Artificial intelligence-based distribution cable health state intelligent evaluation method and system

Abstract

The present application relates to the technical field of power equipment detection, and more particularly to a distribution network cable health state intelligent evaluation method and system based on artificial intelligence, comprising the following steps: S1, obtaining multi-dimensional original data and establishing a unified cable database; S2, selecting a plurality of evaluation indexes from the multi-dimensional original data based on a preset cable state evaluation guide to construct a cable health evaluation system; S3, determining the weight of each evaluation index in the cable health evaluation system; S4, calculating the health score of each cable; S5, matching a corresponding differentiated operation and maintenance strategy according to the health score; S6, dynamically updating the health state of the cable and the differentiated operation and maintenance strategy; and S7, generating a priority repair list. The present application integrates multi-source heterogeneous data, constructs a cable health evaluation system, intelligently allocates weights, and realizes accurate quantitative evaluation of the cable health state and automatic generation of the operation and maintenance strategy.

Description

Technical Field

[0001] This invention relates to the field of power equipment testing technology, and in particular to an intelligent assessment method and system for the health status of distribution network cables based on artificial intelligence. Background Technology

[0002] Currently, data such as distribution network cable basic ledgers, test records, and emergency repair information are scattered across offline documents and multiple heterogeneous business systems, forming serious data silos and hindering unified management and effective integration. Cable health status assessment has long relied on manual offline analysis, which is highly subjective, inefficient, and unable to achieve dynamic evaluation and accurate early warning. Simultaneously, maintenance work lacks scientific and differentiated strategy guidance, often relying on high-frequency, one-size-fits-all testing and maintenance methods, which not only consumes significant human and material resources but may also accelerate cable insulation aging due to excessive testing. Existing technologies lack a systematic method that can integrate multi-source data, construct an intelligent evaluation system, and automatically generate maintenance strategies, thus hindering the digital and intelligent transformation of distribution network cable maintenance management.

[0003] Chinese patent CN117520896A discloses a method, apparatus, computer equipment, storage medium, and computer program product for determining the health status of cable accessories. The method includes: acquiring simulation test data and process data of the cable accessories; fusing multiple state characteristic indicators based on the simulation test data and process data to obtain a state characteristic indicator vector of the cable accessories; obtaining a distance value by measuring the distance between the state characteristic indicator vector and a health status characteristic reference vector; and obtaining a health status assessment result of the cable accessories based on the distance value. However, this solution still suffers from inefficiency, high cost, and inaccurate status assessment due to data silos, manual assessment, and static maintenance. Summary of the Invention

[0004] To address this, the present invention provides an intelligent assessment method for the health status of distribution network cables based on artificial intelligence, which overcomes the problems of low efficiency, high cost, and inaccurate status assessment caused by data silos, manual assessment, and static operation and maintenance in the prior art.

[0005] To achieve the above objectives, on the one hand, the present invention provides an intelligent assessment method for the health status of distribution network cables based on artificial intelligence, comprising: Step S1: Obtain multi-dimensional raw data of distribution network cables from multiple heterogeneous data sources, and standardize and manage the multi-dimensional raw data to establish a unified cable database. The multi-dimensional raw data includes cable static parameters, operating environment parameters, and historical status parameters. Step S2: Based on the preset cable condition evaluation guidelines, select multiple evaluation indicators from the multi-dimensional raw data to construct a cable health evaluation system. Step S3: Determine the weight of each evaluation indicator in the cable health evaluation system; Step S4: Calculate the health score for each cable based on the treated data in the cable database and the weight of each evaluation indicator. Step S5: Based on the health score, the health status of each cable is divided into multiple predefined levels, and a corresponding differentiated operation and maintenance strategy is matched for each level. Step S6: Update the cable database based on the acquired real-time or near-real-time cable operation data, and trigger steps S4 and S5 to dynamically update the cable health status and the differentiated operation and maintenance strategy. Step S7: Based on the dynamically updated health status of all cables, an automatic priority maintenance list is generated, wherein the priority maintenance list sorts the cables according to the handling priority in the differentiated operation and maintenance strategy.

[0006] Further, step S2 includes the following sub-steps: Step S21: Analyze and extract key information from the preset cable condition evaluation guidelines to obtain the basic index pool. Step S22: Map and match the basic index pool with the multi-dimensional raw data in the cable database and define them in a structured manner to obtain a structured cable health evaluation system; Step S23: Store the structured cable health evaluation system in the system's model configuration database.

[0007] Further, step S3 includes the following sub-steps: Step S31: Perform large-scale model reasoning analysis on historical cable fault data and cable evaluation standards to obtain the preliminary weight allocation of each evaluation index. Step S32: Perform domain expert verification on the preliminary weight allocation to obtain expert optimization opinions; Step S33: Based on the expert optimization opinions, iteratively adjust and optimize the preliminary weight allocation to determine the final weight set to be used; Step S34: The final weight set is associated and stored in the cable health evaluation system.

