimization model and ensuring the safe operation and sustainable development of the high-voltage distribution network. Specifically, this method provides an accurate assessment of the correlation strength between nodes through the calculation of spatial correlation strength; ensures the dynamic adjustment of the coupling strength between nodes through the allocation of coupling weights; and provides reliable matrix support for optimization solutions through the generation of the coupling parameter matrix.

[0135] In some embodiments, step 403 involves using a parallel computing architecture to perform cross-node matching between the dynamic weighting coefficients and the main modal features of the vibration spectrum, dynamically correcting the aging risk values ​​in the cable topology partitioning dataset, and iteratively updating the risk gradient direction according to the boundaries of the spatially gridded partitions, outputting a cable aging risk distribution map based on spatial topology constraints, including:

[0136] 601. The dynamic weight coefficients are hashed and partitioned according to the spatial grid partition number, and key-value matching is performed with the main modal features of the vibration spectrum in the corresponding partition through parallel computing nodes to generate dynamic weight coefficients.

[0137] In step 601, the dynamic weighting coefficients are weighting parameters generated based on spatial topology constraints, typically used to dynamically adjust the results of cable aging risk assessment. Spatial gridding partitioning is the process of dividing geospatial data into multiple grid regions according to the segment length of the cable corridor, typically used to simplify data processing and optimize calculations. Hash sharding is the process of storing the dynamic weighting coefficients in shards according to the spatial gridding partition numbers, typically used to optimize data storage and retrieval efficiency. Vibration spectrum master modal characteristics refer to the main frequency components in cable vibration spectrum data, typically used to analyze the source and mode of vibration. Parallel computing nodes are computing nodes used to efficiently process large-scale data, typically implemented using distributed computing technology. Key-value matching is the process of matching dynamic weighting coefficients with vibration spectrum master modal characteristics on different computing nodes, typically used to optimize computational efficiency.

[0138] In this embodiment, the system first reads the dynamic weighting coefficients and performs hash partitioning according to the spatial gridded partition numbers to optimize data storage and retrieval efficiency. Next, the system performs key-value matching between the partitioned dynamic weighting coefficients and the main modal characteristics of the vibration spectrum within the corresponding partition using parallel computing nodes to optimize computational efficiency. Then, the system generates dynamic weighting coefficients based on the matching results, ensuring that the weighting coefficients can dynamically adjust the results of the cable aging risk assessment. Finally, the system stores the dynamic weighting coefficients as structured data for subsequent weighted correction.

[0139] 602. Based on the dynamic weighting coefficient, the aging risk value in the cable topology partition dataset is corrected by weight for each partition. The partition with the higher the weighting coefficient and the greater the energy of the main mode characteristic of the vibration spectrum, the greater the correction magnitude of its aging risk value.

[0140] In step 602, the cable topology partition dataset is a dataset where the spatially gridded partitioned data is distributed and stored across multiple computing nodes, typically used to simplify data processing and optimize computation. The aging risk value refers to the result of cable aging risk assessment, typically used to identify high-risk areas. Weighted correction is a process of correcting the aging risk value in the cable topology partition dataset zone by zone based on dynamic weighting coefficients, typically used to dynamically adjust the risk assessment results. The higher the weighting coefficient and the greater the energy of the dominant mode characteristic of the vibration spectrum in a partition, the greater the correction magnitude for its aging risk value.

[0141] In this embodiment, the system first reads the dynamic weighting coefficients and aging risk values ​​from the cable topology partition dataset to determine the weighting coefficients and vibration spectrum principal modal characteristic energy for each partition. Next, the system performs a weighted correction on a partition-by-partition basis based on the weighting coefficients and vibration spectrum principal modal characteristic energy, ensuring that partitions with higher weighting coefficients and greater vibration spectrum principal modal characteristic energy receive a larger correction in their aging risk values. Then, the system stores the corrected aging risk values ​​as structured data for subsequent calculation of the risk gradient direction.

[0142] 603. Based on the boundary changes of the spatial gridded partition, recalculate the risk gradient direction of the cable aging risk distribution map, and feed the updated risk gradient direction back into the decay function of the dynamic weight coefficient to generate a new round of weight coefficient partitioning.

[0143] In step 603, the boundary of the spatial gridding partition refers to the boundary of the grid region, which is typically used to define the calculation range of the risk gradient direction. The risk gradient direction refers to the direction of change of risk values ​​in the cable aging risk distribution map, which is typically used to assess the spatial distribution characteristics of risk. The attenuation function is a function generated based on the risk gradient direction, which is typically used to dynamically adjust the attenuation trend of the weight coefficients. The weight coefficient partitioning is a weight parameter partitioning generated based on the attenuation function, which is typically used to optimize data storage and retrieval efficiency.

[0144] In this embodiment, the system first reads the boundary changes of the spatial gridded partitions, recalculates the risk gradient direction of the cable aging risk distribution map, and assesses the changing trend of the risk values. Next, the system feeds back the updated risk gradient direction to the decay function of the dynamic weighting coefficients, generating a new round of weighting coefficient slices to ensure that the weighting coefficients can dynamically adjust the risk assessment results. Finally, the system stores the weighting coefficient slices as structured data for subsequent weighted correction. This step ensures that the calculation of the weighting coefficients can be dynamically adjusted through the feedback of the risk gradient direction, providing reliable weighting parameters for subsequent weighted correction.

[0145] 604. Until the rate of change of aging risk values ​​in all spatial gridded partitions is less than the convergence threshold, output a cable aging risk distribution map based on spatial topology constraints according to the weight coefficients.

