Method for supervising the service of electrical power technology

Abstract

The application provides a power technology service guarantee supervision method, including: determining the matching degree evaluation standard of the operation tool specification and the protection measure level according to the real-time operation data set, and determining the operation rationality evaluation criterion by integrating the compliance deviation reason for the unplanned repair judgment result; performing power risk grading analysis on the cascading failure prediction set to judge the existence of the high-risk level, and extracting transformer abnormal sign data to obtain the risk level classification result; integrating the optimized path back to the real-time operation data set according to the difference degree of the optimized repair path and the initial repair operation step process, and obtaining the updated operation guidance set by combining the transformer abnormal sign data; generating a power operation visual report according to the updated operation guidance set, verifying the applicable conditions of not directly starting the alarm when deviating from the service standard process, and obtaining the final compliance determination conclusion.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for supervising and ensuring power technology services. Background Technology

[0002] With the expansion of power systems and the increasing complexity of equipment, the number of potential safety hazards or equipment problems discovered during field operations is growing. These problems are often outside the planned service schedule, placing enormous decision-making pressure on service personnel. How to ensure service standardization while allowing field personnel to flexibly respond to emergencies has become a critical issue that urgently needs to be addressed in the power technology service field. Traditional process specifications are usually designed based on pre-set scenarios, making it difficult to cover the diverse and temporary problems that arise during field operations. For example, service personnel may discover minor cracks or potential electrical hazards in equipment during inspections, but these problems are not within the scope of planned maintenance, and whether or not they are addressed immediately may affect the subsequent stability of power supply. Experience-based decision-making methods, lacking a unified assessment basis, often lead to inconsistent results from different personnel handling the same problem, and may even cause greater risks due to over-intervention or neglect of the problem. In field operations, service personnel need to adjust safety measures, select appropriate tool types, and determine specific repair steps based on the actual situation. The lack of systematic collection and analysis of this information makes it impossible to accurately determine whether the service personnel's decisions deviate from the standard or may cause serious consequences. For example, service personnel may temporarily use non-standard tools for repairs, which may be reasonable in certain circumstances, but may also increase operational risks due to incompatible tools. Furthermore, the implementation of dynamic assessment faces another technical challenge: how to quickly determine the scope of impact of repair operations in complex and ever-changing field environments. The equipment involved in field operations is often highly interconnected with other systems, and a single repair can trigger a chain reaction. For example, when service personnel are handling a minor fault in a transformer, they may need to temporarily disconnect some circuits, which could affect the normal operation of other equipment. If the potential impact of such operations cannot be quickly assessed, the repair actions may actually exacerbate system risks. Therefore, establishing a dynamic rationality judgment mechanism based on information such as the type of tools used by service personnel, adjustments to safety measures, and repair procedures during power technical service field operations, while simultaneously quickly assessing the scope of impact of repair operations, to ensure that service personnel do not deviate from regulations when flexibly responding to emergencies, has become a key issue for the precise supervision and support of power technical services. Summary of the Invention

[0003] This invention provides a method for supervising the guarantee of power technology services, mainly including:

[0004] Collect information on the type of work tools, the level of protective equipment, and the maintenance process; integrate fault handling information on cable damage or transformer malfunctions; and generate a real-time work dataset.

[0005] Based on the real-time operation dataset, the matching degree between tool models and protection equipment levels is identified, compliance deviation factors of unplanned repairs are incorporated, and operational rationality assessment criteria are determined.

[0006] Based on the operational rationality assessment criteria, the maintenance operation process is extracted, the scope of equipment impact is identified, the impact of fault handling plans is assessed, and a cascading fault prediction set is generated.

[0007] The risk classification of the predicted cascading faults is performed, cable damage information is integrated to distinguish deviations from regulations, transformer anomaly data is extracted, and risk level classification results are generated.

[0008] Extract high-risk subsets from the risk level classification results, assess the probability distribution of the repair process, adjust the priority of more than one device or cross-regional operation, and generate an optimized repair path;

[0009] Based on the differences between the optimized repair path and the initial maintenance operation process, the optimized repair path is integrated into the real-time operation dataset to generate an updated operation guidance set.

[0010] Based on the updated set of work instructions, the exemption conditions that deviate from the specifications are identified, and a compliance determination result is generated.

[0011] Furthermore, the system collects information on the type of work tools, the level of protective equipment, and the maintenance procedures, integrates fault handling information for cable damage or transformer malfunctions, and generates a real-time work dataset, including:

[0012] The system collects data on the frequency of tool operation, matches it with the standard procedure database, extracts the insulation level of protective gloves and the safety helmet level, records the installation sequence of grounding wires and the operation sequence of circuit breakers, and generates an initial work record. Based on the initial work record, it detects the wear depth of cable sheaths, extracts the ambient humidity and temperature gradient, or reads the transformer vibration amplitude, compares the fault handling plan with the protection level, and judges the compliance of the work specifications. Based on the compliance of the work specifications, it generates a real-time work dataset containing full-process monitoring.

[0013] Furthermore, based on the real-time operation dataset, the matching degree between tool models and protection equipment levels is identified, compliance deviation factors of unplanned repairs are incorporated, and operational rationality assessment criteria are determined, including:

[0014] The insulation class, withstand voltage, and applicable voltage range of the acquired tools are compared with a standard tool library to obtain the insulation resistance and level of the protective equipment, generating a preliminary matching score. If the preliminary matching score is lower than the standard, the operation sequence is extracted to identify deviations from the procedures in terms of time points and behaviors. Combined with environmental constraints and task urgency, a rationality judgment value is generated. Based on the rationality judgment value, a comprehensive evaluation result is calculated to generate an operational rationality evaluation criterion that includes judgment thresholds and handling rules.

[0015] Furthermore, based on the real-time job dataset, the matching degree between tool specifications and protection levels is identified, compliance deviations from unplanned repairs are incorporated, and operational rationality assessment criteria are determined, including:

[0016] Obtain the tool number and compare it with the standard tool list to calculate the specification deviation of the non-standard tool; based on the specification deviation, query the protection configuration database, calculate the protection parameter margin, and determine the unavailability of the standard tool by combining the tool inventory status and the urgency of the task; based on the unavailability of the standard tool, extract the workspace dimensions and the distance between equipment, calculate the environmental adaptability index, and generate an operational rationality assessment criterion for the use of non-standard tools by combining the adaptation risk and safety risk.

[0017] Furthermore, based on the aforementioned operational rationality assessment criteria, the maintenance operation process is extracted, the equipment impact range is identified, the impact of fault handling plans is assessed, and a cascading fault prediction set is generated, including:

[0018] Obtain the switching of disconnecting switches, the opening and closing of circuit breakers, and the location of grounding wires; identify the equipment number list of transformers, busbars, and feeders; based on the equipment number list, query the power grid connection relationship, trace the power source point and load distribution, and determine the power flow redistribution path; based on the power flow redistribution path, retrieve the historical fault tripping probability, combine it with real-time load data, calculate the node overload probability, and generate a cascading fault prediction set containing faulty equipment and power outage range.

[0019] Furthermore, the predicted cascading failure set is risk-classified, cable damage information is integrated to distinguish deviations from regulations, transformer anomaly data is extracted, and risk level classification results are generated, including:

[0020] The number of users, power outage time, and equipment damage extent are extracted from the cascading failure prediction set to calculate a comprehensive risk index; based on the comprehensive risk index, cable insulation resistance and crack depth are retrieved to identify transformer oil temperature exceeding limits and gas concentration, and to distinguish deviation factors from regulations.

