Power cable intelligent segmentation management method and system

Abstract

The invention discloses a power cable intelligent segmentation management method and system, and belongs to the field of power supply management, and the method comprises the steps: setting a first time sequence segmentation influence coefficient and a second time sequence segmentation influence coefficient, and building a dynamic decision node; acquiring cable layout information of a to-be-monitored power cable; configuring a backtracking time window, and establishing backtracking basic parameters of the power cable to be monitored; according to the backtracking basic parameters, adaptive analysis of balancing task requirements and balancing climate influences is carried out, and a first clustering constraint is generated based on an adaptive analysis result; and establishing a first clustering constraint according to the cable layout information, establishing a second clustering constraint according to the cable layout information, establishing a dynamic segmentation strategy according to the first clustering constraint, the second clustering constraint and the dynamic decision node, allocating an edge joint early warning network, and executing segmentation management based on the edge joint early warning network. A segmentation strategy is dynamically generated based on multi-dimensional data, precise and intelligent early warning and management of the power cable are realized, and the early warning response speed and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the field of power supply management, and in particular to a method and system for intelligent segmentation management of power cables. Background Art

[0002] The safe and reliable operation of power cables is directly related to the stability of power supply. With the continuous expansion and complexity of the power grid, the traditional power cable management method has been unable to meet the needs of modern power grids. At present, the power cable management system adopts fixed-cycle inspections or simple remote monitoring methods. However, this management method cannot adjust the segmentation strategy in real time according to task requirements and environmental changes. Specifically, the existing power cable management adopts a unified management strategy, which makes it difficult to carry out differentiated management according to the characteristics and risk levels of different cable segments. When facing a complex and changeable operating environment, it is impossible to adjust the monitoring frequency and focus in time, resulting in some sections being over-monitored and other sections being under-monitored. In addition, the existing power cable management is often too single when considering influencing factors, focusing mainly on parameters such as temperature or load, while ignoring the impact of comprehensive factors such as climate change and geographical location characteristics on the cable status. Therefore, the power cable management in the existing technology cannot adjust the segmentation strategy in real time according to task requirements and environmental changes, resulting in low early warning efficiency and accuracy. Summary of the invention

[0003] The present application provides a method and system for intelligent segmentation management of power cables, aiming to solve the technical problem in the prior art that power cable management cannot adjust the segmentation strategy in real time according to task requirements and environmental changes, resulting in low warning efficiency and accuracy.

[0004] In view of the above problems, the present application provides a method and system for intelligent segmentation management of power cables.

[0005] The first aspect disclosed in the present application provides a method for intelligent segmentation management of power cables, the method comprising: obtaining a pre-execution task of the power cable to be monitored, establishing a task cycle identifier with the pre-execution task, and setting a first time-series segmentation influence coefficient based on the task cycle identifier; performing a location climate forecast for the power cable to be monitored, and establishing a second time-series segmentation influence coefficient based on the location climate forecast result; inputting the first time-series segmentation influence coefficient and the second time-series segmentation influence coefficient into a dynamic decision model to establish a dynamic decision node; establishing a balanced task demand and a balanced climate impact between nodes with the dynamic decision node; obtaining a cable layout of the power cable to be monitored The cable laying information includes the cable laying structure information and the cable laying fixed environment information; the backtracking time window is configured, the backtracking time window is used to perform the backtracking monitoring of the power cables to be monitored, and the backtracking basic parameters of the power cables to be monitored are established; the adaptation analysis of the balanced task requirements and the balanced climate impact is performed according to the backtracking basic parameters, and the first clustering constraint is generated based on the adaptation analysis result; the second clustering constraint is established according to the cable laying information, and the dynamic segmentation strategy is established according to the first clustering constraint, the second clustering constraint, and the dynamic decision node, the edge joint early warning network is allocated based on the dynamic segmentation strategy, and the segmentation management is performed based on the edge joint early warning network.

[0006] Another aspect disclosed in the present application provides an intelligent segmentation management system for electric power cables, the system comprising: a first influence coefficient module, used to obtain a pre-execution task for the electric power cable to be monitored, establish a task cycle identifier with the pre-execution task, and set a first time-series segmentation influence coefficient based on the task cycle identifier; a second influence coefficient module, used to perform location climate prediction for the electric power cable to be monitored, and establish a second time-series segmentation influence coefficient based on the location climate prediction result; a decision node generation module, used to input the first time-series segmentation influence coefficient and the second time-series segmentation influence coefficient into a dynamic decision model to establish a dynamic decision node; a node balance analysis module, used to establish balanced task requirements and balanced climate impacts between nodes with dynamic decision nodes; a cable information acquisition module, used to obtain the electric power cable to be monitored The cable laying information of the power cable includes the cable laying structure information and the cable laying fixed environment information; the backtracking parameter establishment module is used to configure the backtracking time window, to perform the backtracking monitoring of the power cable to be monitored with the backtracking time window, and to establish the backtracking basic parameters of the power cable to be monitored; the clustering constraint generation module is used for the first clustering constraint module to perform the adaptation analysis of the balanced task requirements and the balanced climate impact according to the backtracking basic parameters, and to generate the first clustering constraint based on the adaptation analysis result; the segmentation management execution module is used to establish the second clustering constraint according to the cable laying information, to establish the dynamic segmentation strategy according to the first clustering constraint, the second clustering constraint and the dynamic decision node, to allocate the edge joint early warning network based on the dynamic segmentation strategy, and to perform segmentation management based on the edge joint early warning network.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: By adopting the method of obtaining the pre-execution task of the power cable to be monitored, establishing the task cycle identification and setting the first time segment influence coefficient, basic data is provided for the subsequent segment management, ensuring that the management strategy can be adjusted according to the specific task requirements; the location climate prediction of the power cable to be monitored is performed, and the second time segment influence coefficient is established based on the prediction result, considering the impact of environmental factors on the cable status and enhancing the adaptability of the management strategy; the first time segment influence coefficient and the second time segment influence coefficient are input into the dynamic decision model to establish a dynamic decision node, realizing the comprehensive consideration of task requirements and environmental factors, and laying the foundation for the formulation of intelligent management strategies; dynamic decision nodes are used to establish balanced task requirements and balanced climate impacts between nodes, ensuring the balance of tasks and environmental factors between different decision nodes, and improving the coordination of the overall management strategy; the cable layout information of the power cable to be monitored is obtained, including structural information and fixed environment information, and the physical characteristics of the cable are taken into consideration, further improving the decision basis; Configure the retrospective time window, conduct retrospective monitoring of the power cables to be monitored, establish retrospective basic parameters, and provide a reference for current decision-making through historical data analysis; conduct adaptation analysis of balanced task requirements and balanced climate impacts based on the retrospective basic parameters, generate the first clustering constraint, realize the combination of historical data and current requirements, and improve the accuracy of decision-making; establish the second clustering constraint based on the cable layout information, and further refine the management strategy to make it more in line with the actual cable layout situation; establish a dynamic segmentation strategy based on the first clustering constraint, the second clustering constraint and the dynamic decision node, allocate the edge joint early warning network, and perform segmentation management based on this, realizing a technical solution for efficient and accurate segmentation management, solving the technical problem that the power cable management in the existing technology cannot adjust the segmentation strategy in real time according to task requirements and environmental changes, resulting in low early warning efficiency and accuracy, and achieving the technical effect of dynamically generating segmentation strategies based on multi-dimensional data, realizing precise and intelligent early warning and management of power cables, and improving the early warning response speed and accuracy.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A schematic flow chart of a method for intelligent segmentation management of power cables is provided for an embodiment of the present application; Figure 2 A structural schematic diagram of a power cable intelligent segmentation management system is provided for an embodiment of the present application.