[0146] In step 604, the convergence threshold is the upper limit of the allowable rate of change of aging risk values, which is usually determined based on the accuracy requirements of the optimization model. The boundary of the spatial gridded partition refers to the boundary of the grid region, which is usually used to define the calculation range of the risk gradient direction. The cable aging risk distribution map based on spatial topology constraints is a risk assessment result generated based on dynamic weight coefficients and weighted corrections, and is used to identify high-risk areas.

[0147] In this embodiment, firstly, the system reads the rate of change of aging risk values ​​for all spatially gridded partitions and compares it with a convergence threshold to determine whether the convergence condition is met. Next, if the convergence condition is met, the system outputs a cable aging risk distribution map based on spatial topology constraints, segmented according to weighted coefficients, to identify high-risk areas. Finally, the system stores the cable aging risk distribution map as visualization data for subsequent planning scheme generation.

[0148] Here is a specific example:

[0149] In a large city, power engineer Liu was responsible for optimizing the aging risk assessment of the distribution network. He first hashed the dynamic weighting coefficients according to spatial grid partition numbers and matched them with the main modal characteristics of the vibration spectrum through parallel computing nodes to generate dynamic weighting coefficients. For example, partition A, due to its high vibration spectrum energy and large weighting coefficient, had a significantly higher aging risk value correction margin than other partitions. Based on the weighting coefficients, Liu corrected the aging risk values ​​zone by zone and recalculated the risk gradient direction according to changes in partition boundaries, updating the weighting coefficient partitions. This process was repeated until the rate of change of aging risk values ​​in all partitions was less than the convergence threshold, ultimately outputting a cable aging risk distribution map based on spatial topology constraints. In practical applications, a commercial area operating under high loads for a long time had its cable aging risks accurately identified and prioritized for upgrades, preventing potential power outages. The power company highly praised Liu's efficient solution, believing it provided crucial support for intelligent management of the distribution network.

[0150] In summary, steps 601 to 604 were used to calculate the dynamic weighting coefficients and optimize the risk distribution map, thereby improving the accuracy and reliability of risk assessment and ensuring the safe operation and sustainable development of the high-voltage distribution network.

[0151] In some embodiments, step 204 involves generating a high-voltage distribution network line capacity expansion scheme based on the optimal Pareto front of the feasible solution set. Simultaneously, based on the risk value gradient direction of the cable aging risk distribution map and the activity index of the spatial cumulative effect inequality constraint, the line impedance parameters in the feasible solution set are ranked by sensitivity to generate a cable replacement priority sequence, including:

[0152] 701. Extract candidate solution sets that satisfy thermodynamic stability and cable aging risk tolerance from the optimal Pareto front of the feasible solution set. Based on the correlation rules between line impedance parameters and node voltage stability margin, screen the solution set subset that makes the voltage deviation rate of key load nodes lower than the preset threshold, and generate the line impedance adjustment amount of high voltage distribution network.

[0153] In step 701, the optimal Pareto front is the set of optimal solutions in the feasible solution set, typically used to generate the final planning scheme. Thermodynamic stability refers to the state where the temperature fluctuation of a cable during operation does not exceed a safe range, typically used to assess the operational safety of the cable. Cable aging risk tolerance refers to the allowable upper limit of cable aging risk, typically determined based on the cable's rated operating conditions and aging assessment standards. Line impedance parameters refer to the impedance characteristics of high-voltage distribution network lines, typically used to assess the electrical performance of the lines. Node voltage stability margin refers to the stability margin of node voltage, typically used to assess the voltage stability of the power grid. Critical load nodes refer to load nodes that have a significant impact on power grid operation, typically used to assess the load characteristics of the power grid. Voltage deviation rate refers to the proportion of difference between the node voltage and the rated voltage, typically used to assess the voltage stability of the power grid. The preset threshold is the allowable upper limit of the voltage deviation rate, typically determined based on the stability requirements of the power grid. The high-voltage distribution network line impedance adjustment amount is a line impedance adjustment scheme generated based on the candidate solution set, typically used to optimize the electrical performance of the power grid.

[0154] In this embodiment, firstly, the system reads the optimal Pareto front of the feasible solution set and extracts candidate solution sets that satisfy thermodynamic stability and cable aging risk tolerance. Next, based on the correlation rules between line impedance parameters and node voltage stability margins, the system filters a subset of solution sets that ensures the voltage deviation rate of critical load nodes is below a preset threshold. Then, the system generates high-voltage distribution network line impedance adjustment amounts based on these solution subsets to optimize the electrical performance of the power grid. Finally, the system stores the high-voltage distribution network line impedance adjustment amounts as structured data for use in generating subsequent cable replacement priority sequences.

[0155] 702. For each line in the solution subset, calculate the sensitivity index of its impedance parameter to the gradient direction of the risk value of the cable aging risk distribution map. The sensitivity index is jointly determined by the risk gradient magnitude of the area where the line is located and the activity index of the spatial cumulative effect inequality constraint.

[0156] In step 702, the sensitivity index is the degree of influence of the line impedance parameter on the gradient direction of the risk value in the cable aging risk distribution map, and is typically used to evaluate the effectiveness of line impedance adjustment. The risk gradient magnitude refers to the magnitude of change in the risk value in the cable aging risk distribution map, and is typically used to evaluate the spatial distribution characteristics of the risk. The activity index of the spatial cumulative effect inequality constraint refers to the activity level of the constraint during the optimization process, and is typically used to evaluate the impact of the constraint on the optimization results.