[0021] Based on the comprehensive risk index and abnormal signs, risk levels are divided, and a risk level classification result containing level identifiers and risk-causing factors is generated.

[0022] Furthermore, a high-risk subset is extracted from the risk level classification results, the probability distribution of the repair process is assessed, the priority of one or more devices or cross-regional operations is adjusted, and an optimized repair path is generated, including:

[0023] High-risk scenarios are selected from the risk level classification results, and device numbers, fault types, repair durations, and location coordinates are extracted to establish a repair task list. Based on the repair task list, historical repair records are queried to construct the time probability distribution of each repair stage and calculate the task completion time interval. Based on the time probability distribution, the number of users affected by the device is sorted, the cross-regional task sequence is adjusted, the task order and personnel allocation are optimized, and an optimized repair path containing task sequences and resource configurations is generated.

[0024] Furthermore, a high-risk subset is extracted from the risk level classification results, the probability distribution of the repair process is assessed, the priority of one or more devices or cross-regional operations is adjusted, and an optimized repair path is generated, including:

[0025] Extract equipment fault codes and repair times from the high-risk subset, query the human resources database, and establish a task list; based on the task list, identify parallel task groups, calculate the distance between work locations, and record personnel conflict status; based on the personnel conflict status, extract equipment hierarchical positions, determine the power outage and restoration sequence, calculate the number of users during power outages, and generate an optimized repair path that includes task timing and personnel allocation.

[0026] Furthermore, based on the differences between the optimized repair path and the initial maintenance operation process, the optimized repair path is integrated into the real-time operation dataset to generate an updated operation guidance set, including:

[0027] Compare the optimized repair path with the initial process, record the task sequence and time differences, and match the emergency response steps; based on the emergency response steps, replace the original process operations, merge the optimized path and abnormal handling requirements, and generate an updated set of work instructions.

[0028] Furthermore, based on the updated set of work instructions, the exemption conditions for deviations from the specifications are identified, and a compliance determination result is generated, including:

[0029] Task nodes and safety measures are extracted from the updated work instruction set to generate a flowchart; based on the flowchart, the operation sequence is compared with the safety procedures, the power outage operation and protective equipment are checked, and the deviation type is recorded; based on the deviation type, exemption conditions are queried, and the compliance degree and deviation rationality are considered to generate a compliance judgment result.

[0030] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0031] This invention discloses a method for supervising and ensuring power technical service support. It proposes an integrated solution for complex business scenarios in power field maintenance, including matching tool specifications and protective measures, ensuring compliance of unplanned repairs, and controlling the risk of cascading failures. The invention extracts tool numbers, protection levels, and work process information from real-time operation datasets, compares them with a standard list, identifies the reasons for compliance deviations in the use of non-standard tools, and establishes reasonableness assessment criteria. For high-risk scenarios, it analyzes cable damage and abnormal transformer symptoms, generates a cascading failure prediction set, and performs risk classification. By optimizing equipment outage sequence and repair paths, and integrating repair time and personnel scheduling, it generates a set of work instructions that balances safety and power supply reliability, ultimately forming a visual report and verifying compliance. This invention significantly improves the safety, efficiency, and compliance of power maintenance, reduces the risk of cascading failures, optimizes cross-regional operation coordination and resource allocation, and ensures power supply reliability. Attached Figure Description

[0032] Figure 1 A flowchart illustrating the method for supervising and ensuring power technology services according to the present invention.

[0033] Figure 2 This is a schematic diagram of the method for supervising and ensuring power technology services according to the present invention. Detailed Implementation

[0034] To further understand the content of this invention, a detailed description of the invention is provided in conjunction with the accompanying drawings and embodiments. The specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0035] like Figure 1-2 The method for supervising power technology service guarantees in this embodiment may specifically include:

[0036] S101. Collect the specifications of the tools used by service personnel, the level of protective equipment, and the maintenance operation process during the operation. When cable damage or transformer abnormality is found, integrate the preset fault handling plan information to obtain a real-time operation dataset.

[0037] S1011. By using vibration and temperature sensors deployed at the power site, the operation frequency data stream of the tools used by the service personnel is collected. According to the tool model code, the standard operating procedure database is matched to extract the corresponding protective glove insulation level parameters and safety helmet protection level information. At the same time, the timing data of the grounding wire installation sequence and circuit breaker operation steps in the maintenance operation are recorded to obtain the initial operation record containing tool usage characteristics and protection configuration information.

[0038] S1012. Regarding the initial operation record, when the cable sheath wear depth is detected to exceed a preset threshold, the humidity value and temperature gradient data of the environment around the cable are extracted from the sensor network. If the transformer oil temperature monitoring value deviates from the standard range, the vibration amplitude sequence is read from the transformer monitoring terminal. By comparing the switching sequence of the disconnecting switch in the fault handling plan library with the current protection level parameters, the compliance of the operation specifications is determined.

[0039] S1013. Based on the compliance with the work specifications, integrate the work tool model parameters, protective equipment wearing status monitoring values, cable damage location coordinate data, and transformer abnormality type identifiers. Through timestamp alignment processing, arrange and combine the discrete data streams collected by each sensor in chronological order to determine a real-time work dataset containing monitoring information for the entire work process.

[0040] Specifically, in one implementation, a power field sensor network achieves comprehensive monitoring of service personnel's operations through various types of sensors deployed in the work area. Vibration sensors are installed on the gripping parts of tools such as insulated operating rods and voltage detectors. By detecting changes in vibration frequency and amplitude, the usage status of the tools is identified. When the vibration frequency remains within the 50-200Hz range for a preset duration, the tool is determined to be in working condition. Temperature sensors are arranged on the tool surface and inside the worker's protective equipment to collect real-time temperature change data, which is used to assess the safety of the working environment and the effectiveness of the protective equipment.

[0041] Specifically, tool model codes are matched in real-time with a standard operating procedure (SOP) database using RFID tags. This SOP database pre-stores the protection requirements for various tools, including the withstand voltage rating of insulating gloves, the protection category of safety helmets, and the resistance range of insulating shoes. When service personnel retrieve tools, the reader automatically identifies the tool code and retrieves the corresponding protection configuration requirements, comparing them with the parameters of the protective equipment actually worn by the personnel. For detecting cable damage, the sensor network combines ultrasonic detectors and image recognition. The ultrasonic detector emits high-frequency sound waves and receives reflected signals, analyzing the distortion of the reflected waveform to determine the integrity of the cable sheath. When the detected damage depth exceeds a preset threshold, an image acquisition device is activated to obtain visual characteristics of the damaged area, including crack length, damaged area, and degree of insulation exposure. Simultaneously, environmental sensors collect humidity and temperature gradients around the damage point; this data is used to assess the impact of the damage on cable operational safety. Transformer anomaly monitoring is achieved through oil temperature sensors and vibration monitoring terminals. When the oil temperature deviates from the standard range and exceeds the set threshold, the vibration monitoring terminal begins to record the vibration amplitude sequence of the transformer body and identifies abnormal vibration modes through spectrum analysis.