[0010] Explanation of the reference numerals: first influence coefficient module 11 , second influence coefficient module 12 , decision node generation module 13 , node balance analysis module 14 , cable information acquisition module 15 , backtracking parameter establishment module 16 , clustering constraint generation module 17 , segment management execution module 18 . DETAILED DESCRIPTION

[0011] The overall idea of ​​the technical solution provided by this application is as follows: The embodiments of the present application provide a method and system for intelligent segmentation management of power cables, which realize real-time optimization of power cable management strategies and precise segmentation warning through comprehensive analysis of multi-dimensional data and application of dynamic decision-making models.

[0012] First, the pre-execution task information and location climate forecast data of the power cables to be monitored are obtained, and the first time-series segmentation influence coefficient and the second time-series segmentation influence coefficient are established to lay the foundation for dynamic decision-making. Subsequently, decision nodes are constructed through the dynamic decision-making model, and a balanced task demand and climate impact relationship is established between nodes. Next, cable layout information and historical monitoring data are introduced, and clustering constraints are generated through retrospective analysis and adaptation analysis to further refine the management strategy. After that, a dynamic segmentation strategy is established based on the obtained factors, and precise segmentation management is performed through the edge joint early warning network, thereby improving the response speed and accuracy of power cable early warning.

[0013] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0014] Embodiment 1, as Figure 1 As shown, an embodiment of the present application provides a method for intelligent segmentation management of power cables, the method comprising: S1: Acquire a pre-execution task of a power cable to be monitored, establish a task cycle identifier with the pre-execution task, and set a first time segment influence coefficient based on the task cycle identifier.

[0015] Specifically, first, obtain the pre-execution tasks of the power cables to be monitored in the future. The pre-execution tasks can be routine inspections, maintenance, etc. The purpose of obtaining the pre-execution tasks is to understand the working status and task arrangements of the power cables. Then, according to the obtained pre-execution tasks, establish the corresponding task cycle identification. The task cycle identification reflects the periodic characteristics of the pre-execution tasks, such as daily, weekly, monthly, etc. Through the task cycle identification, the pre-execution tasks can be classified and divided according to the cycle. After establishing the task cycle identification, the first time segmentation influence coefficient is set based on the task cycle identification. The first time segmentation influence coefficient quantifies the degree of influence of different task cycles on the segmentation management of power cables. By setting this coefficient, the influence of different task cycles on segmentation management is quantified, providing a basis for subsequent dynamic decision-making.

[0016] By obtaining pre-execution tasks, establishing task cycle identifiers, and setting the first time segmentation impact coefficient, basic data and quantitative influencing factors are provided for the subsequent intelligent segmentation management of power cables, and the task characteristics of power cables at different time scales are obtained and converted into quantitative indicators that can be used for segmentation management decisions.

[0017] S2: Execute location climate prediction for the power cable to be monitored, and establish a second time series segment influence coefficient based on the location climate prediction result.

[0018] Specifically, by analyzing the historical climate data of the area where the power cables are located and using the climate prediction model, the climate change trend and the possibility of extreme weather events in the area in the future can be obtained. Among them, the location climate prediction results include changes in multiple climate elements such as temperature, humidity, precipitation, and wind speed. According to the predicted values ​​of these climate elements, the climate environment in which the power cables are located in different time periods can be determined, and the degree of influence of climate conditions on the cables can be evaluated. For example, by analyzing the historical climate data of the past 10 years and using the ARIMA time series model, it is predicted that the average temperature in a certain area in the next month will be 3°C higher than the same period in history, the precipitation will be 20% lower than the same period in history, and there is a 30% probability of high temperature weather with a daily maximum temperature exceeding 35°C in the next month. According to the predicted values ​​of these climate elements, the climate environment in which the power cables are located in different time periods can be determined, and the degree of influence of climate conditions on the cables can be evaluated.

[0019] Then, based on the location climate prediction results, the second time-series segmentation influence coefficient is established. The second time-series segmentation influence coefficient quantifies the degree of influence of different climate conditions on the segmentation management of power cables. For example, extreme weather such as high temperature and heavy precipitation will accelerate the aging of cables and increase the risk of failure, so the corresponding second time-series segmentation influence coefficient will be higher. Under mild climate conditions, the operating environment of the cable is relatively stable, and the corresponding influence coefficient is lower. For example, according to the location climate prediction results, the probability of high temperature weather with a daily maximum temperature exceeding 35°C in the next month is 30%, and high temperature weather will accelerate the aging of cable insulation. Therefore, the "probability of high temperature weather in the next month" is used as an influence factor and multiplied by a weight coefficient to obtain the contribution value of the second time-series segmentation influence coefficient. Similarly, the "extent of the increase in average temperature in the next month compared with the same period in history" is used as another influence factor and multiplied by another weight coefficient to obtain the corresponding contribution value; the contribution values ​​of all influence factors are added together to obtain the second time-series segmentation influence coefficient.

[0020] By establishing the second time-series segmentation influence coefficient, the impact of climate factors on the segmentation management of power cables is quantified, providing a more comprehensive basis for subsequent dynamic decision-making, and incorporating the climate and environmental factors of the power cables into the consideration of segmentation management, thereby improving the accuracy and adaptability of segmentation management.

[0021] S3: Inputting the first time segment influence coefficient and the second time segment influence coefficient into a dynamic decision model to establish a dynamic decision node.

[0022] Specifically, the dynamic decision model can dynamically adjust the computer model of the decision plan according to the real-time input data. The dynamic decision model is based on artificial intelligence algorithms such as reinforcement learning and decision trees. It can autonomously generate and optimize decision plans by learning historical data and analyzing the current state. The input data of the dynamic decision model includes the first time segmentation influence coefficient and the second time segmentation influence coefficient, and the output result is a dynamic decision node. The first time segmentation influence coefficient reflects the working status and task characteristics of the power cable under different task cycles, and the second time segmentation influence coefficient reflects the changing trend of the climate environment in which the power cable is located and the possibility of extreme weather events. These two influence coefficients are input into the dynamic decision model, and the dynamic decision model comprehensively considers the influence of the task cycle and the climate environment to establish a dynamic decision node. Each dynamic decision node represents the optimal segmentation management plan for the power cable in a specific time period. For example, the segmentation management plan of a dynamic decision node: in the next week, the A and B sections of the power cable will be monitored in detail, and the inspection will be carried out twice a day. The detection indicators include wire temperature and tension; the C and D sections will be routinely monitored, and the inspection will be carried out once a week. The detection indicators include wire temperature and insulation resistance.

[0023] Since the decision nodes in the dynamic decision model are dynamically generated based on the impact coefficient of real-time input, the content of the dynamic decision nodes will change dynamically over time. For example, if the impact coefficient of the second time segment changes in the next week (such as the meteorological department issues a high temperature warning), the dynamic decision model will automatically adjust the decision node, increase the monitoring frequency of sections A and B, or add the surface contamination of the conductor as a detection indicator. Through dynamic adjustment, the segment management plan can adapt to changes in task requirements and climate environment in real time, improving the accuracy and timeliness of management.

[0024] By inputting the first time segment influence coefficient and the second time segment influence coefficient into the dynamic decision-making model, a dynamic decision node is established. Each dynamic decision node represents the optimal segment management plan for power cables in a certain time period. It can comprehensively consider the impact of task cycle and climate environment, and dynamically adjust according to real-time input data, thereby realizing the precision and adaptability of power cable management and providing a decision-making basis for subsequent segment management execution.

[0025] S4: Using the dynamic decision nodes to establish balanced task requirements and balanced climate impacts among nodes.

[0026] Specifically, since each dynamic decision node represents the optimal segmented management plan for power cables in a certain time period, there are differences in task requirements and climate impacts between different decision nodes. For example, in the task-intensive time period, the dynamic decision node is assigned more inspection and maintenance tasks; in the time period of frequent extreme weather, the dynamic decision node is assigned more monitoring and protection measures. This difference between nodes will lead to unbalanced segmented management, with management tasks in some time periods being too heavy and in other time periods being too light.