[0157] In this embodiment, firstly, the system reads each line in the solution subset and determines its impedance parameters. Next, the system calculates the risk gradient magnitude of the region where the line is located and assesses the magnitude of the risk value change. Then, the system reads the activity index of the spatial cumulative effect inequality constraint and assesses the impact of the constraint on the optimization results. Finally, the system jointly determines a sensitivity index based on the risk gradient magnitude and the activity index to assess the degree of influence of the line impedance parameters on the risk value gradient direction.

[0158] 703. Sort the lines in the solution subset according to the sensitivity index, mark the lines whose risk gradient direction is opposite to the sensitivity index as high priority, and generate a cable replacement priority sequence by combining the magnitude and direction of impedance adjustment in the line capacity expansion scheme.

[0159] In step 703, the cable replacement priority sequence is a cable replacement plan generated based on sensitivity indicators and line capacity expansion schemes, typically used to guide cable replacement work. The risk gradient direction refers to the direction of change of risk values ​​in the cable aging risk distribution map, typically used to assess the spatial distribution characteristics of risk. The magnitude and direction of impedance adjustment in the line capacity expansion scheme refer to the magnitude and direction of line impedance adjustment, typically used to optimize the electrical performance of the power grid.

[0160] In this embodiment, the system first reads the sensitivity index and sorts the lines in the solution subset. Next, the system marks lines whose risk gradient direction is opposite to that of the sensitivity index as high priority, ensuring that high-priority lines can effectively reduce the risk of cable aging. Then, the system combines the magnitude and direction of impedance adjustment in the line capacity expansion scheme to generate a cable replacement priority sequence. Finally, the system stores the cable replacement priority sequence as structured data for subsequent cable replacement work.

[0161] Here is a specific example:

[0162] In a high-voltage power distribution network optimization project in a certain city, power engineer Wang was responsible for developing line capacity expansion and cable replacement plans. He first extracted candidate solutions that satisfied thermodynamic stability and cable aging risk tolerance from the optimal Pareto front of the feasible solution set. Based on the correlation rules between line impedance parameters and node voltage stability margins, he screened a subset of solutions that reduced the voltage deviation rate of critical load nodes below a preset threshold. For example, the voltage deviation rate of a critical load node in a certain substation decreased from 5% to 2%, significantly improving power supply stability. Next, for each line in the solution subset, Wang calculated the sensitivity index of its impedance parameters to the gradient direction of the risk value in the cable aging risk distribution map. For example, a certain line, located in a high-risk area with high spatial cumulative effect activity, had a larger sensitivity index. Based on the sensitivity index, Wang ranked the lines, marking lines with risk gradient directions opposite to the sensitivity index as high priority. For example, a line connecting an industrial area was prioritized for replacement. Finally, he combined the magnitude and direction of impedance adjustment in the line capacity expansion plan to generate a cable replacement priority sequence. After implementation, the operational stability of the high-voltage distribution network in the area was significantly improved, and the power company highly praised Engineer Wang's solution. This example demonstrates the practicality and efficiency of this method in the optimization of complex distribution networks.

[0163] In summary, steps 701 to 703 optimized the priority of line impedance adjustment and cable replacement, improved the accuracy and reliability of the planning scheme, and ensured the safe operation and sustainable development of the high-voltage distribution network.

[0164] In some embodiments, step 603, which involves recalculating the risk gradient direction of the cable aging risk distribution map based on the boundary changes of the spatial gridded partitions, and feeding the updated risk gradient direction back into the decay function of the dynamic weight coefficients to generate a new round of weight coefficient sharding, includes:

[0165] 801. Based on the boundary change of spatial gridded partitions, calculate the difference rate of cable aging risk values ​​between adjacent partitions and generate a new risk gradient direction vector;

[0166] In step 801, the boundary change of the spatial gridded partition refers to the change in the boundary of the grid area division, which is usually used to define the calculation range of the risk gradient direction. The cable aging risk value difference rate refers to the proportion of risk value difference between adjacent partitions, which is usually used to assess the spatial distribution characteristics of risk. The risk gradient direction vector is a vector generated based on the cable aging risk value difference rate, which is usually used to describe the direction of risk value change.

[0167] In this embodiment, the system first reads the boundary changes of the spatial gridded partitions to determine the range of change in the grid region boundaries. Next, the system calculates the difference rate of cable aging risk values ​​between adjacent partitions to assess the proportion of risk value change. Then, based on the cable aging risk value difference rate, the system generates a new risk gradient direction vector to describe the direction of risk value change. Finally, the system stores the risk gradient direction vector as structured data for subsequent adjustment of the attenuation function.

[0168] 802. Couple the risk gradient direction vector with the decay function of the dynamic weight coefficient, and adjust the ratio parameter of the distance decay factor and the risk correlation factor in the decay function;

[0169] In step 802, the decay function of the dynamic weighting coefficients is a function generated based on the risk gradient direction vector, typically used to dynamically adjust the decay trend of the weighting coefficients. The distance decay factor is a parameter in the decay function used to describe the influence of distance on the weighting coefficients, typically used to assess the trend of weighting coefficients changing with distance. The risk correlation factor is a parameter in the decay function used to describe the influence of risk on the weighting coefficients, typically used to assess the trend of weighting coefficients changing with risk. The proportional parameter is the ratio of the distance decay factor to the risk correlation factor in the decay function, typically used to adjust the decay trend of the decay function.

[0170] In this embodiment, the system first reads the risk gradient direction vector to determine the direction of risk value change. Next, the system couples the risk gradient direction vector with the decay function of the dynamic weight coefficients, adjusting the ratio parameters of the distance decay factor and the risk correlation factor in the decay function. Then, the system updates the decay function according to the adjusted ratio parameters, ensuring that the decay function can dynamically adjust the decay trend of the weight coefficients. Finally, the system stores the updated decay function as structured data for subsequent redefinition of the weight coefficient sharding rules.