[0042] In one possible implementation, the determination of compliance with work procedures is based on a matching mechanism of a fault handling plan database. This database stores standard operating procedures for various fault scenarios, including the switching sequence of disconnect switches, the installation location of grounding wires, and the execution sequence of voltage testing steps. The actual collected operation sequence is compared with the standard procedures in the database, and the time and sequence deviations for each operation step are calculated. When the deviation values ​​are within the allowable range, the work is deemed to comply with the specifications.

[0043] It should be noted that the timestamp alignment process employs a unified clock synchronization mechanism to ensure the temporal consistency of data from all sensors. Each sensor node maintains synchronization with the master clock via a network time protocol, ensuring that the timestamp accuracy of the acquired data reaches the millisecond level. During the alignment process, the system sorts the discrete data streams according to their timestamps and associates and marks multi-source data from the same time point, forming a data matrix in the time dimension.

[0044] Preferably, the construction of the real-time job dataset also includes a data integrity verification step. The integrated data undergoes a continuity check to identify missing data points and outliers. Missing data is then supplemented using interpolation algorithms to ensure the integrity and reliability of the job dataset, providing an accurate data foundation for subsequent dynamic evaluation.

[0045] S102. Based on the real-time operation dataset, determine the matching evaluation criteria for the specifications of operation tools and the level of protective measures. At the same time, incorporate the reasons for compliance deviations into the judgment results of unplanned repairs to determine the criteria for evaluating the rationality of operations.

[0046] The specifications of the currently used tools, including insulation class, withstand voltage, and applicable voltage range, are extracted from the real-time operation dataset. These parameters are compared item by item with the standard power operation tool library to identify the difference between the tool's rated parameters and the on-site operation requirements. Simultaneously, the insulation resistance value and protection level parameters of the protective equipment are obtained. A preliminary matching score between the tool's rated parameters and the on-site operation requirements is obtained by comparing the difference with a preset threshold. If the preliminary matching score is lower than the standard requirements, the operation sequence records of service personnel are extracted from the on-site monitoring data to identify the time points and specific behaviors that deviate from the standard operating procedures. If unplanned repair operations are detected, on-site environmental constraints, including the availability of standard tools, workspace conditions, and task urgency, are retrieved. Appropriate scores are assigned based on the severity of the constraints to determine the reasonableness of the deviation. Based on the aforementioned reasonableness judgment value, an evaluation index set is constructed, comprising four dimensions: tool adaptability, adequate protection, timing deviation, and environmental constraints. Weighting coefficients are assigned according to the degree of impact of each dimension on operational safety. The comprehensive evaluation result is calculated by multiplying the scores of each dimension by their respective weighting coefficients and then summing the results. When the result exceeds a preset safety threshold, the operational behavior is deemed reasonable under the current conditions. Based on the comparison between the comprehensive evaluation result and the safety threshold, deviation cause types, risk level classifications, and compensatory safety measure requirements are integrated to form a set of handling rules for different deviation scenarios. This establishes an operational reasonableness evaluation criterion that includes judgment thresholds, evaluation dimensions, and handling requirements.

[0047] Specifically, in one implementation, tool specification parameters are extracted through the collaborative operation of embedded sensors and a data acquisition terminal. Each power work tool is equipped with an RFID tag, which stores the tool's unique code, rated parameters, and usage restrictions. When service personnel retrieve the tool, the data acquisition terminal automatically reads the tool information, including insulation class, withstand voltage value, applicable voltage range, service life, and inspection validity period. The power work standard tool library pre-establishes a tool configuration requirement matrix for different work scenarios, defining the minimum insulation class, protective equipment configuration standards, and additional requirements for special environments required for each voltage level.

[0048] Specifically, the preliminary matching score is calculated using the difference comparison method. The actual voltage level at the current work site is extracted and compared with the tool's rated applicable voltage to calculate the safety margin between the two. If the actual voltage is 10kV and the tool's rated applicable voltage is 35kV, then the safety margin is 25kV. Simultaneously, the insulation resistance value of the protective equipment is obtained through online testing and compared with the minimum insulation resistance required by the standard. When both the safety margin and insulation resistance meet the requirements, the matching score is full; if either is lower than the standard but higher than the minimum safety threshold, the corresponding score is calculated using linear interpolation.

[0049] For example, deviation behavior identification is achieved through multi-source data fusion at the work site. Surveillance cameras use image recognition algorithms to analyze the actions of service personnel, identifying the execution times of key operational nodes, such as the start time of voltage testing, the completion time of grounding wire installation, and the operation time of disconnecting switches. These time points are compared with the sequence specified in the standard operating procedures. When the execution time of an operation deviates from the standard sequence by more than the allowable range, or the order of operations is reversed, the system determines it as a deviation behavior. Identification of unplanned repair operations is achieved by comparing the actual operation content with the content recorded on the work order. When an operation not listed on the work order is detected, a deviation cause analysis process is triggered.

[0050] In one possible implementation, determining the rationality judgment value requires comprehensive consideration of multiple environmental constraints. The availability status of standard tools is queried in real time through the tool inventory management system. When a required standard tool is under maintenance, inspection, or occupied by other operations, it is recorded as unavailable. Operating space conditions are obtained through on-site measurements, including parameters such as operating space height, width, and distance between equipment. When the space size is less than the minimum operating space requirement for a standard tool, it is judged as space-constrained. Task urgency is calculated based on the scope and duration of the fault's impact. A higher urgency score is assigned if the number of affected users exceeds a preset threshold or the expected power outage time exceeds the specified time limit. The rationality judgment value is obtained by weighted summation of the severity scores of each constraint factor. The weighting coefficients are determined based on historical accident statistics and expert experience. When the weight for tool unavailability is 0.4, the weight for space constraint is 0.3, and the weight for task urgency is 0.3, if the severity scores of the three factors are 80, 60, and 90 respectively, the rationality judgment value is calculated as 80×0.4 + 60×0.3 + 90×0.3. The construction of the multi-dimensional evaluation index set follows the principles of comprehensiveness and independence. The tool adaptability dimension assesses the degree of matching between tool parameters and operational requirements, obtaining a score by calculating the relative deviation between the tool's rated parameters and the actual required parameters. The protection adequacy dimension examines the completeness and effectiveness of protective equipment, including the correct wearing of protective equipment, the suitability of the protection level, and the coverage of the protection range. The timing deviation dimension quantifies the difference between the actual operation timing sequence and the standard timing sequence, calculating the similarity between the two timing sequences using a dynamic time warping algorithm. The environmental constraints dimension reflects the degree of restriction imposed on the operation by the on-site environment, comprehensively considering the impact of environmental factors such as temperature, humidity, wind speed, and visibility on operational safety. The weighting coefficients are determined using the analytic hierarchy process (AHP). By constructing a judgment matrix, each evaluation dimension is compared pairwise to determine its relative importance. In high-voltage operation scenarios, tool adaptability and protection adequacy have relatively large weights, set at 0.35 and 0.30 respectively; while in emergency repair operation scenarios, the weights of timing deviation and environmental constraints are appropriately increased to reflect the actual need for rapid power restoration. The comprehensive evaluation result is obtained by summing the product of the scores of each dimension and their corresponding weights. When the comprehensive evaluation result exceeds the preset safety threshold, it indicates that although there is deviation behavior, the operation is reasonable and necessary under the current conditions.