[0008] Further, step S4 includes the following sub-steps: Step S41: Obtain the original data of the target cable corresponding to each evaluation indicator in the cable health evaluation system from the cable database; Step S42: Perform data cleaning and standardization preprocessing on the raw data of each evaluation indicator to obtain standardized indicator data values; Step S43: The standardized index data value of each evaluation index is weighted and calculated with its corresponding weight in the final weight set to obtain the initial value of the health score of the target cable. Step S44: Standardize and calibrate the initial health score value to output the final health score; Step S45: Associate the unique identifier of the target cable with its calculated final health score and store it in the evaluation results database.

[0009] Further, step S5 includes the following sub-steps: Step S51: Compare the final health score with a predefined status level threshold range to determine the health status level of the target cable. Step S52: Based on the determined health status level of the target cable, match and call the corresponding differentiated operation and maintenance strategy template from the strategy library; Step S53: Instantiate the matched differentiated operation and maintenance strategy template into a specific executable operation and maintenance strategy for the target cable. Step S54: Associate the health status level of the target cable with specific executable operation and maintenance strategies, and output the results to the operation and maintenance management terminal and work order system.

[0010] Further, step S6 includes the following sub-steps: Step S61: Continuously or periodically acquire real-time or near-real-time operating data of the target cable through the data interface; Step S62: After standardizing and processing the real-time or near-real-time operating data, update the corresponding data record in the cable database; Step S63: In response to the update of the cable database, the execution of steps S4 and S5 is automatically triggered; Step S63: Output the updated health status and new differentiated operation and maintenance strategies and push them to the relevant systems and user interfaces.

[0011] Further, step S7 includes the following sub-steps: Step S71: Collect the health status levels of all cables after dynamic updates and the corresponding differentiated operation and maintenance strategies; Step S72: Assign a base priority score to each health status level and fine-tune it based on the specific instructions in the strategy to obtain the comprehensive priority score for each cable. Step S73: Sort all cables to be repaired in descending order according to the comprehensive priority score to generate an initial priority repair list; Step S74: Perform resource conflict verification and optimization on the initial priority maintenance list, and output the final executable priority maintenance list. Step S75: Publish the final priority maintenance list to the operation and maintenance work order system and push it to the mobile terminals of relevant operation and maintenance personnel.

[0012] Furthermore, in step S31, the large model reasoning analysis specifically involves: inputting the historical cable fault data and cable evaluation standards into the power brightening large model, which then performs an index correlation analysis based on knowledge of the power field, and outputs a preliminary weight allocation scheme that includes the relative importance of each evaluation index.

[0013] Furthermore, the cable health evaluation system constructed in step S2 includes evaluation indicators in three dimensions: basic cable condition dimension, including cable importance level, number of intermediate joints, and years of operation; laying method and operating environment dimension, including laying depth, ambient humidity, and load level; and emergency repair and test record dimension, including test data change trend, fault frequency, and maximum fault current.

[0014] On the other hand, the present invention also provides an intelligent assessment system for the health status of distribution network cables based on artificial intelligence, comprising: The data integration module is used to obtain multi-dimensional raw data of distribution network cables from multiple heterogeneous data sources, and to standardize and manage the multi-dimensional raw data to establish a unified cable database. The evaluation system construction module is used to select multiple evaluation indicators from the multi-dimensional raw data based on the preset cable condition evaluation guidelines to construct a cable health evaluation system and determine the weight of each evaluation indicator in the cable health evaluation system. The operation and maintenance strategy generation module is used to calculate the health score of each cable based on the data after treatment in the cable database and the weight of each evaluation indicator. It is also used to divide the health status of each cable into multiple predefined levels based on the health score and match the corresponding differentiated operation and maintenance strategy for each level. The dynamic update module is used to update the cable database based on the acquired real-time or near-real-time operating data of the cable, and trigger the re-triggering of the operation and maintenance strategy generation module to dynamically update the health status of the cable and the differentiated operation and maintenance strategy. The maintenance list generation module is used to automatically generate a priority maintenance list based on the health status of all cables after dynamic updates. The priority maintenance list sorts the cables according to the disposal priority in the differentiated operation and maintenance strategy.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: The method, through step S1, acquires multi-dimensional raw data of distribution network cables from multiple heterogeneous data sources and performs standardized management to establish a unified cable database. This solves the "data silo" problem caused by the dispersion of cable basic ledgers, test records, and emergency repair information across multiple systems and offline locations, providing a complete, standardized, and high-quality data foundation for subsequent intelligent assessment. Furthermore, through step S2, the method selects multiple evaluation indicators from the multi-dimensional raw data based on preset cable condition evaluation guidelines to construct a cable health evaluation system. This achieves the unification and structuring of evaluation dimensions, indicators, and standards, overcoming the problems of inconsistent evaluation standards and strong subjectivity caused by reliance on human experience. In step S3, the method uses a dual-drive mode of "large model reasoning + expert experience" to determine the weight of each evaluation indicator in the cable health evaluation system, ensuring that the weight allocation combines data objectivity and practical rationality, providing a key basis for scientifically calculating health scores. Finally, through step S4, the method uses the managed cable data and... By assigning weights to various indicators and calculating a health score for each cable, the method achieves accurate and automated quantitative evaluation of cable health status, replacing the inefficient manual offline analysis mode and significantly improving assessment efficiency and accuracy. The method further divides cable health status into multiple predefined levels based on the health score in step S5, and matches differentiated operation and maintenance strategies to each level, realizing automatic mapping from status assessment to operation and maintenance decisions. This supports the formulation of scientific and differentiated operation and maintenance strategies, effectively reducing ineffective testing and operation and maintenance costs. The method also dynamically updates the cable database based on real-time or near-real-time operational data in step S6, triggering a recalculation of health scores and strategies. This achieves real-time perception of cable health status and adaptive adjustment of assessment strategies, solving the pain point of traditional methods being unable to achieve dynamic evaluation. Finally, the method automatically generates a maintenance list sorted by priority based on the dynamically updated health status of all cables in step S7, realizing intelligent scheduling and resource optimization of maintenance tasks, effectively reducing overdue backlogs and improving the targeting and efficiency of operation and maintenance work. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the intelligent assessment method for the health status of distribution network cables based on artificial intelligence in this embodiment. Figure 2 This is a schematic diagram of the intelligent health status assessment system for distribution network cables based on artificial intelligence in this embodiment. Detailed Implementation