[0171] 803. Based on the proportional parameter of the adjusted decay function, redefine the dynamic weight coefficient partitioning rules for each spatial gridded partition. The partition with the larger magnitude of the risk gradient direction vector and the higher the proportion of the decay factor has a smaller coverage range of its weight coefficient partitioning.

[0172] In step 803, the dynamic weighting coefficient sharding rule is a weighting coefficient sharding rule generated based on the decay function, typically used to optimize data storage and retrieval efficiency. The magnitude of the risk gradient direction vector refers to the length of the vector, typically used to assess the magnitude of risk value changes. The decay factor ratio is the ratio of the distance decay factor to the risk correlation factor in the decay function, typically used to adjust the decay trend of the decay function. The coverage of the weighting coefficient sharding refers to the storage range of the weighting coefficient shards, typically used to optimize data storage and retrieval efficiency.

[0173] In this embodiment, the system first reads the proportional parameter of the adjusted decay function to determine the ratio of the distance decay factor to the risk correlation factor. Next, based on the proportional parameter, the system redefines the dynamic weight coefficient partitioning rules for each spatial gridded partition, ensuring that partitions with larger magnitudes of the risk gradient direction vector and higher decay factor ratios have smaller weight coefficient partitioning coverage. Then, the system stores the updated weight coefficient partitioning rules as structured data for subsequent cross-node matching.

[0174] 804. By using parallel computing nodes, the updated weight coefficient slices are matched with the main modal features of the vibration spectrum across nodes to verify the consistency between the slice rules and the risk gradient direction vector, and the final weight coefficient slices are output.

[0175] In step 804, parallel computing nodes are computing nodes used to efficiently process large-scale data, typically implemented using distributed computing technology. Cross-node matching is the process of matching the updated weight coefficient shards with the main modal features of the vibration spectrum on different computing nodes, usually used to optimize computational efficiency. Consistency verification is the process of verifying whether the sharding rules are consistent with the risk gradient direction vector, usually used to evaluate the accuracy of the sharding rules. The final weight coefficient sharding is the weight coefficient sharding generated based on consistency verification, usually used to optimize data storage and retrieval efficiency.

[0176] In this embodiment, firstly, the system reads the updated weight coefficient slices and vibration spectrum main modal features, and performs cross-node matching through parallel computing nodes to optimize computational efficiency. Next, the system verifies the consistency between the slice rules and the risk gradient direction vector, evaluating the accuracy of the slice rules. Then, the system outputs the final weight coefficient slices based on the consistency verification results, ensuring that the slice rules are consistent with the risk gradient direction vector. Finally, the system stores the final weight coefficient slices as structured data for subsequent risk assessment.

[0177] Here is a specific example:

[0178] In a power distribution network optimization project in an industrial park, power engineer Li was responsible for updating the risk gradient direction of the cable aging risk distribution map and generating weighted coefficient partitions. He first calculated the difference rate of cable aging risk values ​​between adjacent partitions based on the boundary changes of the spatial gridded partitions. For example, the risk value difference rate between partition A and partition B increased from 10% to 15%, generating a new risk gradient direction vector. Next, Li coupled the risk gradient direction vector with the attenuation function of the dynamic weighted coefficients, adjusting the ratio of the distance attenuation factor to the risk correlation factor in the attenuation function. For example, he increased the ratio of the risk correlation factor from 0.6 to 0.8 to more accurately reflect the impact of high-risk areas. Based on the adjusted attenuation function, he redefined the dynamic weighted coefficient partitioning rules for each partition. For example, because partition C has a larger magnitude of the risk gradient direction vector and a higher attenuation factor ratio, the coverage area of ​​its weighted coefficient partitioning was reduced to half of its original size. Finally, Li Gong used parallel computing nodes to perform cross-node matching between the updated weight coefficient slices and the main modal features of the vibration spectrum, verifying the consistency between the slice rules and the risk gradient direction vector. For example, the slice rules for partition D perfectly matched the risk gradient direction, ensuring the accuracy of the risk distribution. After outputting the final weight coefficient slices, Li Gong successfully optimized the cable aging risk assessment model, providing a scientific basis for the power distribution network transformation of the industrial park, and received high recognition from the project team. This example demonstrates the efficiency and practicality of this method in the optimization of complex power distribution networks.

[0179] In summary, steps 801 to 804 optimized the risk gradient direction and dynamic weight coefficients, improving the accuracy and reliability of risk assessment and ensuring the safe operation and sustainable development of the high-voltage distribution network.

[0180] In some embodiments, step 104, which involves correcting the line corridor expansion parameters within the high-voltage distribution network planning area based on the high-voltage distribution network line capacity expansion scheme and cable replacement priority sequence, combined with the partial discharge frequency domain characteristic change trend in the cable vibration spectrum data, and outputting a dynamic planning graph containing cable life prediction nodes and capacity expansion paths, includes:

[0181] 901. Based on the high-priority lines marked in the cable replacement priority sequence, extract the frequency domain characteristic variation trend of partial discharge in the corresponding cable vibration spectrum data;

[0182] In step 901, the cable replacement priority sequence is a cable replacement plan generated based on sensitivity indicators and line capacity expansion schemes, typically used to guide cable replacement work. High-priority lines are those marked with higher priority in the cable replacement priority sequence, typically used for priority cable replacement. The variation trend of partial discharge frequency domain characteristics in cable vibration spectrum data refers to the variation law of partial discharge signal in the frequency domain, typically used to assess the aging trend of cables.