[0051] In one embodiment, the formation process of operational rationality assessment criteria includes three stages: rule extraction, classification and organization, and standardization. The rule extraction stage identifies typical deviation scenarios from a large number of historical operational cases and analyzes the correlation between the causes of deviation, remedial measures, and the final results. The classification and organization stage categorizes deviation scenarios according to cause type, such as tool-related causes, environmental causes, and emergency repair causes, with each category corresponding to different judgment thresholds and assessment focuses. The standardization stage integrates the scattered remedial rules into a unified assessment criterion system, clarifying the judgment threshold ranges for various deviation scenarios, the safety baselines that must be met, and recommended compensation measures.

[0052] For example, when the deviation is due to the lack of standard tools, the evaluation criteria stipulate that the insulation level of the alternative tools must not be lower than 80% of that of the standard tools, and require additional safety protection measures, such as increasing the safety distance, shortening the operation time, and increasing the number of monitoring personnel.

[0053] The tool numbers currently used by service personnel are extracted from the real-time operation dataset and compared with the standard operation tool list to identify the specification differences and applicable scope of non-standard tools. The matching relationship between the current protection level and the safety requirements of non-standard tools is obtained. The necessity and urgency of tool replacement in unplanned repair operations are analyzed. It is determined whether temporary replacement caused by on-site environmental limitations and the lack of standard tools falls within the scope of compliance deviation, and operational rationality assessment criteria are formed.

[0054] The RFID tag number and QR code information of the handheld tools used by service personnel are extracted from the real-time operation dataset. The tool configuration requirements for the corresponding voltage level and operation type are retrieved from the standard operation tool list. By comparing the tool insulation level, withstand voltage value, and leakage current index, the relative percentage deviation of each parameter from the standard value is calculated, and the maximum deviation is taken as the specification deviation value. Based on the specification deviation value, the protective measure configuration database is queried to extract the insulation resistance test value and dielectric strength parameter of the current protective equipment. The margin for protective parameters exceeding the minimum requirements of non-standard tools is calculated. If the margin is greater than a preset safety threshold, the on-site tool inventory status and task urgency rating are obtained from the operation record to determine if standard tools are unavailable. For cases where standard tools are unavailable, the operating space dimensions and equipment distance values ​​are extracted from the on-site environmental monitoring data. When the space dimension is less than the minimum space required for standard tool operation, the spatial constraint degree value is recorded. Simultaneously, the environmental adaptability index is obtained by multiplying the applicability score of non-standard tools in confined spaces by the spatial constraint degree value. The environmental adaptability index is compared with the preset compliance deviation judgment threshold. When the index value is within the allowable range, the adaptation risk score caused by the deviation of tool electrical performance and the operation safety risk score caused by the difference in mechanical strength are calculated respectively. Based on the combination of the risk level intervals of the two scores, the operational rationality assessment criteria for the temporary use of non-standard tools are determined.

[0055] Specifically, in one implementation, non-standard tools are identified through a multi-coding mechanism for precise location. Each tool is equipped with an RFID tag that stores the manufacturer's code, batch number, specifications, and certificate of conformity number, while the QR code contains the tool's usage history and maintenance information. When service personnel retrieve a tool from the tool library, a handheld terminal automatically scans and parses this encoded information, comparing it in real time with a pre-established list of standard operating tools. This list is categorized and stored according to different voltage levels and job types.

[0056] Specifically, the specification deviation value is calculated using a multi-parameter weighted method. The insulation class value, withstand voltage test value, and measured leakage current value of the tool are extracted and their differences are calculated compared to the corresponding parameters of the standard tool. For insulation class deviation, a corresponding score is deducted for each decrease in class; for withstand voltage deviation, a percentage is calculated based on the ratio of the actual value to the standard value; and for leakage current deviation, the degree of deviation is determined by the multiple by which it exceeds the standard limit. The three deviation values ​​are multiplied by their respective weighting coefficients and then summed. These weighting coefficients are dynamically adjusted according to the risk level of different operating scenarios to obtain the comprehensive specification deviation value.

[0057] For example, the matching assessment of protective measures and non-standard tools involves multi-level safety margin analysis. When the insulation level of non-standard tools is lower than the standard requirements, the system automatically retrieves real-time monitoring data from the protective equipment database, including insulation resistance test values ​​for insulating gloves, withstand voltage test results for safety helmets, and leakage current detection values ​​for insulating shoes. These protective parameters need to compensate for the inadequacy of tool performance. By calculating the margin value of protective parameters exceeding the minimum safety requirements of non-standard tools, it is determined whether the work safety conditions are met.

[0058] For example, when using a Class II insulation tester instead of a Class III tester, the insulation resistance of the protective gloves needs to be at least 1.5 times the standard value to create sufficient safety redundancy. Simultaneously, tool inventory management records will be checked to identify whether standard tools are unavailable due to maintenance, inspection, or being occupied. Only when standard tools are truly unavailable will the suitability assessment process for non-standard tools be initiated.

[0059] It is important to note that a quantitative assessment of workspace constraints is a crucial step in determining the appropriateness of using non-standard tools. On-site environmental monitoring uses laser rangefinders to acquire three-dimensional spatial data around the work site, including horizontal distances between equipment, vertical heights, and the range of operable angles. Standard tools typically have long insulating rods or large operating radii, limiting their use in confined spaces.

[0060] In one possible implementation, the formation of the environmental adaptability index comprehensively considers two dimensions: the degree of spatial constraint and the applicability of the tool. The degree of spatial constraint is calculated by the ratio of the actual available space to the minimum operating space of the standard tool. A ratio less than 1 indicates that the space is limited, and the smaller the ratio, the more severe the constraint. The applicability score of the non-standard tool is based on a comprehensive evaluation of its physical dimensions, operational flexibility, and historical usage records in confined spaces. The product of the two forms the environmental adaptability index, which reflects the feasibility of using non-standard tools under specific spatial conditions. When the spacing between switchgear in a substation is only 0.8 meters, and the standard insulated operating rod is 1.2 meters long and cannot be used normally, the shorter non-standard operating rod, although with a slightly lower insulation level, may have a higher environmental adaptability index. In this case, the use of non-standard tools is reasonable.

[0061] Preferably, compliance deviation determination adopts a tiered threshold mechanism. Three threshold levels are preset, each corresponding to a different risk tolerance. The first threshold is applicable to general maintenance work, allowing for a relatively large deviation range; the second threshold is applicable to live-line work, with a lower tolerance for deviation; and the third threshold is applicable to high-voltage work, allowing only extremely small deviations. Environmental adaptability indicators are compared with the corresponding threshold levels to determine whether the deviation is within acceptable limits. Furthermore, a dual risk assessment mechanism comprehensively evaluates the risks of using non-standard tools from both tool compatibility and operational safety perspectives. The compatibility risk score is calculated by analyzing the probability of failure that may result from deviations in the tool's electrical parameters, including a comprehensive score for insulation failure risk, insufficient withstand voltage risk, and excessive leakage current risk. The operational safety risk score considers factors such as differences in mechanical strength, reduced ease of operation, and personnel familiarity, assessing the impact of each factor on operational safety using fault tree analysis. The two risk scores are mapped to a preset risk level matrix, which is divided into four areas: low risk, medium risk, high risk, and unacceptable.