[0017] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0018] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0019] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0020] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0021] Please see Figure 1 The diagram shown is a flowchart of the intelligent assessment method for the health status of distribution network cables based on artificial intelligence in this embodiment. The system includes: Step S1: Obtain multi-dimensional raw data of distribution network cables from multiple heterogeneous data sources, and standardize and manage the multi-dimensional raw data to establish a unified cable database. The multi-dimensional raw data includes cable static parameters, operating environment parameters, and historical status parameters. Step S2: Based on the preset cable condition evaluation guidelines, select multiple evaluation indicators from the multi-dimensional raw data to construct a cable health evaluation system. Step S3: Determine the weight of each evaluation indicator in the cable health evaluation system; Step S4: Calculate the health score for each cable based on the treated data in the cable database and the weight of each evaluation indicator. Step S5: Based on the health score, the health status of each cable is divided into multiple predefined levels, and a corresponding differentiated operation and maintenance strategy is matched for each level. Step S6: Update the cable database based on the acquired real-time or near-real-time cable operation data, and trigger steps S4 and S5 to dynamically update the cable health status and the differentiated operation and maintenance strategy. Step S7: Based on the dynamically updated health status of all cables, an automatic priority maintenance list is generated, wherein the priority maintenance list sorts the cables according to the handling priority in the differentiated operation and maintenance strategy.

[0022] Specifically, this method is applied to the intelligent assessment of the health status of distribution network cables. By integrating multi-source heterogeneous data, a cable health evaluation system is constructed and intelligent weight allocation is achieved, enabling accurate quantitative assessment of cable health status and automatic generation of operation and maintenance strategies. In step S1, the method acquires multi-dimensional raw data of distribution network cables from multiple heterogeneous data sources and performs standardized governance to establish a unified cable database. This solves the "data silo" problem caused by the dispersion of cable basic ledgers, test records, and emergency repair information across multiple offline systems, providing a complete, standardized, and high-quality data foundation for subsequent intelligent assessment. In step S2, based on preset cable status evaluation guidelines, the method selects multiple evaluation indicators from the multi-dimensional raw data to construct a cable health evaluation system. This achieves the unification and structuring of evaluation dimensions, indicators, and standards, overcoming the problems of inconsistent evaluation standards and strong subjectivity caused by reliance on human experience. In step S3, the method uses a dual-drive mode of "large model reasoning + expert experience" to determine the weight of each evaluation indicator in the cable health evaluation system, ensuring that the weight allocation is both data objective and practically reasonable, providing a basis for scientific calculation of health status. The method provides crucial information and, through step S4, calculates a health score for each cable based on the treated cable data and the weights of various indicators. This achieves accurate and automated quantitative evaluation of cable health status, replacing the inefficient manual offline analysis mode and significantly improving assessment efficiency and accuracy. Step S5 further divides cable health status into multiple predefined levels based on the health score and matches differentiated operation and maintenance strategies to each level, achieving automatic mapping from status assessment to operation and maintenance decisions. This supports the formulation of scientific and differentiated operation and maintenance strategies, effectively reducing ineffective testing and operation and maintenance costs. Step S6 dynamically updates the cable database based on real-time or near-real-time operating data and triggers the recalculation of health scores and strategies, achieving real-time perception of cable health status and adaptive adjustment of assessment strategies. This addresses the pain point of traditional methods being unable to achieve dynamic evaluation. Step S7 automatically generates a maintenance list sorted by priority based on the dynamically updated health status of all cables, achieving intelligent scheduling and resource optimization of maintenance tasks, effectively reducing overdue backlogs and improving the targeting and efficiency of operation and maintenance work.