[0183] In this embodiment, the system first reads the cable replacement priority sequence to determine high-priority lines. Next, the system extracts the cable vibration spectrum data corresponding to the high-priority lines and analyzes the frequency domain characteristic variation trend of partial discharge. Then, the system stores the frequency domain characteristic variation trend as structured data for subsequent correction of line corridor expansion parameters.

[0184] 902. Based on the correlation rules between the frequency domain characteristic change trend and the line corridor expansion parameters, the high-priority line corridor expansion parameters within the high-voltage distribution network planning area are iteratively corrected.

[0185] In step 902, the line corridor extension parameter refers to the planned extension range of the line corridor, which is usually determined based on geospatial data and power grid demand. The correlation rule between the frequency domain characteristic change trend and the line corridor extension parameter is a correlation rule generated based on the frequency domain characteristic change trend, and is usually used to guide the correction of the line corridor extension parameter.

[0186] In this embodiment, the system first reads the correlation rules between the frequency domain characteristic change trend and the line corridor extension parameters to determine the correction direction. Next, the system iteratively corrects the extension parameters of high-priority line corridors within the high-voltage distribution network planning area to ensure that the extension parameters can adapt to the frequency domain characteristic change trend. Then, the system stores the corrected line corridor extension parameters as structured data for subsequent binding to the cable life prediction model.

[0187] 903. Bind the corrected line corridor extension parameters to the cable lifetime prediction model. The cable lifetime prediction model calculates the remaining lifetime nodes based on the cumulative effect of the change trend of partial discharge frequency domain characteristics, and associates the topology of the capacity expansion path with the risk suppression effect in the priority sequence.

[0188] In step 903, the cable life prediction model is a prediction model generated based on the trend of partial discharge frequency domain characteristics, and is typically used to assess the remaining life of cables. The remaining life node is the cable remaining life assessment result generated based on the cable life prediction model, and is typically used to guide cable replacement work. The topology of the capacity expansion path refers to the planned path for the expansion of high-voltage distribution network lines, and is typically determined based on line corridor expansion parameters and optimization schemes. The risk mitigation effect refers to the mitigation effect of high-priority lines in the cable replacement priority sequence on the risk of cable aging, and is typically used to assess the priority of cable replacement.

[0189] In this embodiment, firstly, the system reads the corrected line corridor expansion parameters and binds them to the cable lifetime prediction model. Next, the system calculates the remaining lifetime nodes based on the cumulative effect of the partial discharge frequency domain characteristic change trend, thus assessing the cable's remaining lifetime. Then, the system associates the topology of the capacity expansion path with the risk mitigation effect in the priority sequence to ensure that the capacity expansion path can effectively suppress the cable aging risk. Finally, the system stores the remaining lifetime nodes and the topology of the capacity expansion path as structured data for subsequent dynamic programming graph generation.

[0190] 904. By integrating the corrected extended parameters, cable life prediction nodes, and topological constraints of the capacity expansion path, a dynamic programming graph is generated that includes dynamic parameter update rules in the time dimension and risk hot zone markings in the spatial dimension.

[0191] In step 904, the dynamic programming map is a planning map that includes dynamic parameter update rules in the time dimension and risk hot zone markings in the spatial dimension. It is typically used to guide the long-term planning and maintenance of high-voltage distribution networks. The dynamic parameter update rules in the time dimension are parameter update rules generated based on cable life prediction models and are typically used to dynamically adjust planning schemes. The risk hot zone markings in the spatial dimension are risk markings generated based on cable aging risk distribution maps and are typically used to identify high-risk areas.

[0192] In this embodiment, the system first reads the corrected extended parameters, cable life prediction nodes, and topological constraints of the capacity expansion path to determine the dynamic parameter update rules in the time dimension and the risk hot zone markers in the spatial dimension. Next, the system generates a dynamic programming graph containing the dynamic parameter update rules in the time dimension and the risk hot zone markers in the spatial dimension, ensuring that the planning scheme can dynamically adapt to cable aging risks and grid expansion needs. Then, the system stores the dynamic programming graph as visual data for subsequent planning scheme generation.

[0193] Here is a specific example:

[0194] In a high-voltage power distribution network renovation project in a certain city, an electrical engineer was responsible for generating a dynamic programming graph. He first extracted the frequency domain characteristics of partial discharge from the frequency domain features of high-priority lines marked in the cable replacement priority sequence, such as lines connecting an industrial area, and found that the discharge frequency increased significantly over time. Based on the correlation rules between this trend and the line corridor expansion parameters, he iteratively corrected the corridor expansion parameters of high-priority lines. For example, he adjusted the expansion width of a certain line segment from 5 meters to 8 meters to meet future capacity expansion needs. Next, he bound the corrected expansion parameters to a cable life prediction model, calculating the remaining life nodes based on the cumulative effect of the partial discharge frequency domain characteristic change trend. For example, if the remaining life of a certain line is shortened from 10 years to 7 years, it should be prioritized for inclusion in the renovation plan. Simultaneously, he correlated the topology of the capacity expansion path with the risk mitigation effect in the priority sequence. For example, the newly added capacity expansion path bypasses high-risk areas, significantly reducing the probability of potential failures. Finally, he integrated the revised extended parameters, cable life prediction nodes, and topological constraints of the capacity expansion path to generate a dynamic programming graph that includes dynamic parameter update rules in the time dimension and risk hotspot markings in the spatial dimension. This graph provided the power company with a clear transformation path and priorities, ensuring the efficient implementation of the project and receiving high praise from leadership. This example demonstrates the scientific validity and practicality of this method in high-voltage distribution network planning.