[0062] For example, when the adaptation risk is at a medium level and the safety risk is low, tool use falls within the low-to-medium risk zone of the matrix. In this case, with the addition of compensatory measures such as enhanced monitoring and shortened operation time, the use of non-standard tools is deemed reasonable. Based on the risk level combination, the system automatically matches the corresponding operational requirements and safety assurance measures, forming a structured assessment criterion. This provides clear judgment criteria and operational guidance for on-site personnel to use non-standard tools under special circumstances.

[0063] S103. When the operational rationality assessment criteria exceed the set range, extract the maintenance operation execution process from the real-time operation dataset, identify the impact range of power equipment, and assess the impact range of the fault handling plan without directly activating the alarm to obtain a cascading fault prediction set.

[0064] When the value of the operational rationality assessment criterion exceeds a preset threshold, the operation sequence record of the maintenance operation is extracted from the real-time operation dataset, including the switching time of the disconnecting switch, the opening and closing sequence of the circuit breaker, and the installation location of the grounding wire. The transformers, busbars, and feeder circuits involved in the current operation are identified through the power grid wiring diagram, and a list of power equipment numbers directly affected by the maintenance is determined. Based on the list of power equipment numbers, the power grid connection relationship database is queried to obtain the upstream power supply point and downstream load distribution of each equipment. If a certain equipment is detected to be under maintenance, all user nodes and backup power supply paths within its power supply range are traced to determine the power flow redistribution path caused by the maintenance operation. For key nodes on the power flow redistribution path, the equipment tripping probability under the same load level is retrieved from the historical fault database. Combined with the current real-time load data, the probability value of overload occurring at each node after the power flow transfer is calculated through the state transition matrix to obtain the equipment fault propagation probability distribution. The fault propagation probability distribution is compared with a preset risk threshold. Without triggering an automatic alarm, the affected equipment sequence, fault propagation path, and expected power outage range data are integrated to form a cascading fault prediction set containing potentially faulty equipment, the number of affected users, and the recovery time.

[0065] Specifically, in one implementation, the determination of whether the operation rationality assessment criteria exceed the threshold adopts a multi-level triggering mechanism. When the assessment value exceeds the first-level threshold, the impact range analysis process is initiated, but a silent state is maintained without triggering audible and visual alarms. The operation sequence records in the real-time job dataset are sorted by timestamp, and each record includes the operation object number, operation type code, execution time, and operator identifier.

[0066] Specifically, power equipment identification is achieved by parsing the topology of the power grid wiring diagram. The power grid wiring diagram stores equipment connection relationships in the form of nodes and branches. Nodes represent convergence points such as buses and transformers, while branches represent connecting elements such as lines and switches. Based on the disconnector switch number in the operation record, the corresponding branch is located in the wiring diagram. A breadth-first search algorithm is used to traverse all nodes and branches directly connected to that branch to identify the set of equipment directly affected by the operation, including upstream power supply-side equipment and downstream load-side equipment.

[0067] For example, determining power flow redistribution paths involves complex power flow transfer mechanisms in the power grid. When a line or transformer is taken out of service for maintenance, the power that originally flowed through that device needs to be transferred via alternative paths. First, the rated capacity, current load rate, and impedance parameters of the devices are extracted from the power grid interconnection database. Then, all possible backup power supply paths are identified. Each backup path consists of multiple devices connected in series or parallel, and the total impedance and available capacity of each path are calculated. Power flow allocation follows the principle of minimum impedance, with power preferentially flowing to the path with lower impedance. When the main transformer is out of service, the 10kV bus load it supplies is transferred to an adjacent substation via a tie switch, changing the power flow path from a single path to a detour via the tie line. This power flow transfer leads to an increase in the load on the tie line and adjacent transformers, requiring an assessment of the operational risks of these devices under the new load levels.

[0068] It should be noted that the historical fault database stores a large number of equipment fault records under different operating conditions. Each record includes the load level, ambient temperature, runtime, and fault type at the time of the fault. Based on the current real-time load data, fault cases under similar operating conditions are searched in the historical database, and the frequency of occurrence of various fault types is statistically analyzed.

[0069] In one possible implementation, the state transition matrix is ​​constructed based on the Markov property of device states. The operating states of power equipment can be categorized into normal, abnormal, and fault states, and the transition probability between states is closely related to the equipment load rate. Each element of the matrix represents the probability value of a device transitioning from one state to another. When the equipment load rate exceeds 80% of its rated capacity, the probability of transitioning from a normal state to an abnormal state increases significantly. The transition probability of each state is calculated by statistically analyzing the state transition frequency under different load levels in historical data. For devices experiencing a sudden load surge after a power flow shift, the state transition probability is recalculated using the updated load rate to identify critical devices with higher fault risks. The fault propagation probability distribution is obtained by multiplying the fault probability of each device by its position weight in the power grid; the position weight reflects the degree of impact of device faults on the system. The risk threshold setting considers the safety margin of the power grid and the user's power supply reliability requirements. A tiered risk threshold is preset; when the fault propagation probability exceeds the low-risk threshold but does not reach the high-risk threshold, a warning state is entered without triggering an automatic alarm. This mechanism allows maintenance personnel to be aware of potential risks in advance and formulate countermeasures without interfering with normal operations. Furthermore, the formation process of the cascading failure prediction set integrates multi-dimensional impact assessment results. The sequence of affected devices is arranged according to the order of failure propagation, with each device labeled with its failure probability and expected failure time. The failure propagation path describes the possible paths for the failure to spread from the initial device to other devices, with each node on the path representing a device that may fail. The expected power outage range is determined by tracing the power supply users of each device, including the number of affected users, user type, and importance level.

[0070] For example, when a 110kV transformer is shut down for maintenance, its load is transferred to another transformer, causing the load factor of the first transformer to rise from 60% to 85%. The system predicts that the probability of it tripping due to overload after operating under high load for 2 hours is 15%. If this transformer trips, it will cause power loss on two downstream 10kV feeders, affecting approximately 3,000 residential users and 5 industrial users. The cascading fault prediction dataset records this complete fault change process, providing decision support for dispatchers and enabling early risk prevention.

[0071] S104. Perform power risk classification analysis on the cascading fault prediction set to determine the existence of high-risk levels, and extract transformer abnormality data to obtain risk level classification results.

[0072] For each fault scenario in the cascading fault prediction set, numerical values ​​are extracted based on three dimensions: the number of affected users, the duration of the power outage, and the degree of equipment damage. The number of users is mapped to corresponding scores according to preset intervals, the power outage time is converted into hours and multiplied by a time weighting coefficient, and the degree of equipment damage is converted into a loss score based on repair costs. The three scores are added together to obtain a comprehensive risk index. If the comprehensive risk index exceeds the high-risk threshold, the insulation resistance test value and outer sheath crack depth data of the damaged location are retrieved from the cable monitoring database. When the insulation resistance is lower than the standard value or the crack depth exceeds the allowable limit, the cable damage is determined to constitute a deviation from safety regulations. At the same time, the number of times the oil temperature exceeds the limit, the records of abnormal oil levels, and the gas concentration value are obtained from the transformer online monitoring device to identify the type of abnormal transformer symptoms. Based on the types of abnormal signs, the severity of cable damage, and the comprehensive risk index value, scenarios with risk indices exceeding preset thresholds and exhibiting multiple abnormal signs are classified as extremely high-risk, scenarios with only a single abnormal sign are classified as high-risk, scenarios with no abnormal signs but a high risk index are classified as medium-risk, and the rest are classified as low-risk, forming a risk level classification result that includes level identifiers and descriptions of risk-causing factors.