[0023] Specifically, the process of acquiring multi-dimensional raw data of distribution network cables from multiple heterogeneous data sources and standardizing and managing this multi-dimensional raw data to establish a unified cable database involves: Extracting multi-dimensional raw data in parallel from multiple heterogeneous data sources, such as the power grid resource business platform, production management system (PMS), online monitoring system, and offline data entry spreadsheets, through configured data interfaces and connectors. This data includes: cable static parameters (e.g., asset code, commissioning year, cable type, length, number of intermediate joints); operating environment parameters (e.g., laying method, ambient humidity, soil corrosion level, historical load data); and historical status parameters (e.g., annual preventative test records, live-line testing data, fault repair records). Data cleaning rules are applied to the extracted multi-dimensional raw data, including but not limited to: filling missing values, correcting obvious errors (e.g., test values ​​exceeding physical limits), removing duplicate records, and unifying all timestamps to a standard format to obtain regularized data. The regularized data is then subjected to structured transformation and semantic mapping to obtain standardized data. The structured transformation includes: parsing unstructured or semi-structured data (such as the text description of an inspection report) in the regularized data into structured fields using natural language processing (NLP) technology; The semantic mapping includes: unifying homonymous or heteronymous fields from different systems in the regularized data according to a preset cable data standard dictionary. For example, the “cable type” in system A and the “cable model” in system B are uniformly mapped to the “cable_model” field, and its value is uniformly set to a standard code (such as “YJLV22-8.7 / 15-3*300”). Standardized data is loaded into a unified cable database through ETL (Extract, Transform, Load) or ELT (Extract, Load, Transform) processes. It is then stored and linked according to a predefined data model (e.g., divided into "Basic Ledger Table," "Test Record Table," "Operating Environment Table," etc.), completing the physical centralization and fusion of data. During the data loading process, the data source, transformation process, and loading time are recorded to establish data lineage. Simultaneously, data quality verification scripts are run regularly to check the completeness and accuracy of key fields in the database, ensuring that the unified cable database remains in a high-quality and traceable state, providing a reliable data foundation for subsequent analysis.

[0024] Specifically, step S2 includes the following sub-steps: Step S21: Analyze and extract key information from the preset cable condition evaluation guidelines to obtain the basic index pool. Step S22: Map and match the basic index pool with the multi-dimensional raw data in the cable database and define them in a structured manner to obtain a structured cable health evaluation system; Step S23: Store the structured cable health evaluation system in the system's model configuration database.

[0025] Specifically, step S21 involves parsing and extracting key information from the preset cable condition evaluation guidelines to obtain a basic indicator pool. This includes calling standard documents such as the "Cable Line Condition Evaluation Guidelines" stored in the knowledge base. Natural language processing technology is used to parse the guideline text, identify and extract all recommended or mandatory condition quantity suggestions, forming a basic indicator pool containing preliminary evaluation items. Step S22, which maps and matches the multi-dimensional raw data in the basic indicator pool and the cable database to obtain a structured cable health evaluation system, specifically involves: Step S221, Data Mapping: Associating and matching each evaluation item in the basic indicator pool with specific data fields in the cable database to ensure that each indicator has data support. For example, mapping the "main insulation insulation resistance" indicator to specific fields in the "insulation resistance test record" table; Step S222, Quantification and Structuring: Formulating clear quantification rules and type definitions (such as quantitative or qualitative indicators) for successfully mapped indicators, and organizing all indicators according to three dimensions: "basic cable conditions," "laying method and operating environment," and "emergency repair and test records," to construct a hierarchical cable health evaluation system. Step S23, storing the structured cable health evaluation system into the system's model configuration database, specifically involves: persistently storing the constructed cable health evaluation system (including indicator names, types, quantification rules, and dimensional information) into the model configuration database, completing the construction of the evaluation system, and preparing for subsequent weight allocation and health score calculation.

[0026] Specifically, step S3 includes the following sub-steps: Step S31: Perform large-scale model reasoning analysis on historical cable fault data and cable evaluation standards to obtain the preliminary weight allocation of each evaluation index. Step S32: Perform domain expert verification on the preliminary weight allocation to obtain expert optimization opinions; Step S33: Based on the expert optimization opinions, iteratively adjust and optimize the preliminary weight allocation to determine the final weight set to be used; Step S34: The final weight set is associated and stored in the cable health evaluation system.

[0027] Specifically, step S31, which involves performing large-scale model reasoning analysis on historical cable fault data and cable evaluation standards to obtain the preliminary weight allocation for each evaluation indicator, involves inputting historical cable fault data (including fault type, fault location, fault cause, etc.) stored in the database and textualized cable evaluation standards (such as the "Cable Line Condition Evaluation Guidelines") into a pre-trained power health large-scale model. Based on its embedded power domain knowledge and learning of the correlation patterns of massive amounts of data, the large-scale model performs reasoning calculations and outputs a weight set containing the preliminary weights corresponding to each evaluation indicator in the cable health evaluation system. Step S32, which involves verifying the preliminary weight allocation with domain experts to obtain expert optimization opinions, specifically involves presenting the preliminary weight allocation results obtained in step S1 to one or more experts in the fields of power operation and maintenance and cable technology through the system's expert interaction interface. Based on their long-term practical experience and understanding of the specific operational needs of the local power grid, the experts judge the rationality of the preliminary weights and submit adjustment suggestions through this interface, thus forming expert optimization opinions. Step S33, which iteratively adjusts and optimizes the preliminary weight allocation based on the expert optimization opinions to determine the final weight set, specifically involves the system's weight calculation module receiving the expert optimization opinions. This module, based on the opinions (e.g., whether a certain indicator's weight should be increased or decreased), performs corresponding mathematical adjustments and rebalancing of the preliminary weight allocation, generating a new, optimized weight set. This process can be iterated multiple times until the weight allocation result simultaneously satisfies both data patterns and expert experience. Finally, the weight set determined at this point is solidified as the final weight set and stored in the system. Step S34, which associates and stores the final weight set with the cable health evaluation system, specifically involves binding each weight value in the final weight set with the corresponding evaluation index in the cable health evaluation system, and persistently storing them together in the system's model configuration database, thereby completing a complete evaluation model that combines data objectivity and practical rationality and can be immediately used for health score calculation.