[0027] Therefore, on the basis of dynamic decision nodes, balanced task requirements and balanced climate impacts are established between nodes. Balanced task requirements refer to adjusting and balancing the task requirements of different decision nodes while keeping the overall task requirements unchanged, so that the amount of management tasks in each time period is as balanced as possible. For example, some tasks in the task-intensive time period are transferred to the time period with fewer tasks, or some preventive maintenance tasks are arranged in the time period with fewer tasks to balance the distribution of tasks in the entire management cycle. Balanced climate impacts refer to adjusting and balancing the climate impacts of different decision nodes while keeping the overall climate impact unchanged, so that the adaptability of the management plans in each time period to climate change is as consistent as possible. For example, for time periods with frequent extreme weather, the monitoring frequency and protective measures in adjacent time periods are appropriately increased to improve the climate adaptability in the entire management cycle; for time periods with relatively stable climate conditions, the monitoring and protection requirements in adjacent time periods are appropriately simplified to balance management costs and management effects.

[0028] In the process of establishing balanced task requirements and balanced climate impacts, multiple factors need to be considered comprehensively, such as the overall task volume, climate change trends, management costs, etc. For example, optimization algorithms (such as integer programming, dynamic programming, etc.) can be used to solve the optimal balanced solution, or heuristic algorithms (such as simulated annealing, genetic algorithms, etc.) can be used to quickly search for approximate optimal solutions. The purpose of establishing balanced task requirements and balanced climate impacts is to ensure that power cables are managed relatively evenly and consistently throughout the management cycle, avoid drastic fluctuations in management tasks and management effects, improve the integrity and stability of segmented management, and further optimize and adjust dynamic decision nodes, so that segmented management solutions are smoother and more coherent, and more able to adapt to the long-term management needs of power cables.

[0029] By establishing balanced task requirements and balanced climate impacts between dynamic decision-making nodes, the time distribution and climate adaptability of the segmented management plan are optimized, the integrity and stability of management are improved, thereby achieving precise and long-term management.

[0030] S5: Acquire cable laying information of the power cable to be monitored, wherein the cable laying information includes cable laying structure information and cable laying fixing environment information.

[0031] Specifically, by consulting historical data such as engineering design drawings and construction records, the structural parameters and environmental conditions of cable laying are extracted to obtain the cable laying information of the power cables to be monitored, including cable laying structure information and cable laying fixed environment information. Cable laying structure information refers to the spatial laying method and structural characteristics of power cables, including cable laying paths (such as along streets, through forest areas, etc.), laying forms (such as overhead, underground, etc.), cable models and specifications (such as conductor cross-sectional area, insulation grade, etc.), pole tower types and spacing, etc., reflecting the characteristics of the physical structure of power cables and is an important factor affecting the stress, heat dissipation, and insulation performance of cables. By obtaining cable laying structure information, the spatial distribution and structural characteristics of the power cables to be monitored are mastered, providing basic data for subsequent segment management. Cable laying fixed environment information refers to the surrounding environment and fixed conditions of the power cables, including the geological conditions (such as soil quality, topography, etc.) through which the cables pass, the distribution of surrounding buildings and trees, the type and status of cable support and fixing devices (such as poles, guy wires, etc.), etc., reflecting the characteristics of the environment in which the power cables are located. By obtaining the fixed environment information of the cables, the characteristics and potential risks of the environment in which the power cables to be monitored are understood, providing an environmental basis for subsequent segment management.

[0032] By obtaining the cable layout information of the power cables to be monitored, including the cable layout structure information and the cable layout fixing environment information, the spatial distribution characteristics of the power cables and the characteristics of the environment in which they are located are mastered, and the influencing factors of the power cables' stress, heat dissipation, insulation, corrosion, etc. are reflected, providing an important basis for precise segmentation management.

[0033] S6: configuring a retrospective time window, performing retrospective monitoring of the power cable to be monitored with the retrospective time window, and establishing retrospective basic parameters of the power cable to be monitored.

[0034] Specifically, the backtracking time window refers to a period of time before the current moment, which is used to perform backtracking analysis on the historical monitoring data of the power cable. The length of the backtracking time window is set according to actual needs, for example, from a few months to a year. By configuring the backtracking time window, the time range of the historical monitoring data of the power cable to be monitored can be determined to facilitate its analysis and mining.

[0035] After determining the retrospective time window, retrospective monitoring is performed on the power cables to be monitored within the retrospective time window. Retrospective monitoring refers to the process of collecting, organizing and analyzing the historical monitoring data of the power cables to be monitored within the retrospective time window. Among them, the historical monitoring data includes data in multiple dimensions such as the electrical parameters of the cable (such as current, voltage, power, etc.), environmental parameters (such as temperature, humidity, wind speed, etc.), and mechanical parameters (such as tension, vibration, tilt, etc.). Through retrospective monitoring, the operating status and performance of the power cables to be monitored in the past period of time can be fully understood, providing a historical basis for subsequent segment management.

[0036] After completing the retrospective monitoring, the retrospective basic parameters of the power cable to be monitored are established based on the retrospective monitoring data. Retrospective basic parameters refer to the key indicators that reflect the historical operation characteristics of the power cable, which are extracted by analyzing and mining the retrospective monitoring data, including the average current, maximum current, average temperature, maximum temperature, average wind speed, maximum wind speed, conductor vibration frequency, tower inclination angle, etc. of the cable. By establishing the retrospective basic parameters, the operating status and performance level of the power cable in the retrospective time window can be quantitatively described, providing a quantitative basis for subsequent segment management decisions. For example, by statistically analyzing the current data in the retrospective time window, the average current and maximum current of the cable and other parameters are obtained to reflect the load level and overload risk of the cable; by analyzing the temperature data in the retrospective time window, the average temperature and maximum temperature of the cable and other parameters are obtained to reflect the heat dissipation status and overtemperature risk of the cable; by performing spectrum analysis on the vibration data in the retrospective time window, the main frequency and amplitude of the conductor vibration and other parameters are obtained to reflect the mechanical vibration characteristics and fatigue risk of the cable.

[0037] By configuring the retrospective time window, retrospective monitoring is carried out on the monitored power cables, and the retrospective basic parameters are established based on the retrospective monitoring data. The historical operation big data of the power cables is fully utilized, and key indicators reflecting the characteristics of the cables are extracted through data mining, providing a comprehensive and quantitative basis for subsequent segmented management decisions.

[0038] S7: Performing an adaptation analysis of the equilibrium task requirements and the equilibrium climate impacts according to the backtracking basic parameters, and generating a first clustering constraint based on the adaptation analysis result.

[0039] Specifically, adaptation analysis refers to matching and associating the retrospective basic parameters with the balanced task requirements and balanced climate impacts, and evaluating the degree of adaptation and correlation between the two. Through adaptation analysis, the inherent connection between the operating characteristics of power cables in historical periods and the current management tasks and environmental conditions is revealed, providing a basis for optimizing the segmented management plan. When conducting adaptation analysis, the retrospective basic parameters are compared and associated with the balanced task requirements and balanced climate impacts in multiple dimensions. For example, the average current parameter of the cable is compared with the load prediction value in the task requirements to evaluate the degree of matching between the power supply capacity of the cable and the load requirements; the maximum temperature parameter of the cable is compared with the high temperature warning value in the climate impact to evaluate the degree of adaptability of the heat dissipation capacity of the cable to the ambient temperature; the failure history of the cable is associated with the task requirements and climate impacts to evaluate the cable failure risk under different tasks and climate conditions. Through adaptation analysis, the degree of adaptation matrix or degree of adaptation between the retrospective basic parameters and the balanced task requirements and balanced climate impacts is obtained. The fitness matrix is ​​a multidimensional matrix, in which each element represents the fitness between a certain backtracking basic parameter and a certain task requirement or climate influencing factor, and the value range is [0, 1] or [-1, 1], etc. The fitness level is a level label obtained by discretizing the fitness matrix, such as "high fitness", "moderate fitness", "low fitness", etc.