[0073] Specifically, in one implementation, the risk classification of the cascading failure prediction set adopts a quantitative assessment mechanism. Each failure scenario includes three types of basic information: equipment failure sequence, impact range data, and recovery time estimate. The system extracts key indicators from this information to quantify the risk.

[0074] Specifically, the comprehensive risk index is calculated based on standardized, multi-dimensional data. The number of affected users is mapped to 1, 3, and 5 points for 0-500 households, 500-2000 households, and over 2000 households, respectively. The power outage duration is measured in hours, with each hour assigned a weighting coefficient of 0.5. The degree of equipment damage is mapped to 2, 4, and 6 points based on the estimated repair cost as a percentage of the original equipment value, with thresholds of 10%, 30%, and 50%. The comprehensive risk index, obtained by adding these three scores, is classified as high-risk when it exceeds 10 points. This quantitative method allows for the comparison and comprehensive assessment of different types of risk factors within a unified evaluation system.

[0075] It should be noted that the monitoring of cable damage is achieved through distributed fiber optic sensors. These sensors are laid along the cable to detect strain and temperature changes in the insulation layer in real time. When an abnormal signal is detected, the system automatically locates the damage site and measures the insulation resistance. A cable is considered severely damaged if the insulation resistance is below 5 megohms or the crack depth exceeds 30% of the insulation layer thickness.

[0076] For example, the identification of abnormal transformer symptoms relies on comprehensive monitoring of multiple parameters. The oil temperature sensor records the temperature value once per minute, and when it exceeds 85 degrees Celsius for 10 consecutive minutes, it is recorded as an over-limit; the oil level gauge monitors the oil level, and if it deviates from the normal range by more than 10 centimeters, it is recorded as an abnormality; the gas relay detects the gas generated by internal faults in the transformer, and outputs an alarm signal when the concentration exceeds a set threshold.

[0077] In one possible implementation, risk levels are classified using a tiered judgment rule. Extremely high risk corresponds to a comprehensive risk index greater than 15, with both severe cable damage and multiple transformer anomalies present; high risk is defined as a risk index of 10-15 or the presence of a single severe anomaly; medium risk is defined as a risk index of 5-10 with no severe anomalies; all other situations are classified as low risk. Each risk level corresponds to different handling priorities and response measures, forming a complete risk management system.

[0078] S105. Obtain information on high-risk subsets from the risk level classification results, assess the probability distribution of repair operation steps, and adjust the priority of unplanned repair judgments when more than one piece of equipment is being repaired simultaneously or when cross-regional operation coordination is involved, and determine the optimized repair path.

[0079] From the risk level classification results, fault scenarios marked as high-risk and extremely high-risk are selected. For each scenario, the equipment number, fault type, estimated repair time, and geographical coordinates of the equipment are extracted. Based on the power supply topology, the upstream and downstream connections of each device are determined, and a repair task list including equipment priority identifiers is established. For each task in the repair task list, repair records of the same fault type are queried from the historical maintenance database. The percentage of the total repair time for each of the five stages—voltage testing, isolation, grounding, repair, and restoration—is statistically analyzed. Based on the mean and variance of the duration of each stage in historical data, a time probability distribution for each repair stage is constructed. The estimated completion time interval for each task is calculated based on the time probability distribution. When the time intervals of multiple tasks overlap, they are sorted from largest to smallest number of affected users. Considering the personnel dispatch distance across substation areas, tasks with greater distances are postponed, resulting in an adjusted task start timetable. Using the timing arrangement in the task start timetable, with the goal of minimizing the total power outage time, and under the constraint that the interval between adjacent tasks is not less than a preset value, an optimized repair path including task sequence, resource allocation, and estimated completion time is determined by progressively optimizing the task execution order and personnel allocation.

[0080] Specifically, in one implementation, the extraction of the high-risk subset is achieved through multi-layered screening. First, all scenarios marked as high-risk and extremely high-risk are identified from the risk level classification results. Each scenario is associated with a unique fault code, which contains information such as the fault occurrence time, equipment identification, fault type, and scope of impact. By parsing the fault codes, the basic attributes and fault characteristics of the affected equipment are extracted.

[0081] Specifically, the establishment of the repair task list needs to comprehensively consider the importance of the equipment in the power grid. The power supply topology is represented by an adjacency matrix, and the element values ​​in the matrix reflect the electrical connection strength between the equipment. For critical node equipment such as transformers and busbars, the system assigns higher priority weights; for end-point distribution equipment, the priority is relatively lower. Each repair task includes elements such as equipment number, fault description, estimated repair time, number of personnel required, and a list of special tools and equipment.

[0082] For example, the construction of the time probability distribution involves statistical analysis of a large amount of historical maintenance data. Maintenance records of the same or similar fault types from the past three years are extracted from the database. Each record details the entire time process from fault occurrence to repair completion. The repair process is divided into five standard stages: voltage detection, isolation, grounding, repair, and restoration. Each stage has a clear start and end point. The voltage detection stage begins with the arrival of the workers on site and ends with confirmation that the equipment has no voltage; the isolation stage includes disconnecting relevant switches and hanging warning signs; the grounding stage involves the installation of temporary grounding wires; the repair stage is the actual fault handling process; and the restoration stage includes removing temporary measures and restoring power. Through statistical analysis, the average percentage of each stage in the total repair time is calculated, such as voltage detection 5%, isolation 10%, grounding 8%, repair 65%, and restoration 12%. Simultaneously, the standard deviation of each stage's duration is calculated to assess the uncertainty of time. Based on the mean and standard deviation, a normal or log-normal distribution is used to fit the time probability density function of each stage.

[0083] It should be noted that the identification of task time conflicts is based on the judgment of overlapping probability intervals. The completion time interval of each task at a 95% confidence level is calculated according to the probability distribution function. When the time intervals of two or more tasks overlap, it is determined to be a time conflict.

[0084] In one possible implementation, task priority adjustment follows a multi-factor comprehensive evaluation principle. The number of affected users is obtained by querying the power supply marketing system, including the specific number of residential, commercial, and industrial users. The number of users is converted into a standardized impact score, with industrial users having a higher weighting coefficient than commercial users, and commercial users having a higher weighting coefficient than residential users. The personnel dispatch distance for cross-regional operations is calculated using a geographic information system, taking into account actual road distances and travel time. When a task requires dispatching professionals from other substations, the additional travel time is added to the task's estimated start time. For cross-regional operations exceeding 50 kilometers, they are automatically scheduled after local tasks are completed to avoid wasting time due to frequent personnel travel.

[0085] Preferably, the task start timetable is generated using an iterative adjustment method. The initial timetable is arranged according to equipment priority from high to low, and then the time intervals between adjacent tasks are checked. A preset safety interval, typically set to 30 minutes, considers necessary steps such as job transfer, tool preparation, and safety briefings. When an insufficient interval is found, subsequent tasks are postponed until the interval requirement is met. Furthermore, determining the optimized repair path involves multi-objective trade-offs. The total power outage time includes both fault outage time and planned outage time; the system reduces repeated power outages by adjusting the task execution order.

[0086] For example, when two faulty devices are located on the same feeder, they are scheduled for repair within the same power outage window to avoid multiple power outages and restorations. Personnel allocation considers professional skill matching; high-voltage equipment faults require operators with the corresponding qualifications, and the system optimizes allocation based on a personnel skill matrix.