[0028] Specifically, step S4 includes the following sub-steps: Step S41: Obtain the original data of the target cable corresponding to each evaluation indicator in the cable health evaluation system from the cable database; Step S42: Perform data cleaning and standardization preprocessing on the raw data of each evaluation indicator to obtain standardized indicator data values; Step S43: The standardized index data value of each evaluation index is weighted and calculated with its corresponding weight in the final weight set to obtain the initial value of the health score of the target cable. Step S44: Standardize and calibrate the initial health score value to output the final health score; Step S45: Associate the unique identifier of the target cable with its calculated final health score and store it in the evaluation results database.

[0029] Specifically, step S41, obtaining the original data of the target cable corresponding to each evaluation indicator in the cable health evaluation system from the cable database, specifically involves: based on the indicator items defined by the cable health evaluation system stored in the model configuration database, the system retrieves all corresponding original data of the target cable from the cable database through the database query interface, such as obtaining specific values ​​or status records of the cable such as "years of operation", "number of intermediate joints", "most recent insulation resistance value", and "laying environment type". Step S42 involves data cleaning and standardization preprocessing of the raw data for each evaluation indicator to obtain standardized indicator data values. Specifically, this involves preprocessing the raw data obtained in step S1 to ensure data quality and dimensional consistency. For numerical indicators (such as insulation resistance value), outlier removal and missing value filling are performed; For non-numerical indicators (such as laying environment), they are converted into uniform numerical scores according to predefined quantification rules (e.g., "pipeline" = 5, "direct burial" = 3, "tunnel" = 1). Finally, all indicator data values ​​are mapped to a unified scoring range (e.g., 0-1 or 0-100) through normalization (e.g., Min-Max standardization) to form standardized indicator data values.

[0030] In step S43, the standardized indicator data value of each evaluation indicator is weighted and calculated according to its corresponding weight in the final weight set to obtain the initial health score of the target cable. Specifically, the system's scoring calculation engine calls the following formula to calculate: Initial health score = Σ (standardized indicator data value of the i-th indicator × weight of the i-th indicator); Here, i iterates from 1 to the number of all evaluation metrics. This calculation integrates state information from all dimensions into a single, quantifiable score. Step S44, which normalizes and calibrates the initial health score to output the final health score, specifically involves mapping the initial health score calculated in step S43 to a preset, easily understood final score range (e.g., 0 to 100 points) through linear transformation or other calibration functions. This step ensures that the scoring standards for all cables are consistent, and that the scoring results are intuitively reflected in subsequent status classifications. Step S45, which associates the unique identifier of the target cable with its calculated final health score and stores it in the evaluation result database, specifically involves writing the cable ID (or asset code) along with the calculated final health score and calculation timestamp into the evaluation result database. This storage action provides a direct data foundation for subsequent health status classification, strategy generation, and historical trend analysis.

[0031] Specifically, step S5 includes the following sub-steps: Step S51: Compare the final health score with a predefined status level threshold range to determine the health status level of the target cable. Step S52: Based on the determined health status level of the target cable, match and call the corresponding differentiated operation and maintenance strategy template from the strategy library; Step S53: Instantiate the matched differentiated operation and maintenance strategy template into a specific executable operation and maintenance strategy for the target cable. Step S54: Associate the health status level of the target cable with specific executable operation and maintenance strategies, and output the results to the operation and maintenance management terminal and work order system.

[0032] Specifically, step S51, comparing the final health score with a predefined status level threshold range to determine the health status level of the target cable, involves the system calling a status level threshold range rule stored in the policy library. This rule explicitly defines the score range corresponding to different status categories, for example: Good: Score ≥ 90 points; Stability: 75 points ≤ Score < 90 points; Attention: 60 points ≤ Score < 75 points; Address the vacancy promptly: Score < 60 points; The system automatically compares the final health score of the target cable with the above rules and classifies it into the corresponding health status level. Step S52, which involves matching and calling the corresponding differentiated operation and maintenance strategy template from the strategy library based on the determined health status level of the target cable, specifically involves: each health status level has one or more standardized differentiated operation and maintenance strategy templates pre-stored in the strategy library. The system automatically matches and calls the corresponding template based on the level determined in step S1. The strategy template must contain at least the following core processing instructions: Test cycle instructions: specify the time interval for the next test or inspection; Priority handling instructions: Define their sorting position in the maintenance work order; Specific handling measures are recommended: specific maintenance actions are suggested.