[0195] In summary, steps 901 to 904 optimized the line corridor expansion and cable life prediction, improved the accuracy and reliability of the planning scheme, and ensured the safe operation and sustainable development of the high-voltage distribution network.

[0196] Figure 2 This application provides a schematic diagram of the structure of a power distribution network integrated planning system in power engineering design, as shown in the embodiment of the present application. Figure 2 As shown, the system includes:

[0197] Module 21 acquires geospatial data, historical operation data, and cable vibration spectrum data monitored by fiber optic sensor network within the high-voltage distribution network planning area.

[0198] Analysis module 22 performs distributed storage and correlation analysis on the geospatial data and the cable vibration spectrum data through a parallel computing architecture with a mixed integer programming parallel acceleration algorithm, and generates a cable aging risk distribution map based on spatial topology constraints, wherein the spatial topology constraints are dynamically defined by the geospatial relationship between the substation coordinates and the line corridor in the geospatial data.

[0199] The generation module 23 performs multi-objective collaborative optimization on the cable aging risk distribution map, the load peak-valley difference in the historical operation data, and the line corridor expansion parameters in the geospatial data based on the second-order cone programming model, and generates high-voltage distribution network line capacity expansion schemes and cable replacement priority sequences respectively. The constraints of the multi-objective collaborative optimization include the real-time threshold of the cable joint temperature fluctuation signal and the spatial accumulation effect of the partial discharge frequency domain characteristics.

[0200] The correction module 24, based on the high-voltage distribution network line capacity expansion scheme and cable replacement priority sequence, and combined with the partial discharge frequency domain characteristic change trend in the cable vibration spectrum data, corrects the line corridor expansion parameters in the high-voltage distribution network planning area, and outputs a dynamic planning map including cable life prediction nodes and capacity expansion paths.

[0201] Figure 2 The aforementioned power engineering design distribution network integrated planning system can execute... Figure 1 The implementation principle and technical effects of the power distribution network integrated planning method in the power engineering design described in the illustrated embodiment will not be repeated here. The specific operation methods of each module and unit in the power distribution network integrated planning system in the above embodiment have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0202] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A power distribution network integrated planning method in electric power engineering design, characterized in that, The method comprises the following steps: obtain the geographic spatial data, historical operation data and cable vibration spectrum data monitored by the optical fiber sensing network in the planning area of the high-voltage distribution network, wherein the cable vibration spectrum data refers to the vibration frequency characteristics of the cable monitored by the optical fiber sensing network in real time, including the partial discharge frequency domain characteristics and vibration amplitude; perform distributed storage and correlation analysis on the geographic spatial data and the cable vibration spectrum data through a mixed integer programming parallel acceleration algorithm and a parallel computing architecture to generate a cable aging risk distribution map based on spatial topology constraints; perform multi-objective collaborative optimization on the cable aging risk distribution map, the load peak-valley difference in the historical operation data and the line corridor expansion parameter in the geographic spatial data based on a second-order cone programming model to generate a high-voltage distribution network line capacity expansion scheme and a cable replacement priority sequence, wherein the multi-objective collaborative optimization takes maximizing the capacity expansion benefit, minimizing the aging risk and the reconstruction cost as objective functions, and the constraint conditions include the real-time threshold of the cable joint temperature fluctuation signal and the spatial cumulative effect of the partial discharge frequency domain characteristics, and the solution is obtained through relaxation processing and decomposition and coordination iteration; correct the line corridor expansion parameter in the planning area of the high-voltage distribution network according to the high-voltage distribution network line capacity expansion scheme and the cable replacement priority sequence, in combination with the change trend of the partial discharge frequency domain characteristics in the cable vibration spectrum data, and output a dynamic planning atlas containing a cable life prediction node and a capacity expansion path, wherein the correction process refers to the correlation rules based on the change trend of the partial discharge frequency domain characteristics and the line corridor expansion parameter, and the expansion width and length key indicators are adjusted iteratively to ensure the adaptation to the cable aging state; wherein the multi-objective collaborative optimization on the cable aging risk distribution map, the load peak-valley difference in the historical operation data and the line corridor expansion parameter based on the second-order cone programming model to generate the high-voltage distribution network line capacity expansion scheme and the cable replacement priority sequence comprises: construct a collaborative optimization model taking the risk value gradient in the cable aging risk distribution map, the time series deviation of the load peak-valley difference and the cost coefficient of the line corridor expansion parameter as optimization variables; wherein the collaborative optimization model is a multi-objective optimization model for comprehensively considering the cable aging risk, load fluctuation and economic cost factors to generate an optimal planning scheme; the risk value gradient refers to the change rate of the risk value in the cable aging risk distribution map for evaluating the spatial distribution characteristics of the risk; the time series deviation of the load peak-valley difference refers to the change of the difference between the peak value and the valley value of the grid load with time for evaluating the fluctuation characteristics of the grid load; and the cost coefficient of the line corridor expansion parameter refers to the economic cost of the line corridor expansion planning, which is determined based on the line length, terrain complexity and construction difficulty factors; define a set of thermodynamic constraint equations based on the real-time threshold of the cable joint temperature fluctuation signal, couple the energy integral of the partial discharge frequency domain characteristics in the cable vibration spectrum data in the spatial dimension with the preset safety threshold, and generate a spatial cumulative effect inequality constraint. The inequality constraint related to the spatial cumulative effect in the thermodynamic constraint equation set is relaxed by introducing an auxiliary variable, the auxiliary variable quantifies the constraint deviation and constructs a dynamic relaxation boundary; the spatial topological correlation of the cable aging risk distribution map is mapped to the boundary condition of the second-order cone programming model, and the coupling parameter matrix is defined by the topological adjacency relationship; based on the constraint relationship of the coupling parameter matrix and the dynamic relaxation boundary, an alternating iteration updating mechanism of the main variable and the auxiliary variable is adopted, the main variable updating stage solves the thermodynamic state parameter under the fixed relaxation boundary, the auxiliary variable updating stage dynamically adjusts the deviation and the penalty coefficient of the relaxation boundary according to the current main variable residual and the topological connection strength, and the iteration is stopped until the residual converges to the preset threshold; in the iteration process, the intermediate solution meeting the dynamic relaxation boundary and the coupling parameter matrix constraint is screened synchronously, the solution conflicting with the spatial topological correlation is eliminated by non-dominated sorting, and a feasible solution set with the target function value descending and the relaxation boundary deviation converging with the iteration is generated, wherein the calculation method of the feasible solution set includes: the auxiliary variable quantifies the constraint deviation, the topological adjacency relationship points to the direct physical connection of the node, the coupling parameter matrix element reflects the node correlation strength, the main variable solves the thermodynamic parameter, the auxiliary variable adjusts the relaxation boundary, and the iteration is stopped until the residual converges, the intermediate solution meeting the constraint is screened and the feasible solution set is generated by non-dominated sorting; According to the optimal Pareto front of the feasible solution set, a line capacity increasing scheme of the high-voltage distribution network is generated, and based on the risk value gradient direction of the cable aging risk distribution map and the activity index of the spatial cumulative effect inequality constraint, the line impedance parameters in the feasible solution set are sorted in sensitivity, and a cable replacement priority sequence is generated.