[0040] Subsequently, based on the adaptation degree matrix or adaptation degree level obtained by the adaptation analysis, the first clustering constraint is generated. The first clustering constraint refers to the parameter similarity constraint condition that needs to be met when the power cables are segmented and clustered. Specifically, the retrospective basic parameters and balanced task requirements / balanced climate impacts with a higher degree of adaptation should be clustered into the same cable segment; while the parameters and factors with a lower degree of adaptation should be clustered into different cable segments. In this way, it is ensured that the parameter similarity within each cable segment is high, while the parameter differences between segments are large, which helps to achieve refined and differentiated segment management. For example, if the adaptation analysis finds that the average current parameter of cable A is highly adapted to the load prediction value of a heavy-load line, while the average current parameter of cable B is poorly adapted to the load prediction value, a first clustering constraint is generated: cable A should be clustered with the heavy-load line in the same management segment, and cable B should not be clustered with the heavy-load line.

[0041] Through adaptation analysis, the retrospective basic parameters are matched and associated with the equilibrium task requirements and equilibrium climate impacts in multiple dimensions, and the first clustering constraint is generated based on the adaptation analysis results. The connection between the historical operation characteristics of the cable and the current management needs is established, which provides parameter similarity constraints for subsequent segment clustering, helps to achieve refined and differentiated segment management, and improves the pertinence and effectiveness of cable management.

[0042] S8: Establish a second clustering constraint according to the cable laying information, establish a dynamic segmentation strategy according to the first clustering constraint, the second clustering constraint, and the dynamic decision node, allocate an edge joint warning network based on the dynamic segmentation strategy, and perform segmentation management based on the edge joint warning network.

[0043] Specifically, the second clustering constraint refers to the cable layout similarity constraint that needs to be met when the power cables are segmented and clustered. The cable layout information includes the cable layout structure information and the cable layout fixed environment information, which reflects the spatial distribution characteristics of the power cables and the characteristics of the environment in which they are located. Cables with similar layout structures and similar layout environments usually have similar physical properties and external influencing factors, so they should be clustered into the same management segment. For example, for cables with the same conductor cross-sectional area, the same conductor material, and the same tower model, their physical parameters such as conductivity, mechanical strength, thermal expansion and contraction characteristics are very close, so a second clustering constraint can be established: requiring these cables to be clustered into the same management segment. For another example, for cables with similar terrain conditions, similar distribution of surrounding buildings, and similar ambient temperatures, the external physical influences (such as wind load, ice load, solar radiation, etc.) they are subject to are usually very similar, so a second clustering constraint can also be established: requiring these cables to be clustered into the same management segment as much as possible.

[0044] After establishing the first clustering constraint and the second clustering constraint, they are integrated with the established dynamic decision nodes to establish a dynamic segmentation strategy. The dynamic segmentation strategy refers to a strategy for dynamically segmenting power cables based on multiple clustering constraints and decision nodes. It stipulates what segmentation schemes should be adopted at different time nodes, different task requirements, and different environmental conditions, as well as how to dynamically adjust and optimize between different segments. Among them, the establishment of the dynamic segmentation strategy adopts artificial intelligence algorithms such as multi-objective optimization, decision tree learning, and reinforcement learning. By modeling and optimizing multiple constraints and decision variables such as the first clustering constraint, the second clustering constraint, and the dynamic decision node, a comprehensive and dynamically changing segmentation strategy is obtained. It can adaptively adjust the segmentation plan according to multiple factors such as the historical characteristics of the cable, current needs, and future predictions, balance various management goals, and achieve dynamic optimization of segmentation management.

[0045] After that, the edge joint warning network is further allocated based on the dynamic segmentation strategy. The edge joint warning network is a distributed and collaborative cable monitoring and warning system, which consists of multiple monitoring terminals, communication nodes, data processing units, etc. distributed on different cable segments. By mapping the dynamic segmentation strategy to the edge joint warning network, differentiated monitoring and warning of different cable segments can be achieved, improving the accuracy of monitoring and the timeliness of warning. For example, according to the dynamic segmentation strategy, important cable segments and high-risk cable segments are allocated to high-performance monitoring terminals and high-speed communication nodes in the edge joint warning network to achieve key monitoring and rapid warning of these cables; while ordinary cable segments and low-risk cable segments are allocated to conventional monitoring terminals and low-speed communication nodes in the edge joint warning network to achieve daily monitoring and regular warning of these cables. After that, segment management is performed based on the allocated edge joint warning network. Segment management refers to the implementation of differentiated and refined management measures for different cable segments based on dynamic segmentation strategies and edge joint warning networks, including real-time monitoring, status assessment, risk warning, maintenance decision-making, etc. By continuously collecting monitoring data from each cable segment through the edge joint early warning network and performing analysis and decision-making based on dynamic segmentation strategies, it is possible to achieve all-round, full-cycle, and full-process management of cables and improve the safety, reliability, and economy of cable operation.

[0046] By establishing the second clustering constraint based on the cable laying information, and then integrating the first clustering constraint, the second clustering constraint, and the dynamic decision node, a dynamic segmentation strategy is established. Based on the dynamic segmentation strategy, an edge joint early warning network is allocated. Segmentation management is performed according to the edge joint early warning network. The intelligent upgrade of power cables from static segmentation to dynamic management is fully realized, and efficient, accurate, and coordinated segmentation management of cables throughout their life cycle is achieved.

[0047] Furthermore, the embodiment of the present application also includes: Obtain cable material information of the power cable to be monitored, perform a material retrospective impact assessment of the cable based on the cable material information, and establish an aging retrospective score; Conducting a seasonal climate change evaluation on the power cables to be monitored, establishing a climate cycle, obtaining the frequency of occurrence of extreme weather in the region, and constructing a seasonal climate change score and an extreme weather event frequency score respectively according to the climate cycle and the frequency of occurrence; A comprehensive score is established based on the aging backtracking score, seasonal climate change score, and extreme weather event frequency score, and the length of the backtracking time window is determined by the comprehensive score to complete the backtracking time window configuration.

[0048] In a feasible implementation, firstly, the cable material information of the power cable to be monitored is obtained, and the cable material information includes the conductor material (such as copper, aluminum, etc.), insulation material (such as cross-linked polyethylene, silicone rubber, etc.), sheath material (such as polyvinyl chloride, polyethylene, etc.), etc. Different materials have different physical and chemical properties, so their aging processes and influencing factors are also different. Based on the obtained cable material information, the cable material retrospective impact evaluation is performed. Material retrospective impact evaluation refers to analyzing the aging laws and influencing factors of different materials during long-term operation, and evaluating the degree of influence of material aging on cable performance. For example, for copper conductors, the evaluation indicators include conductivity, tensile strength, elongation, etc., and the influencing factors include temperature, humidity, vibration, etc.; for cross-linked polyethylene insulation, the evaluation indicators include insulation resistance, dielectric loss, breakdown strength, etc., and the influencing factors include temperature, electric field strength, partial discharge, etc. Through the material retrospective impact evaluation, a quantitative aging retrospective score is obtained, which reflects the aging degree and performance attenuation of the cable material during long-term operation. The higher the score, the more serious the aging of the cable material, the greater the impact on the cable performance, and the longer the retrospective time window is required for monitoring and analysis.

[0049] Then, seasonal climate change evaluation is conducted on the area where the monitored power cables are located. The climate conditions in different regions vary greatly, so it is necessary to analyze the impact of climate factors on cables in a targeted manner. First, a climate cycle is established. The climate cycle reflects the periodic change law of regional climate on an annual scale, such as the alternation of four seasons, the synchronization of rain and heat, etc. Based on the meteorological data of the region for many years, Fourier analysis, wavelet analysis and other methods are used to extract the climate cycle characteristics. Secondly, the frequency of extreme weather events in the region is obtained. Extreme weather events such as typhoons, rainstorms, and ice disasters will pose a serious threat to power cables, so they need to be paid special attention to. The frequency of occurrence of extreme weather events in historical meteorological data is calculated by counting the number of occurrences of extreme weather events. Then, according to the climate cycle and the frequency of occurrence of extreme weather events, seasonal climate change scores and extreme weather event frequency scores are constructed respectively. The seasonal climate change score reflects the degree of influence of the periodic characteristics of the regional climate on the performance of the cable. The fuzzy comprehensive evaluation, hierarchical analysis and other methods are used to associate the climate cycle characteristics with the impact on the performance of the cable to form a seasonal climate change score. The extreme weather event frequency score reflects the degree of threat that extreme weather events pose to cable performance. The event frequency is mapped to a fixed scoring range. The higher the frequency, the higher the extreme weather event frequency score.