[0033] Step S53, which instantiates the matched differentiated operation and maintenance strategy template into a specific executable operation and maintenance strategy for the target cable, specifically involves the system combining the abstract strategy template with the specific information of the target cable (such as asset code and geographical location) to generate a specific executable operation and maintenance strategy. For example: For cables in "good" condition, the strategy is instantiated as: "Extend the test period to 48 months; include in the routine inspection list, priority: low"; For cables with a status of "Concern", the strategy is instantiated as follows: "Shorten the test cycle to 12 months; include in the key concern list, priority: medium; recommend partial discharge detection"; For cables with a status of "resolve the defect as soon as possible", the strategy is instantiated as: "Generate maintenance work order immediately; test cycle is 0 (execute immediately); handling priority: highest".

[0034] Step S54, which associates the health status level of the target cable with specific executable operation and maintenance strategies and outputs them to the operation and maintenance management terminal and work order system, specifically involves binding the final determined health status level and specific executable operation and maintenance strategies with the cable ID, pushing them to the operation and maintenance personnel's visualization terminal for display through the system interface, and automatically creating or updating the corresponding inspection or maintenance tasks in the work order system. This step completes the closed loop from intelligent analysis to production instructions.

[0035] Specifically, step S6 includes the following sub-steps: Step S61: Continuously or periodically acquire real-time or near-real-time operating data of the target cable through the data interface; Step S62: After standardizing and processing the real-time or near-real-time operating data, update the corresponding data record in the cable database; Step S63: In response to the update of the cable database, the execution of steps S4 and S5 is automatically triggered; Step S64: Output the updated health status and new differentiated operation and maintenance strategies and push them to the relevant systems and user interfaces.

[0036] Specifically, step S61, which involves continuously or periodically acquiring real-time or near-real-time operating data of the target cable through a data interface, involves the system acquiring real-time or near-real-time operating data of the target cable through a preset data acquisition interface (such as an API interface with a SCADA system or online monitoring device) in a streaming or timed polling manner. This data includes, but is not limited to: real-time load current, cable joint temperature, partial discharge monitoring signals, and ambient temperature and humidity. Step S62, which involves standardizing and processing the real-time or near-real-time operational data before updating the corresponding data record in the cable database, specifically involves: For the new data acquired in step S61, the same standardized processing procedure (including data cleaning, format conversion, outlier handling, etc.) as the initial data integration step is applied, and the data is written into the cable database, overwriting or appending it to the historical records corresponding to the target cable. For example, the operating parameter table is updated with new load data, and new partial discharge detection results are added to the test record table. This ensures that the database always reflects the latest status of the cable. Step S63, which automatically triggers the execution of steps S4 and S5 in response to an update of the cable database, specifically involves: Triggering Mechanism: After the cable database completes a data update, it generates a "data update event" or sets a status flag. The system's task scheduler listens for this event or flag and automatically triggers a re-evaluation task for the target cable accordingly. Execution step S4: The triggered task first calls step S4, which will recalculate the health score of the target cable based on the latest data in the updated cable database, as well as the stored cable health evaluation system and final weight set, to obtain a score that reflects its latest status. Step S5: Next, the task calls step S5, which is based on the new health score calculated in S4, redetermines the health status level of the cable, and matches and generates a new differentiated operation and maintenance strategy accordingly. Step S64, which outputs and pushes the updated health status and new differentiated operation and maintenance strategies to relevant systems and user interfaces, specifically involves the system pushing the new health status level and new differentiated operation and maintenance strategies obtained after reassessment to the operation and maintenance work order system, the visual monitoring dashboard, and the mobile terminals of relevant operation and maintenance personnel via message queues or API interfaces. If the status deteriorates (e.g., from "stable" to "concerned"), the system will generate alarm information and increase the push priority.

[0037] Specifically, step S7 includes the following sub-steps: Step S71: Collect the health status levels of all cables after dynamic updates and the corresponding differentiated operation and maintenance strategies; Step S72: Assign a base priority score to each health status level and fine-tune it based on the specific instructions in the strategy to obtain the comprehensive priority score for each cable. Step S73: Sort all cables to be repaired in descending order according to the comprehensive priority score to generate an initial priority repair list; Step S74: Perform resource conflict verification and optimization on the initial priority maintenance list, and output the final executable priority maintenance list. Step S75: Publish the final priority maintenance list to the operation and maintenance work order system and push it to the mobile terminals of relevant operation and maintenance personnel.