2. The method of claim 1, wherein, The spatial topological constraint is dynamically defined by the geographical spatial relationship between the substation coordinates and the line corridor in the geographical space data; The geographical space data and the cable vibration frequency spectrum data are distributedly stored and associated analyzed by a mixed integer programming parallel acceleration algorithm parallel computing architecture, to generate a cable aging risk distribution map based on the spatial topological constraint, including: The geographical space data is spatially gridded and partitioned based on the obtained segmented length of the line corridor, to generate a distributedly stored cable topological partition data set; The local discharge frequency domain features in the cable vibration frequency spectrum data are mapped to the corresponding cable topological partition data set according to the spatial gridding partition, to extract the vibration frequency spectrum main mode features in each partition; Based on the coordinate distance between the substation and the line corridor and the risk gradient direction, a dynamic weight coefficient of the spatial topological constraint is defined; The dynamic weight coefficient and the vibration frequency spectrum main mode features are cross-node matched by the parallel computing architecture, to dynamically correct the aging risk values in the cable topological partition data set, and to iteratively update the risk gradient direction according to the boundary of the spatial gridding partition, and output the cable aging risk distribution map based on the spatial topological constraint.

3. The method of claim 1, wherein, The spatial topological correlation of the cable aging risk distribution map is mapped to the boundary condition of the second-order cone programming model, and the coupling parameter matrix is defined by the topological adjacency relationship, including: extracting pairs of nodes directly topologically adjacent in the cable aging risk distribution map, calculating spatial correlation strength based on cable aging risk difference value and physical connection length between nodes; allocating coupling weight for each pair of nodes according to the spatial correlation strength, the coupling weight being negatively correlated with the spatial correlation strength, and the coupling weight of high-density area exponentially decaying with the increase of the number of adjacent nodes, combining the maximum allowed relaxation deviation dynamically set according to the area topology density to generate asymmetric element values of the coupling parameter matrix; mapping the non-diagonal elements of the coupling parameter matrix to the second-order cone constraint boundary conditions of the relaxation optimization model, and constraining the relaxation deviation vectors of adjacent nodes to satisfy that the module length of the difference value vector is not more than the product of the coupling weight and the maximum allowed relaxation deviation; multiplying the coupling weight by the maximum allowed relaxation deviation of the area where the coupling weight is located to generate matrix element values, and directly defining the coupling parameter matrix according to the matrix element values based on the dynamic setting of the area topology density.

4. The method of claim 2, wherein, cross-node matching the dynamic weight coefficient with the vibration frequency spectrum main modal characteristics through a parallel computing architecture, dynamically correcting the aging risk values in the cable topology partition data set, and iteratively updating the risk gradient direction according to the boundary of the spatial gridding partition to output the cable aging risk distribution map based on the spatial topology constraint, including: hashing the dynamic weight coefficient according to the number of the spatial gridding partition, and performing key-value matching with the vibration frequency spectrum main modal characteristics in the corresponding partition through a parallel computing node to simultaneously generate the dynamic weight coefficient; weighting and correcting the aging risk values in the cable topology partition data set based on the dynamic weight coefficient, and the higher the weight coefficient and the greater the energy of the vibration frequency spectrum main modal characteristics, the greater the correction amplitude of the aging risk value of the partition; recomputing the risk gradient direction of the cable aging risk distribution map according to the boundary change of the spatial gridding partition, and feeding back the updated risk gradient direction to the attenuation function of the dynamic weight coefficient to generate a new round of weight coefficient slicing; until the change rate of the aging risk values of all spatial gridding partitions is less than the convergence threshold, outputting the cable aging risk distribution map based on the spatial topology constraint according to the weight coefficient slicing.