[0087] For example, a 220kV substation experienced two high-risk faults simultaneously: oil leakage in the main transformer and partial discharge in the 10kV switchgear. The main transformer affected 15,000 users, and the switchgear affected 3,000 users. Based on historical data, repairing the main transformer oil leakage typically takes 4-6 hours, while repairing the switchgear requires 2-3 hours. Considering that repairing the main transformer would require deploying professionals from a maintenance base 50 kilometers away, the switchgear fault was prioritized. This allowed the waiting time to be used for tasks with a smaller impact but faster restoration capabilities, thus shortening the overall power outage time.

[0088] Extract the repair time and personnel configuration requirements of each power device from the high-risk subset information, establish equipment maintenance time windows and personnel scheduling tables, identify the distribution of equipment requiring maintenance and the scope of work areas within the same time period, analyze the power outage sequence and power restoration path of equipment in cross-regional operations, adjust the repair order of transformers over cables and main lines over branch lines, and rearrange the execution order of unplanned repairs according to the impact of power outages of more than one device on user power supply to form an optimized repair path.

[0089] Fault codes for each power device are extracted from the high-risk subset information. The average repair time and standard deviation for the same fault type are obtained by querying the historical maintenance database. Simultaneously, the skill levels and numbers of high-voltage electricians, relay protection workers, and cable specialists are extracted from the human resources database. A task list is established, including equipment number, repair time interval, and required personnel types. Based on the repair time intervals of each device in the task list, parallel task groups with overlapping time periods are identified. The work location coordinates of each task are queried from a geographic information database, and the actual distance between tasks is calculated. When multiple tasks require the same type of work and the distance exceeds a preset threshold, personnel conflict status and allocation requirements are recorded. For personnel conflict status, the hierarchical position of each device in the power supply network is extracted from the power grid topology database. Priority values ​​are assigned in descending order: transformers in the first level, busbars in the second level, and feeders in the third level. Simultaneously, the electrical interlocking conditions for power outage operations are identified, and the power outage and restoration sequence of equipment meeting safety regulations is determined. Using the aforementioned power outage and restoration sequence, the number of users during the power outage is calculated based on the product of the number of power-supplying users for each device and the duration of the power outage. By comparing the total number of users during the outage generated by different task execution sequences, the sequence with the smallest value is selected as the optimization result, and an optimized repair path is formed that includes task execution sequence, personnel allocation, and scope of impact.

[0090] Specifically, in one implementation, the high-risk equipment repair task list is constructed based on multi-source data fusion. A unique identifier for each piece of equipment is extracted from the high-risk subset, and this identifier is used to link to a historical fault database, an equipment ledger database, and a human resources management database. The historical fault database stores detailed records of all fault repairs over the past five years, including fault type, handling method, actual time taken, and information on personnel involved.

[0091] Specifically, the determination of the repair time interval employs statistical methods to process historical data. For each fault type, the 100 most recent similar fault repair records are selected, and the mean and standard deviation of the repair time are calculated. The repair time interval is set as the lower limit of the mean minus the standard deviation, and the upper limit of the mean plus twice the standard deviation, thus covering more than 95% of actual repair scenarios. Personnel requirements are obtained by analyzing personnel configuration patterns in historical records. High-voltage equipment faults typically require 2 high-voltage electricians and 1 relay protection worker, while cable faults require 3 cable specialists.

[0092] For example, the process of identifying time conflicts in parallel tasks involves interval overlap judgment and resource competition analysis. The repair time interval of each task is projected onto the time axis. When the projected intervals of two or more tasks intersect, it is further checked whether these tasks require personnel of the same job type. Geographic distance calculation considers not only straight-line distance but also the actual road network and traffic conditions. A geographic information service interface is called to obtain the actual driving routes and estimated travel times between task locations. When the distance between tasks exceeds 30 kilometers or the travel time exceeds 1 hour, the system determines that a separate work team is needed to avoid the time waste and safety risks caused by frequent personnel transfers between different locations.

[0093] It should be noted that the priority weights assigned to the power grid topology hierarchy follow the principle of power supply reliability. As a key device for power conversion, the transformer's failure will cause the entire power supply area to lose power, so it is assigned the highest weight of 10; the busbar connects multiple feeders, and its impact range is next, so the weight is set to 7; the feeder directly supplies power to users, so the weight is 5; the weights of branch lines and distribution transformers are 3 and 2, respectively.

[0094] In one possible implementation, the identification of electrical interlocking conditions is based on power grid operation procedures and equipment operation logic. Electrical interlocking refers to the requirement that certain equipment operations must follow a specific sequence; violation of this sequence may result in short circuits, arcing, or personal injury. A pre-established interlocking rule base includes interlocking between disconnecting switches and circuit breakers, interlocking between grounding switches and disconnecting switches, and sequence constraints for busbar switching operations. When it is determined that a certain piece of equipment needs repair, all relevant interlocking conditions are automatically queried, identifying the sequence of equipment that must be operated first.

[0095] For example, before repairing a 10kV feeder switch, the load-side circuit breaker must be disconnected first, then the line-side isolating switch opened, and finally the grounding switch closed. This order cannot be reversed. By following the interlocking conditions, the determined power outage and restoration sequence satisfies electrical safety requirements while reducing operational steps and power outage time.

[0096] Preferably, the number of users during a power outage is calculated using a cumulative statistical method. Each device is associated with a list of users it supplies, including user type, power capacity, and importance level. The number of users during a power outage equals the number of affected users multiplied by the outage duration, expressed in user-hours. For important users such as hospitals and schools, their weighting factor is set to 3 times that of ordinary users; for industrial users, the weighting factor is 2 times. All possible task execution sequences are enumerated, with each sequence corresponding to a repair order arrangement. Furthermore, the optimization path selection process is achieved by comparing the total number of users during different sequences. A branch-and-bound method is used to reduce computation; when the cumulative number of users during a certain sequence exceeds the current optimal solution, the branch is pruned and the search is stopped. Through this optimization algorithm, in a scenario containing 10 repair tasks, the search space is reduced from 3.62 million permutations to less than 1,000 effective sequences.

[0097] For example, in a certain power supply area, three high-risk faults occurred simultaneously: an abnormal oil level in the 220kV main transformer affecting 8,000 households, reduced insulation on the 10kV busbar affecting 3,000 households, and overheating of a 10kV cable joint affecting 500 households. Considering that repairing the main transformer would take 4 hours but required dispatching experts from afar, the system prioritized repairing the busbar fault affecting 3,000 households, utilizing the 2 hours while waiting for experts to arrive. Then, the main transformer was addressed, and finally, the cable was repaired. This arrangement reduced the total number of households affected by power outages from the initially estimated 44,000 household-hours to 31,000 household-hours, minimizing the impact of power outages while ensuring safety.

[0098] S106. Based on the degree of difference between the optimized repair path and the initial repair operation steps, the optimized path is integrated back into the real-time operation dataset. At the same time, combined with the transformer abnormality data, an updated operation guidance set is obtained.