[0038] Specifically, step S71, which gathers the health status levels of all cables after dynamic updates and the corresponding differentiated operation and maintenance strategies, involves the system reading the health status levels of all cables in the current assessment period (such as "address the defect as soon as possible" or "pay attention") and the differentiated operation and maintenance strategies generated in step S5 (which include the "handling priority" instruction) from the assessment result database in batches. Step S72 assigns a base priority score to each health status level and fine-tunes it based on specific instructions in the strategy to obtain the comprehensive priority score for each cable. Specifically, the base score assignment involves the system calling a predefined status-base score mapping table, for example: "Fill the vacancy as soon as possible" -> Base score 100; "Follow" -> Base score 70; "Stable" -> Base score 30; "Good" -> Base score 0; Score fine-tuning: Points are added or subtracted from the base score based on specific instructions within the differentiated operation and maintenance strategy. For example, if the strategy includes the tag "power supply related to political issues," an additional 20 points are awarded to the overall priority score; if it includes "high load importance," an additional 15 points are awarded. This is used to calculate the overall priority score for each cable. Step S73, which involves sorting all cables to be inspected in descending order based on the comprehensive priority score to generate an initial priority inspection list, specifically involves the system's sorting algorithm filtering out all cables with a comprehensive priority score greater than 0 (or another threshold), and arranging them in descending order of score to form an initial priority inspection list. The cable with the highest score is placed at the top of the list, indicating that it requires the highest priority for processing. Step S74, which involves verifying and optimizing the initial priority maintenance list to output the final executable priority maintenance list, specifically involves the system comparing the initial priority maintenance list with existing constraints such as work orders, personnel, and materials. Conflict checking: Check if there are resource conflicts among multiple tasks in the list (e.g., multiple tasks are assigned to the same operations team at the same time). Optimization and adjustment: While keeping the core priority order as much as possible, make minor adjustments to the list to form a final priority maintenance list that can be executed in terms of time, space and resources. Step S75, which involves publishing the final priority maintenance list to the operation and maintenance work order system and pushing it to the mobile terminals of relevant operation and maintenance personnel, specifically involves publishing the final priority maintenance list to the work order system through the system integration interface, automatically generating specific maintenance work orders, and simultaneously pushing the summary of the list and the highest priority task to the mobile applications of relevant operation and maintenance teams and responsible persons in real time to guide them in carrying out on-site work.

[0039] Specifically, in step S31, the large model reasoning analysis involves inputting the historical cable fault data and cable evaluation standards into the power brightening large model, which then performs an index correlation analysis based on knowledge of the power field and outputs a preliminary weight allocation scheme that includes the relative importance of each evaluation index.

[0040] Specifically, the cable health evaluation system constructed in step S2 includes evaluation indicators in three dimensions: basic cable condition dimension, including cable importance level, number of intermediate joints, and years of operation; laying method and operating environment dimension, including laying depth, ambient humidity, and load level; and emergency repair and test record dimension, including test data change trend, fault frequency, and maximum fault current.

[0041] Please see Figure 2 As shown, this is a schematic diagram of the intelligent health status assessment system for distribution network cables based on artificial intelligence in this embodiment. The system includes: The data integration module is used to obtain multi-dimensional raw data of distribution network cables from multiple heterogeneous data sources, and to standardize and manage the multi-dimensional raw data to establish a unified cable database. The evaluation system construction module is used to select multiple evaluation indicators from the multi-dimensional raw data based on the preset cable condition evaluation guidelines to construct a cable health evaluation system and determine the weight of each evaluation indicator in the cable health evaluation system. The evaluation system construction module is connected to the data integration module. The operation and maintenance strategy generation module is used to calculate the health score of each cable based on the treated data in the cable database and the weight of each evaluation indicator. It is also used to divide the health status of each cable into multiple predefined levels based on the health score and match the corresponding differentiated operation and maintenance strategy for each level. The operation and maintenance strategy generation module is connected to the evaluation system construction module. The dynamic update module is used to update the cable database based on the acquired real-time or near-real-time operating data of the cable, and to trigger the operation and maintenance strategy generation module to dynamically update the health status of the cable and the differentiated operation and maintenance strategy. The dynamic update module is connected to the operation and maintenance strategy generation module. The maintenance list generation module is used to automatically generate a priority maintenance list based on the health status of all cables after dynamic updates. The priority maintenance list sorts the cables according to the disposal priority in the differentiated operation and maintenance strategy. The maintenance list generation module is connected to the dynamic update module.

[0042] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. An intelligent assessment method for the health status of distribution network cables based on artificial intelligence, characterized in that, include: Step S1: Obtain multi-dimensional raw data of distribution network cables from multiple heterogeneous data sources, and standardize and manage the multi-dimensional raw data to establish a unified cable database. The multi-dimensional raw data includes cable static parameters, operating environment parameters, and historical status parameters. Step S2: Based on the preset cable condition evaluation guidelines, select multiple evaluation indicators from the multi-dimensional raw data to construct a cable health evaluation system. Step S3: Determine the weight of each evaluation indicator in the cable health evaluation system; Step S4: Calculate the health score for each cable based on the treated data in the cable database and the weight of each evaluation indicator. Step S5: Based on the health score, the health status of each cable is divided into multiple predefined levels, and a corresponding differentiated operation and maintenance strategy is matched for each level. Step S6: Update the cable database based on the acquired real-time or near-real-time cable operation data, and trigger steps S4 and S5 to dynamically update the cable health status and the differentiated operation and maintenance strategy. Step S7: Based on the dynamically updated health status of all cables, an automatic priority maintenance list is generated, wherein the priority maintenance list sorts the cables according to the handling priority in the differentiated operation and maintenance strategy.

2. The intelligent assessment method for the health status of distribution network cables based on artificial intelligence according to claim 1, characterized in that, Step S2 includes the following sub-steps: Step S21: Analyze and extract key information from the preset cable condition evaluation guidelines to obtain the basic index pool. Step S22: Map and match the basic index pool with the multi-dimensional raw data in the cable database and define them in a structured manner to obtain a structured cable health evaluation system; Step S23: Store the structured cable health evaluation system in the system's model configuration database.