[0050] Subsequently, a comprehensive score is established based on the aging retrospective score, seasonal climate change score, and extreme weather event frequency score. The comprehensive score reflects the comprehensive impact of multiple factors such as cable materials and regional climate conditions on cable performance, and is the direct basis for determining the length of the retrospective time window. For example, the three scoring indicators are integrated into a comprehensive score using weighted average, geometric average, and other methods. Afterwards, the length of the retrospective time window is determined based on the comprehensive score. Among them, the higher the comprehensive score value, the more serious the aging of the cable material and the worse the regional climate environment, so a longer retrospective time window is required for comprehensive historical data analysis; conversely, if the comprehensive score value is low, the length of the retrospective time window is appropriately shortened.

[0051] Through multiple dimensions such as cable material evaluation and regional climate evaluation, the key factors affecting cable performance were quantitatively evaluated, and the length of the traceback time window was determined based on the comprehensive score, which more comprehensively and objectively reflected the actual operating status of the cable, helped select the optimal traceback time window, and improved the pertinence and effectiveness of cable monitoring and analysis.

[0052] Furthermore, the embodiment of the present application also includes: Configure the comprehensive scoring formula as follows: ; in, For the comprehensive rating, Represent any scoring feature, Features The corresponding weight, Characterization characteristics The scoring function is: ; ; ; Characterizes aging retrospective scoring, and is the linear coefficient, Characterize the cable aging speed characteristics based on cable material information, Characterize seasonal climate change scores, Characterize the amplitude coefficient, Characterization cycle, is the input variable of seasonal climate change, Characterize the frequency score of extreme weather events, and is the adjustment coefficient, is the frequency of occurrence; ; is the length of the lookback time window, and They are the shortest and longest optional values ​​of the time window respectively.

[0053] In a preferred embodiment, a specific comprehensive score calculation method is given. By configuring a comprehensive score formula, the aging retrospective score, seasonal climate change score and extreme weather event frequency score are quantitatively integrated to obtain a comprehensive score that comprehensively reflects the cable status, and the length of the retrospective time window is determined based on the comprehensive score.

[0054] Specifically, the comprehensive scoring formula adopts the form of weighted summation to obtain the comprehensive score. The comprehensive scoring formula is: ;in, For the comprehensive rating, Represents any scoring indicator, with a value range of 1 to 3, corresponding to the aging retrospective score, seasonal climate change score, and extreme weather event frequency score; Indicates The weight coefficient of each scoring indicator reflects the importance of the indicator in the comprehensive scoring; Indicates The scoring function of the scoring indicators is used to map the original indicator value into a unified scoring interval. For the scoring functions of the three scoring indicators, the specific function forms are given. The scoring function of the aging retrospective scoring is: ,in Indicates the cable aging speed characteristics obtained based on cable material information. and is a linear coefficient used to adjust the slope and intercept of the linear function, reflecting the positive correlation between the cable aging speed characteristics and the aging retrospective score. The scoring function of the seasonal climate change score is: ,in is the input variable of seasonal climate change, is the amplitude coefficient, is the period, describing the periodic characteristics of seasonal climate change. The scoring function for the frequency score of extreme weather events is: ,in The frequency of extreme weather events, and is the adjustment coefficient, which describes the nonlinear relationship between the frequency of extreme weather events and their impact.

[0055] In getting the overall score Afterwards, according to The size of the backtracking time window is determined by ,Right now ,in and They are the minimum and maximum values ​​of the time window length, respectively, reflecting the negative correlation between the comprehensive score and the length of the traceback time window, that is, the higher the comprehensive score, the worse the cable condition, and the longer the traceback time window is required; conversely, the lower the comprehensive score, the better the cable condition, and the traceback time window can be appropriately shortened. and , you can set the value range of the backtracking time window length according to actual needs.

[0056] By configuring a comprehensive scoring formula, multiple cable status scoring indicators are quantitatively integrated, and multiple influencing factors such as cable material aging, regional climate change, extreme weather events, etc. are comprehensively considered. It can objectively and accurately evaluate the overall status of the cable and determine the optimal backtracking time window length accordingly, providing an important basis for subsequent cable segmentation management.

[0057] Furthermore, the embodiment of the present application also includes: Perform similarity clustering of segmentation results by using the dynamic segmentation strategy to generate segmentation similarity clustering results; Establishing a benchmark early warning network according to the segmented similarity clustering results, and performing incremental learning based on the similarity clustering results through the benchmark early warning network to build multiple early warning networks mapped with dynamic segmentation strategies; The joint influence of each early warning network is established based on the dynamic segmentation strategy, and the edge joint early warning network allocation is completed through the joint influence.

[0058] In a feasible implementation, accurate mapping and efficient deployment of the warning network and dynamic segmentation strategy are achieved through segment similarity clustering, incremental learning of the baseline warning network and joint impact analysis of the warning network.

[0059] First, the power cables are segmented using a dynamic segmentation strategy, and the segmentation results are clustered similarly. Similarity clustering refers to dividing cable segments with similar characteristics into the same cluster cluster, so that the similarity of the segments within the cluster is the highest, and the difference between the segments between clusters is the largest. For example, clustering algorithms such as k-means and DBSCAN are used, or clustering methods based on deep learning, such as autoencoders and convolutional neural networks, are used. By performing similarity clustering on the segmentation results, several segmentation similarity clusters are obtained. The cable segments in each cluster cluster have similar electrical characteristics, mechanical characteristics, environmental characteristics, etc., and similar warning strategies and warning thresholds can be adopted. The segmentation clustering method based on similarity can greatly reduce the complexity of the design and deployment of the warning network, and improve the accuracy and real-time performance of the warning decision. Then, based on the segmentation similarity clustering results, a benchmark warning network is established. The benchmark warning network is a standardized warning model for a certain type of cable segment, including the key monitoring indicators, warning thresholds, decision rules, etc. of this type of segment. According to expert experience, historical data analysis and other methods, a benchmark early warning network is designed for each segmented cluster. After that, the benchmark early warning network is used to perform incremental learning on the cable segments in each cluster. According to the newly collected monitoring data, the parameters and structure of the early warning network are continuously updated and optimized so that it can adapt to the changing trend of the cable status. Specifically, the cable segment data in each cluster is input into the corresponding benchmark early warning network, and the network weights, thresholds and other parameters are fine-tuned and optimized through machine learning algorithms such as back propagation and gradient descent, so as to obtain personalized early warning networks for different cable segments and obtain multiple early warning networks. Through the incremental learning of the benchmark early warning network, while ensuring the consistency and interpretability of the early warning decision, the individual characteristics and dynamic change information of each cable segment are fully utilized to improve the adaptability and accuracy of the early warning network. At the same time, the incremental learning method based on similar clustering results reduces the training complexity and computational cost of the early warning network, so that the early warning network can be quickly and flexibly deployed to the edge side to achieve distributed real-time early warning.