5. The method of claim 1, wherein, generating a high-voltage distribution network line capacity expansion scheme according to the optimal Pareto front of the feasible solution set, and sorting the line impedance parameters in the feasible solution set according to the sensitivity based on the risk value gradient direction of the cable aging risk distribution map and the activity index of the spatial cumulative effect inequality constraint to generate a cable replacement priority sequence, including: extracting a candidate solution set satisfying the thermodynamic stability and cable aging risk tolerance from the optimal Pareto front of the feasible solution set, and generating a high-voltage distribution network line impedance adjustment amount based on the correlation rule of the line impedance parameter and the node voltage stability margin; for each line in the solution subset, calculating the sensitivity index of the impedance parameter of the line to the risk value gradient direction of the cable aging risk distribution map, the sensitivity index being determined by the risk gradient amplitude of the area where the line is located and the activity index of the spatial cumulative effect inequality constraint; According to the sensitivity index, the lines in the solution subset are sorted, the lines with the risk gradient direction opposite to the sensitivity index are marked as high priority, and the impedance adjustment amount in the line capacity expansion scheme is combined to generate a cable replacement priority sequence, wherein the thermodynamic stability refers to the temperature not exceeding the safety threshold, the aging risk tolerance refers to the risk value not exceeding the preset level, the sensitivity index reflects the response degree of the impedance parameter to the risk change, and the lines opposite to the risk gradient direction are high priority.

6. The method of claim 4, wherein, According to the boundary change of the spatial grid partition, the risk gradient direction of the cable aging risk distribution map is recalculated, and the updated risk gradient direction is fed back to the attenuation function of the dynamic weight coefficient to generate a new round of weight coefficient partition, including: Based on the boundary change of the spatial grid partition, the cable aging risk value difference rate between adjacent partitions is calculated to generate a new risk gradient direction vector; The risk gradient direction vector is coupled with the attenuation function of the dynamic weight coefficient to adjust the proportion parameter of the distance attenuation factor and the risk correlation factor in the attenuation function; Based on the proportion parameter of the adjusted attenuation function, the dynamic weight coefficient partition rule of each spatial grid partition is redefined, wherein the longer the modulus of the risk gradient direction vector and the higher the proportion of the attenuation factor, the smaller the coverage range of the weight coefficient partition; The updated weight coefficient partition and the vibration frequency spectrum main modal characteristics are matched across nodes by parallel computing nodes to verify the consistency of the partition rule and the risk gradient direction vector, and the final weight coefficient partition is output.

7. The method of claim 1, wherein, According to the line capacity expansion scheme of the high-voltage distribution network and the cable replacement priority sequence, the local discharge frequency domain feature change trend in the cable vibration frequency spectrum data is combined to correct the line corridor expansion parameter in the high-voltage distribution network planning area, and a dynamic planning map containing cable life prediction nodes and capacity expansion paths is output, including: According to the high-priority line marked in the cable replacement priority sequence, the frequency domain feature change trend of local discharge in the corresponding cable vibration frequency spectrum data is extracted, wherein the frequency domain feature change trend is determined by comparing data at different times; Based on the association rule between the frequency domain feature change trend and the line corridor expansion parameter, the high-priority line corridor expansion parameter in the high-voltage distribution network planning area is iteratively corrected, and the iteration correction needs to verify the parameter adaptation to geographical conditions, construction feasibility and cost until the parameter is reasonable; The corrected line corridor expansion parameter is bound with the cable life prediction model, which calculates the remaining life nodes according to the cumulative effect of the local discharge frequency domain feature change trend, and associates the topology of the capacity expansion path with the risk suppression effect in the priority sequence; The dynamic planning map containing the parameter dynamic update rule in the time dimension and the risk hot area label in the space dimension is generated by fusing the corrected expansion parameter, the cable life prediction node and the topology constraint of the capacity expansion path.

8. A power distribution network comprehensive planning system in a power engineering design, used for executing the power distribution network comprehensive planning method in any one of claims 1-7, characterized in that, including: The acquisition module acquires geographic space data, historical operation data and cable vibration spectrum data monitored by the optical fiber sensing network in the high-voltage power distribution network planning area. The cable vibration spectrum data refers to the vibration frequency characteristics of the cable monitored by the optical fiber sensing network in real time, including the partial discharge frequency domain characteristics and vibration amplitude. The analysis module performs distributed storage and correlation analysis on the geographic space data and the cable vibration spectrum data through a mixed integer programming parallel acceleration algorithm parallel computing architecture to generate a cable aging risk distribution map based on spatial topology constraints. The spatial topology constraints are dynamically defined by the geographic spatial relationship between the substation coordinates and the line corridor in the geographic space data. The generation module performs multi-objective collaborative optimization on the cable aging risk distribution map, the load peak-valley difference in the historical operation data and the line corridor expansion parameters in the geographic space data based on a second-order cone programming model to generate a high-voltage power distribution network line capacity expansion scheme and a cable replacement priority sequence, respectively. The constraint conditions of the multi-objective collaborative optimization include the real-time threshold of the cable joint temperature fluctuation signal and the spatial cumulative effect of the partial discharge frequency domain characteristics. The multi-objective collaborative optimization takes maximizing the capacity expansion benefit, minimizing the aging risk and the modification cost as the objective function, and solves it through relaxation processing and decomposition coordination iteration. The correction module corrects the line corridor expansion parameters in the high-voltage power distribution network planning area according to the high-voltage power distribution network line capacity expansion scheme and the cable replacement priority sequence, combined with the change trend of the partial discharge frequency domain characteristics in the cable vibration spectrum data, and outputs a dynamic planning atlas containing cable life prediction nodes and capacity expansion paths. The correction process refers to the correlation rules based on the change trend of the partial discharge frequency domain characteristics and the line corridor expansion parameters, and adjusts the expansion width and length key indicators through iteration to ensure the adaptation to the cable aging state.

Images