3. The intelligent assessment method for the health status of distribution network cables based on artificial intelligence according to claim 1, characterized in that, Step S3 includes the following sub-steps: Step S31: Perform large-scale model reasoning analysis on historical cable fault data and cable evaluation standards to obtain the preliminary weight allocation of each evaluation index. Step S32: Perform domain expert verification on the preliminary weight allocation to obtain expert optimization opinions; Step S33: Based on the expert optimization opinions, iteratively adjust and optimize the preliminary weight allocation to determine the final weight set to be used; Step S34: The final weight set is associated and stored in the cable health evaluation system.

4. The intelligent assessment method for the health status of distribution network cables based on artificial intelligence according to claim 1, characterized in that, Step S4 includes the following sub-steps: Step S41: Obtain the original data of the target cable corresponding to each evaluation indicator in the cable health evaluation system from the cable database; Step S42: Perform data cleaning and standardization preprocessing on the raw data of each evaluation indicator to obtain standardized indicator data values; Step S43: The standardized index data value of each evaluation index is weighted and calculated with its corresponding weight in the final weight set to obtain the initial value of the health score of the target cable. Step S44: Standardize and calibrate the initial health score value to output the final health score; Step S45: Associate the unique identifier of the target cable with its calculated final health score and store it in the evaluation results database.

5. The intelligent assessment method for the health status of distribution network cables based on artificial intelligence according to claim 1, characterized in that, Step S5 includes the following sub-steps: Step S51: Compare the final health score with a predefined status level threshold range to determine the health status level of the target cable. Step S52: Based on the determined health status level of the target cable, match and call the corresponding differentiated operation and maintenance strategy template from the strategy library; Step S53: Instantiate the matched differentiated operation and maintenance strategy template into a specific executable operation and maintenance strategy for the target cable. Step S54: Associate the health status level of the target cable with specific executable operation and maintenance strategies, and output the results to the operation and maintenance management terminal and work order system.

6. The intelligent assessment method for the health status of distribution network cables based on artificial intelligence according to claim 1, characterized in that, Step S6 includes the following sub-steps: Step S61: Continuously or periodically acquire real-time or near-real-time operating data of the target cable through the data interface; Step S62: After standardizing and processing the real-time or near-real-time operating data, update the corresponding data record in the cable database; Step S63: In response to the update of the cable database, the execution of steps S4 and S5 is automatically triggered; Step S63: Output the updated health status and new differentiated operation and maintenance strategies and push them to the relevant systems and user interfaces.

7. The intelligent assessment method for the health status of distribution network cables based on artificial intelligence according to claim 1, characterized in that, Step S7 includes the following sub-steps: Step S71: Collect the health status levels of all cables after dynamic updates and the corresponding differentiated operation and maintenance strategies; Step S72: Assign a base priority score to each health status level and fine-tune it based on the specific instructions in the strategy to obtain the comprehensive priority score for each cable. Step S73: Sort all cables to be repaired in descending order according to the comprehensive priority score to generate an initial priority repair list; Step S74: Perform resource conflict verification and optimization on the initial priority maintenance list, and output the final executable priority maintenance list. Step S75: Publish the final priority maintenance list to the operation and maintenance work order system and push it to the mobile terminals of relevant operation and maintenance personnel.

8. The intelligent assessment method for the health status of distribution network cables based on artificial intelligence according to claim 3, characterized in that, In step S31, the large model reasoning analysis specifically involves: inputting the historical cable fault data and cable evaluation standards into the power brightening large model, which then performs an index correlation analysis based on knowledge of the power field, and outputs a preliminary weight allocation scheme that includes the relative importance of each evaluation index.

9. The intelligent assessment method for the health status of distribution network cables based on artificial intelligence according to claim 2, characterized in that, The cable health evaluation system constructed in step S2 includes evaluation indicators in three dimensions: basic cable condition dimension, including cable importance level, number of intermediate joints, and years of service; laying method and operating environment dimension, including laying depth, ambient humidity, and load level; and emergency repair and test record dimension, including test data change trend, fault frequency, and maximum fault current.

10. A system applied to the intelligent assessment method for the health status of distribution network cables based on artificial intelligence as described in any one of claims 1 to 9, characterized in that, include: The data integration module is used to obtain multi-dimensional raw data of distribution network cables from multiple heterogeneous data sources, and to standardize and manage the multi-dimensional raw data to establish a unified cable database. The evaluation system construction module is used to select multiple evaluation indicators from the multi-dimensional raw data based on the preset cable condition evaluation guidelines to construct a cable health evaluation system and determine the weight of each evaluation indicator in the cable health evaluation system. The operation and maintenance strategy generation module is used to calculate the health score of each cable based on the data after treatment in the cable database and the weight of each evaluation indicator. It is also used to divide the health status of each cable into multiple predefined levels based on the health score and match the corresponding differentiated operation and maintenance strategy for each level. The dynamic update module is used to update the cable database based on the acquired real-time or near-real-time operating data of the cable, and trigger the re-triggering of the operation and maintenance strategy generation module to dynamically update the health status of the cable and the differentiated operation and maintenance strategy. The maintenance list generation module is used to automatically generate a priority maintenance list based on the health status of all cables after dynamic updates. The priority maintenance list sorts the cables according to the disposal priority in the differentiated operation and maintenance strategy.

Images