[0060] After obtaining multiple early warning networks for different cable segments, the joint impact between early warning networks is further analyzed to achieve collaborative optimization and overall deployment of edge-side early warning networks. The joint impact analysis of early warning networks mainly includes two aspects: one is the logical relationship and dependency between different early warning networks, and the other is the degree of influence of different early warning networks on the overall state of the cable. For the logical relationship and dependency between early warning networks, a directed acyclic graph of the early warning network is constructed based on the segmentation logic and decision-making process in the dynamic segmentation strategy. Among them, each node represents an early warning network, and the directed edges between nodes represent the dependency and sequence of early warning decisions. By analyzing the topological structure and critical path of the directed acyclic graph of the early warning network, the execution process of the early warning decision is optimized to improve the speed and accuracy of the early warning response. For the degree of influence of the early warning network on the overall state of the cable, sensitivity analysis, feature importance ranking and other methods are used to quantify the influence weight of each early warning network on the key performance indicators of the cable (such as failure rate, availability, etc.). By calculating the influence weight of the early warning network, the priority and resource allocation strategy of different early warning networks on the edge side are determined to achieve the optimal configuration and dynamic scheduling of the early warning network. Subsequently, the deployment plan of the edge joint warning network was obtained by comprehensively considering the logical dependencies and influence weights of the warning network, clarifying the deployment location, execution order, resource allocation, etc. of each warning network on the edge side, and realizing accurate mapping and collaborative optimization of the warning network and dynamic segmentation strategy. Based on the edge joint warning network, distributed, real-time, and collaborative cable status monitoring and warning can be realized, greatly improving the intelligence level and management efficiency of cable operation and maintenance.

[0061] Through precise mapping and efficient deployment of the early warning network and dynamic segmentation strategy, the similarity and dynamic characteristics of cable segmentation are fully utilized, the accuracy, real-time and scalability of early warning decisions are improved, and important technical support is provided for intelligent cable operation and maintenance.

[0062] Furthermore, the embodiment of the present application also includes: Performing segmented monitoring of the power cables to be monitored through the edge joint early warning network to generate independent segmented early warning results; Activate the joint analysis layer of the edge joint warning network, perform edge joint analysis with the joint analysis layer, and establish a joint segmented warning result; The independent segmented warning results and the combined segmented warning results are verified and identified, and a verification warning result is reported based on the verification and identification result.

[0063] In a preferred embodiment, first, the edge joint early warning network that has been deployed is used to monitor the power cables to be monitored in sections. Through the sensors, smart terminals and other IoT devices deployed on each cable segment, the multi-source heterogeneous monitoring data such as the current, voltage, temperature, vibration, etc. of the cable are collected in real time, and these data are transmitted to the early warning network node on the edge side for local processing and analysis. In each early warning network node, feature extraction, pattern recognition and anomaly detection are performed on the local monitoring data to generate a local early warning result for the cable segment, that is, an independent segment early warning result, which reflects the operating status and potential risks of a single cable segment, including various types of early warning information such as local overload, local overtemperature, and local discharge of the cable. By processing and analyzing a large amount of segmented monitoring data in parallel on the edge side, the real-time and scalability of the early warning analysis are improved, and the computing and storage pressure of the central cloud platform is reduced. At the same time, since each early warning network node only processes local monitoring data, the individual characteristics and environmental factors of the cable segment can be fully utilized to improve the pertinence and accuracy of the early warning analysis.

[0064] In order to further improve the comprehensiveness and consistency of early warning analysis, a joint analysis layer is introduced in the edge joint early warning network. The joint analysis layer realizes the global status assessment and risk prediction of the entire cable by connecting and coordinating various early warning network nodes. Specifically, the joint analysis layer organizes each early warning network node into a hierarchical and orderly analysis process in the form of a directed acyclic graph. Each node represents an early warning network node, and the directed edges between nodes represent data transmission and control dependencies, thereby realizing collaborative analysis and result fusion between early warning network nodes. When an early warning network node generates an independent segmented early warning result, the result is transmitted to the next node in the directed acyclic graph for further analysis. The next node associates the received local early warning result with the local monitoring data to generate higher-level early warning information. The recursive propagation and analysis is repeated until the root node of the directed acyclic graph is reached, and the global early warning result of the entire cable, that is, the joint segmented early warning result, can be obtained. The joint segmented early warning result integrates the local early warning information of each cable segment and takes into account the association influence and transmission effect between segments, so it can more comprehensively and accurately reflect the health status and risk level of the entire cable. Based on the joint segmented warning results, the systemic risks and cascading failures of cables can be discovered in a timely manner, providing a more reliable basis for cable operation and maintenance decisions.

[0065] Subsequently, the independent segment warning results and the joint segment warning results are verified and identified. The purpose of verification and identification is to eliminate uncertain factors such as false alarms and missed alarms in the warning results through cross-validation and comprehensive analysis of multi-source data, and to improve the confidence and interpretability of the warning results. For example, the rationality and consistency of the independent segment warning results and the joint segment warning results are checked and verified by using prior knowledge such as physical models and electrical laws, and warning information that violates physical constraints or logical contradictions is eliminated; the independent segment warning results and the joint segment warning results are compared and analyzed and thresholds are checked by using reference information such as historical operation data and simulation test data to identify abnormal deviations and unreasonable fluctuations in the warning results; the independent segment warning results and the joint segment warning results are semantically understood and analyzed by using artificial intelligence methods such as expert experience and decision rules to identify semantic errors and contextual contradictions in the warning results. By conducting a comprehensive quality diagnosis and credibility assessment of the independent segment warning results and the joint segment warning results from multiple dimensions, verified and confirmed verification warning results are obtained, providing more complete and reliable decision support information for cable management.

[0066] Furthermore, the embodiment of the present application also includes: Obtain the warning identification and evaluation mark of the edge joint warning network; Establishing adaptive feedback of dynamic decision nodes through the early warning identification and evaluation identification; An adaptive compensation of the dynamic decision model is performed based on the constructed adaptation feedback.

[0067] In a feasible implementation, first, obtain the warning recognition evaluation mark of the edge joint warning network. The warning recognition evaluation mark is a set of indicators for evaluating and scoring the warning performance and recognition quality of the warning network, reflecting the key performance of the warning network in practical applications, such as effectiveness, accuracy, and reliability. For example, in the actual operation stage of the warning network, through manual review and post-analysis of the warning results, the accuracy rate, missed alarm rate, false alarm rate and other indicators of the warning network are statistically analyzed, and the warning network is evaluated to obtain the warning recognition evaluation mark. Then, the construction adaptation feedback of the dynamic decision node is established using the warning recognition evaluation mark. Construction adaptation feedback is a mechanism for associating and mapping the evaluation results of the warning network with the construction process of the decision model. By identifying the advantages and disadvantages of the warning network, the structure and parameters of the decision model are dynamically adjusted to achieve the coordinated optimization of the two. Specifically, first, the warning identification and evaluation marks are classified and aggregated according to different cable segments and decision nodes to form an adaptive evaluation vector for the decision process; then, the correlation and sensitivity between the adaptive evaluation vector and the construction parameters of the decision node are analyzed to identify the key decision factors that have the greatest impact on the warning performance; then, according to the value and change trend of the adaptive evaluation vector, the construction parameters of the decision node, such as segment boundaries, feature selection, decision threshold, etc., are dynamically adjusted to adapt to the actual performance of the warning network. By establishing the construction adaptation feedback of the dynamic decision node, the two-way interaction and co-evolution of the decision model and the warning network are realized. On the one hand, the evaluation results of the warning network can provide data support and improvement direction for the optimization of the decision model, helping the decision model to better adapt to the actual cable environment and monitoring needs; on the other hand, the dynamic adjustment of the decision model can also provide guidance and constraints for the training and updating of the warning network, helping the warning network to identify and locate cable risks more accurately and efficiently.

[0068] Subsequently, based on the construction of adaptive feedback, adaptive compensation is performed on the dynamic decision model to improve the adaptability, robustness and interpretability of the decision model. For example, based on the key decision factors identified by the construction of adaptive feedback, corresponding compensation items or correction factors are introduced into the decision model to dynamically adjust the parameter values ​​and threshold judgments in the decision process, so that the decision results are closer to the actual cable status. By performing adaptive compensation on the dynamic decision model, dynamic alignment and continuous optimization of the decision process and the warning results are achieved, effectively reducing the deviation and misjudgment between the decision model and the actual cable status, improving the accuracy and reliability of the decision plan, and continuously improving the generalization and adaptability of the decision model, so that it can better cope with complex and changing cable environments and monitoring needs.

[0069] Furthermore, the embodiment of the present application also includes: Configuring a warning timing cycle based on the recognition result of the edge joint warning network; Performing response effect verification in the warning time sequence cycle, and establishing a progressive response feedback factor based on the response effect verification result; Feedback optimization of early warning identification is performed through the progressive response feedback factor.

[0070] In a preferred embodiment, the warning timing cycle refers to the time interval or frequency of warning analysis and decision response for a specific cable segment within a continuous time range. A reasonable warning timing cycle can ensure the timeliness and accuracy of the warning while avoiding the waste of computing resources and human intervention costs caused by too frequent warning decisions. For example, according to the edge joint warning network, for important trunk cables, an hourly warning cycle is configured to achieve 24-hour uninterrupted monitoring; for secondary branch cables, a daily warning cycle is configured, and a warning analysis is performed once a day; for low-risk auxiliary cables, a weekly or monthly warning cycle is configured, and spot checks and inspections are performed regularly. Then, based on the configured warning timing cycle, the response effect of the warning decision is tracked and verified to obtain response effect verification. Response effect verification refers to evaluating the impact and improvement of the warning decision on the actual cable operation status after each warning timing cycle, so as to judge the effectiveness and timeliness of the warning decision. For example, by comparing the key performance indicators of the cable before and after the early warning decision, such as failure rate, availability, power quality, etc., the effect of the early warning decision on the improvement of the cable health status is evaluated; the execution of the early warning decision and the input of manpower and material resources are analyzed to evaluate the operability and economy of the early warning decision; feedback from cable operation and maintenance personnel and users is collected to evaluate the interpretability and user experience of the early warning decision. Through multi-dimensional response effect verification, quantitative effect evaluation indicators are obtained, such as the percentage of failure rate reduction, the percentage of availability improvement, the percentage of labor cost savings, etc. Based on these indicators, a progressive response feedback factor is established to measure the change trend of the response effect of the early warning decision over time. Afterwards, based on the change trend of the progressive response feedback factor, the length of the early warning time sequence cycle is adaptively adjusted, and the early warning cycle is extended when the response effect is improved, and the early warning cycle is shortened when the response effect is reduced.

[0071] Through feedback optimization of early warning identification, dynamic alignment and continuous improvement of early warning decisions and actual response effects can be achieved, effectively improving the adaptability and robustness of the early warning network, enabling it to quickly respond to changes in cable status, and continuously learn and evolve, to achieve precise operation and maintenance of cable segment management.

[0072] In summary, the intelligent segmentation management method for power cables provided in the embodiments of the present application has the following technical effects: Obtain the pre-execution task of the power cable to be monitored, establish the task cycle identifier with the pre-execution task, and set the first time segment impact coefficient based on the task cycle identifier, which provides basic data for the management strategy, ensures that the strategy can be adjusted according to the specific task requirements, and improves the pertinence of management. Perform the location climate prediction of the power cable to be monitored, establish the second time segment impact coefficient based on the location climate prediction results, consider the impact of environmental factors on the cable status, and enhance the environmental adaptability of the management strategy. Input the first time segment impact coefficient and the second time segment impact coefficient into the dynamic decision model, establish a dynamic decision node, realize the comprehensive consideration of task requirements and environmental factors, and lay the foundation for the formulation of intelligent management strategies. Establish balanced task requirements and balanced climate impacts between nodes with dynamic decision nodes, ensure the balance of tasks and environmental factors between different decision nodes, and improve the coordination of the overall management strategy. Obtain the cable layout information of the power cable to be monitored, which includes the cable layout structure information and the cable layout fixed environment information, taking the physical characteristics of the cable into consideration, and further improving the decision basis. Configure the retrospective time window, use the retrospective time window to perform retrospective monitoring of the power cables to be monitored, establish the retrospective basic parameters of the power cables to be monitored, and provide a reference for current decision-making through historical data analysis to enhance the reliability of decision-making. Perform adaptation analysis of the balanced task requirements and balanced climate impacts based on the retrospective basic parameters, generate the first clustering constraint based on the adaptation analysis results, realize the combination of historical data and current requirements, and improve the accuracy of decision-making. Establish the second clustering constraint based on the cable layout information, establish a dynamic segmentation strategy based on the first clustering constraint, the second clustering constraint, and the dynamic decision node, allocate the edge joint warning network based on the dynamic segmentation strategy, perform segmentation management based on the edge joint warning network, comprehensively consider the above-mentioned factors, and dynamically generate segmentation strategies based on multi-dimensional data to achieve precise and intelligent early warning and management of power cables, and improve the early warning response speed and accuracy.

[0073] Embodiment 2 is based on the same inventive concept as the method for intelligent segmentation management of power cables in the above embodiment. Figure 2 As shown, an embodiment of the present application provides an intelligent segmentation management system for power cables, the system comprising: A first influence coefficient module 11 is used to obtain a pre-execution task of the power cable to be monitored, establish a task cycle identifier with the pre-execution task, and set a first time-series segment influence coefficient based on the task cycle identifier; A second influence coefficient module 12, used to perform location climate prediction of the power cable to be monitored, and establish a second time series segmented influence coefficient based on the location climate prediction result; A decision node generation module 13, used for inputting the first time segment influence coefficient and the second time segment influence coefficient into a dynamic decision model to establish a dynamic decision node; A node balance analysis module 14, used to establish balanced task requirements and balanced climate impacts between nodes using the dynamic decision nodes; The cable information acquisition module 15 is used to acquire the cable layout information of the power cable to be monitored, wherein the cable layout information includes the cable layout structure information and the cable layout fixing environment information; A backtracking parameter establishing module 16 is used to configure a backtracking time window, perform backtracking monitoring of the power cable to be monitored with the backtracking time window, and establish backtracking basic parameters of the power cable to be monitored; A cluster constraint generation module 17, configured for the first cluster constraint module to perform an adaptation analysis of the equilibrium task requirements and the equilibrium climate impact according to the backtracking basic parameters, and to generate a first cluster constraint based on the adaptation analysis result; The segmentation management execution module 18 is used to establish a second clustering constraint according to the cable laying information, establish a dynamic segmentation strategy according to the first clustering constraint, the second clustering constraint, and the dynamic decision node, allocate an edge joint warning network based on the dynamic segmentation strategy, and perform segmentation management based on the edge joint warning network.

[0074] Furthermore, the traceback parameter establishment module 16 includes the following execution steps: Obtain cable material information of the power cable to be monitored, perform a material retrospective impact assessment of the cable based on the cable material information, and establish an aging retrospective score; Conducting a seasonal climate change evaluation on the power cables to be monitored, establishing a climate cycle, obtaining the frequency of occurrence of extreme weather in the region, and constructing a seasonal climate change score and an extreme weather event frequency score respectively according to the climate cycle and the frequency of occurrence; A comprehensive score is established based on the aging backtracking score, seasonal climate change score, and extreme weather event frequency score, and the length of the backtracking time window is determined by the comprehensive score to complete the backtracking time window configuration.

[0075] Furthermore, the traceback parameter establishment module 16 also includes the following execution steps: Configure the comprehensive scoring formula as follows: ; in, For the comprehensive rating, Represent any scoring feature, Features The corresponding weights, Characterization characteristics The scoring function is: ; ; ; Characterizes aging retrospective scoring, and is the linear coefficient, Characterize the cable aging speed characteristics based on cable material information, Characterize seasonal climate change scores, Characterize the amplitude coefficient, Characterization cycle, is the input variable of seasonal climate change, Characterize the frequency score of extreme weather events, and is the adjustment coefficient, is the frequency of occurrence; ; is the length of the lookback time window, and They are the shortest and longest optional values ​​of the time window respectively.

[0076] Furthermore, the segment management execution module 18 includes the following execution steps: Perform similarity clustering of segmentation results by using the dynamic segmentation strategy to generate segmentation similarity clustering results; Establishing a benchmark early warning network according to the segmented similarity clustering results, and performing incremental learning based on the similarity clustering results through the benchmark early warning network to build multiple early warning networks mapped with dynamic segmentation strategies; The joint influence of each early warning network is established based on the dynamic segmentation strategy, and the edge joint early warning network allocation is completed through the joint influence.

[0077] Furthermore, the embodiment of the present application also includes a warning result verification module, which includes the following execution steps: Performing segmented monitoring of the power cables to be monitored through the edge joint early warning network to generate independent segmented early warning results; Activate the joint analysis layer of the edge joint warning network, perform edge joint analysis with the joint analysis layer, and establish a joint segmented warning result; The independent segmented warning results and the combined segmented warning results are verified and identified, and a verification warning result is reported based on the verification and identification result.

[0078] Furthermore, the embodiment of the present application also includes a decision model optimization module, which includes the following execution steps: Obtain the warning identification and evaluation mark of the edge joint warning network; Establishing adaptive feedback of dynamic decision nodes through the early warning identification and evaluation identification; An adaptive compensation of the dynamic decision model is performed based on the constructed adaptation feedback.

[0079] Furthermore, the embodiment of the present application also includes a warning feedback optimization module, which includes the following execution steps: Configuring a warning timing cycle based on the recognition result of the edge joint warning network; Performing response effect verification in the warning time sequence cycle, and establishing a progressive response feedback factor based on the response effect verification result; Feedback optimization of early warning identification is performed through the progressive response feedback factor.

[0080] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0081] Furthermore, the first or second mentioned above may not only represent an order relationship, but may also represent a specific concept, and / or refer to multiple elements that can be selected individually or in full. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these changes and variations.

Claims

1. A method for intelligent segmentation management of power cables, characterized in that: The method comprises: Acquire a pre-execution task of the power cable to be monitored, establish a task cycle identifier with the pre-execution task, and set a first time segmentation influence coefficient based on the task cycle identifier; Performing location climate prediction for the power cable to be monitored, and establishing a second time series segmented influence coefficient based on the location climate prediction result; Inputting the first time segment influence coefficient and the second time segment influence coefficient into a dynamic decision model to establish a dynamic decision node; Establishing balanced task requirements and balanced climate impacts among nodes using the dynamic decision nodes; Acquire cable laying information of the power cable to be monitored, wherein the cable laying information includes cable laying structure information and cable laying fixing environment information; Configure a backtracking time window, perform backtracking monitoring of the power cable to be monitored with the backtracking time window, and establish backtracking basic parameters of the power cable to be monitored; Performing an adaptation analysis of the equilibrium task requirements and the equilibrium climate impact according to the backtracking basic parameters, and generating a first clustering constraint based on the adaptation analysis result; A second clustering constraint is established according to the cable laying information, a dynamic segmentation strategy is established according to the first clustering constraint, the second clustering constraint, and the dynamic decision node, an edge joint warning network is allocated based on the dynamic segmentation strategy, and segmentation management is performed based on the edge joint warning network.

2. A method for intelligent segmentation management of power cables according to claim 1, characterized in that: The configuration backtracking time window further includes: Obtain cable material information of the power cable to be monitored, perform a material retrospective impact assessment of the cable based on the cable material information, and establish an aging retrospective score; Conducting a seasonal climate change evaluation on the power cables to be monitored, establishing a climate cycle, obtaining the frequency of occurrence of extreme weather in the region, and constructing a seasonal climate change score and an extreme weather event frequency score respectively according to the climate cycle and the frequency of occurrence; A comprehensive score is established based on the aging backtracking score, seasonal climate change score, and extreme weather event frequency score. The length of the backtracking time window is determined by the comprehensive score to complete the backtracking time window configuration.

3. A method for intelligent segmentation management of power cables according to claim 2, characterized in that: The method of establishing a comprehensive score based on the aging backtracking score, the seasonal climate change score, and the extreme weather event frequency score, and determining the length of the backtracking time window based on the comprehensive score to complete the backtracking time window configuration, further includes: Configure the comprehensive scoring formula as follows: ； in, For the comprehensive rating, Represent any scoring feature, Features The corresponding weight, Characterization characteristics The scoring function is: ； ； ； Characterizes aging retrospective scoring, and is the linear coefficient, Characterize the cable aging speed characteristics based on cable material information, Characterize seasonal climate change scores, Characterize the amplitude coefficient, Characterization cycle, is the input variable of seasonal climate change, Characterize the frequency score of extreme weather events, and is the adjustment coefficient, is the frequency of occurrence; ； is the length of the lookback time window, and They are the shortest and longest optional values ​​of the time window respectively.

4. The method for intelligent segmentation management of power cables according to claim 1, characterized in that: The allocating edge joint warning network based on the dynamic segmentation strategy and performing segmentation management based on the edge joint warning network also includes: Perform similarity clustering of segmentation results by using the dynamic segmentation strategy to generate segmentation similarity clustering results; Establishing a benchmark early warning network according to the segmented similarity clustering results, and performing incremental learning based on the similarity clustering results through the benchmark early warning network to build multiple early warning networks mapped with dynamic segmentation strategies; The joint influence of each early warning network is established based on the dynamic segmentation strategy, and the edge joint early warning network allocation is completed through the joint influence.

5. A method for intelligent segmentation management of power cables according to claim 4, characterized in that: The method further comprises: Performing segmented monitoring of the power cables to be monitored through the edge joint early warning network to generate independent segmented early warning results; Activate the joint analysis layer of the edge joint warning network, perform edge joint analysis with the joint analysis layer, and establish a joint segmented warning result; The independent segmented warning results and the combined segmented warning results are verified and identified, and a verification warning result is reported based on the verification and identification result.

6. The method for intelligent segmentation management of power cables according to claim 1, characterized in that: The method further comprises: Obtain the warning identification and evaluation mark of the edge joint warning network; Establishing adaptive feedback of dynamic decision nodes through the early warning identification and evaluation identification; An adaptive compensation of the dynamic decision model is performed based on the constructed adaptation feedback.

7. The method for intelligent segmentation management of power cables according to claim 1, characterized in that: The method further comprises: Configuring a warning timing cycle based on the recognition result of the edge joint warning network; Performing response effect verification in the warning time sequence cycle, and establishing a progressive response feedback factor based on the response effect verification result; Feedback optimization of early warning identification is performed through the progressive response feedback factor.

8. An intelligent segmentation management system for power cables, characterized in that: A method for intelligent segmentation management of power cables according to any one of claims 1 to 7, the system comprising: A first influence coefficient module, the first influence coefficient module is used to obtain a pre-execution task of the power cable to be monitored, establish a task cycle identifier with the pre-execution task, and set a first time-series segment influence coefficient based on the task cycle identifier; A second influence coefficient module, the second influence coefficient module is used to perform location climate prediction of the power cable to be monitored, and establish a second time series segmented influence coefficient based on the location climate prediction result; A decision node generation module, wherein the decision node generation module is used to input the first time segment influence coefficient and the second time segment influence coefficient into a dynamic decision model to establish a dynamic decision node; A node balance analysis module, the node balance analysis module is used to establish balanced task requirements and balanced climate impacts between nodes using the dynamic decision nodes; A cable information acquisition module, the cable information acquisition module is used to obtain cable laying information of the power cable to be monitored, the cable laying information includes cable laying structure information and cable laying fixing environment information; A backtracking parameter establishment module, the backtracking parameter establishment module is used to configure a backtracking time window, perform backtracking monitoring of the power cable to be monitored with the backtracking time window, and establish backtracking basic parameters of the power cable to be monitored; A cluster constraint generation module, wherein the cluster constraint generation module is used for the first cluster constraint module to perform an adaptation analysis of the equilibrium task requirements and the equilibrium climate impact according to the backtracking basic parameters, and to generate a first cluster constraint based on the adaptation analysis result; A segmentation management execution module is used to establish a second clustering constraint according to the cable laying information, establish a dynamic segmentation strategy according to the first clustering constraint, the second clustering constraint, and the dynamic decision node, allocate an edge joint warning network based on the dynamic segmentation strategy, and perform segmentation management based on the edge joint warning network.

Images