[0099] The optimized repair path was compared with the initial repair workflow. Differences in task execution order, start time, and personnel allocation were compared item by item. The number of order adjustments, the number of minutes the time was advanced or delayed, and the number of personnel changes were recorded. When any difference exceeded a preset threshold, it was marked as a significant difference, resulting in a difference marker list. Based on this list, the most recent oil temperature sequence, gas relay activation count, and vibration sensor amplitude data were obtained from the transformer online monitoring device. By comparing these with standard operating parameters, abnormal parameters exceeding the normal range were identified, and the corresponding emergency response steps were matched against preset fault handling procedures. The corresponding operations in the original workflow were replaced with these emergency response steps. The task order, adjusted time schedule, and transformer anomaly handling requirements in the optimized path were merged into the corresponding fields of the real-time operation dataset, forming an updated operation guidance set containing the corrected operation sequence and safe operating procedures.

[0100] Specifically, in one implementation, path difference comparison is achieved through a structured comparison method. The optimized repair path and the initial repair workflow are parsed into task sequence lists, each task containing attributes such as task number, execution order, start time, duration, and assigned personnel. By comparing the attribute values ​​of tasks with the same task number one by one, the specific changes are identified.

[0101] Specifically, the difference measurement adopts a categorical statistical method. For task sequence differences, the system calculates the change in the task's position in the sequence; for time differences, it records the number of minutes the start time is advanced or delayed; for personnel differences, it counts the number of personnel number changes. Preset thresholds are determined based on historical operation adjustment experience; differences are marked as significant when the sequence adjustment exceeds 3 positions, the time change exceeds 30 minutes, or more than half of the personnel changes occur. Each marked difference item includes three elements: original value, new value, and magnitude of change, forming a structured difference marker list. Transformer abnormal parameter identification is based on multi-parameter comprehensive judgment. Oil temperature numerical sequences are obtained by sampling every 5 minutes, and the temperature rise rate over the past 2 hours is calculated; an abnormality is determined when the rate exceeds 5 degrees Celsius per hour. The number of gas relay actuations is recorded by a cumulative counter; more than 3 actuations within 24 hours are considered abnormal. Vibration amplitude is collected by an accelerometer; an abnormality is triggered when the peak value exceeds 1.5 times the average value during normal operation.

[0102] It should be noted that the fault handling procedure query uses a pattern matching method. A pre-defined mapping table between fault symptoms and handling measures is provided, with each combination of abnormal parameters corresponding to specific emergency handling steps.

[0103] For example, a rapid rise in oil temperature combined with a gas activation triggers a "primary internal fault" mode. The corresponding handling steps include increasing the monitoring frequency, preparing a backup transformer, and arranging a power outage for inspection.

[0104] In one possible implementation, the job instruction set is updated through field replacement and addition. The records requiring modification are located in the real-time job dataset. The original task order field is replaced with the optimized order, the time schedule field is updated to the adjusted time, and transformer anomaly handling requirements are added to the remarks field. The updated job instruction set contains complete operating procedures, safety precautions, and emergency response plans, providing comprehensive operational guidance for on-site personnel.

[0105] S107. Generate a power operation visualization report based on the updated work instruction set, verify the applicable conditions for not directly activating alarms when deviating from the service standard process, and obtain the final compliance judgment conclusion.

[0106] Based on the task sequence, operation content, and safety measures data in the updated work instruction set, the name, execution time, completion status, and responsible personnel information of each task node are extracted. Data conversion generates flowchart nodes and connections, which are then rendered into a visual report containing a work progress bar, personnel allocation table, and equipment status indicators. The operation sequence displayed in the visual report is compared item by item with the preset power safety operation procedures. This checks whether the power outage operation disconnects the load before isolating, whether the safe distance meets the specified distance, and whether protective equipment is complete. When deviations are found, the deviation type and severity are recorded, and it is determined whether the deviation value exceeds the allowable deviation range. For each deviation, a preset exemption condition list is consulted. If the deviation is due to emergency repair requiring rapid power restoration, on-site equipment failure preventing normal operation, or special handling to protect personnel safety, and the risk caused by the deviation does not reach a high-risk level, then the condition of not triggering an alarm is verified. The final compliance determination is obtained by combining the ratio of the number of compliant items to the total number of items and the result of the deviation rationality judgment.

[0107] Specifically, in one implementation, the visualization report is built based on structured data transformation. The system reads fields such as task number, task name, planned time, actual time, execution status, and responsible person from the updated work instruction set, and converts these discrete data into graphical display elements. The flowchart uses nodes to represent tasks and arrows to indicate the execution order; the Gantt chart uses time on the horizontal axis and task items on the vertical axis, with bar lengths indicating task durations; equipment status is indicated by different colors for three states: normal, faulty, and under maintenance.

[0108] Specifically, the comparison of power safety operating procedures employs a rule-matching mechanism. A pre-set operating procedure database includes mandatory requirements such as disconnecting the load switch before opening the isolating switch during power outages, maintaining a safe distance of at least 0.7 meters while working on live lines, and wearing insulated gloves and shoes for high-voltage work. During the comparison, the system checks each item to ensure the actual operation meets the requirements. For items that do not meet the requirements, the deviation is categorized into three levels: minor deviation, moderate deviation, and serious deviation. Minor deviation refers to slight adjustments in the operating sequence that do not affect safety; moderate deviation refers to some deficiencies in safety measures but manageable risks; serious deviation refers to violations of mandatory safety regulations and the existence of potential safety hazards.

[0109] For example, the verification of exemption conditions involves a comprehensive judgment of multiple factors. Emergency repair scenarios include situations such as large-scale power outages requiring rapid restoration, power outages for important users, and risks of equipment failure escalating. Equipment restrictions cover situations such as switch mechanisms being stuck and unable to operate normally, temporary shortages of safety tools, and narrow working spaces making it difficult to maintain standard safe distances. The principle of prioritizing personnel safety refers to simplifying operating procedures to reduce personnel exposure time during severe weather such as thunderstorms and heavy rain.

[0110] It should be noted that the quantitative assessment of compliance is achieved through compliance rate calculation. The total number of all inspection items and the number of items conforming to the regulations are counted, and the compliance rate percentage is calculated. When the compliance rate is higher than 90% and all mandatory items are met, it is considered fully compliant; when the compliance rate is between 70% and 90% and there are no serious deviations, it is considered basically compliant; when the compliance rate is lower than 70% or there are serious deviations, it is considered non-compliant.

[0111] In one possible implementation, the final compliance assessment includes evaluations across multiple dimensions. In addition to compliance scores, it includes explanations of the reasonableness of deviations, descriptions of risk control measures, and improvement recommendations. Integrating this information into a structured assessment report clearly identifies the compliance level of the operation, key risk points, and subsequent precautions, providing a basis for management decisions and on-site personnel to improve their operations.

[0112] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from the spirit or essential characteristics of this application. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of this application is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within this application. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for supervising the guarantee of power technical services, characterized in that, include: Collect information on the type of work tools, the level of protective equipment, and the maintenance process; integrate fault handling information on cable damage or transformer malfunctions; and generate a real-time work dataset. Based on the real-time operation dataset, the matching degree between tool models and protection equipment levels is identified, compliance deviation factors of unplanned repairs are incorporated, and operational rationality assessment criteria are determined. Based on the operational rationality assessment criteria, the maintenance operation process is extracted, the equipment impact range is identified, the impact of the fault handling plan is assessed, and a cascading fault prediction set is generated. This includes: obtaining the switching of disconnecting switches, the opening and closing of circuit breakers, and the location of grounding wires; identifying the equipment number list of transformers, busbars, and feeders; querying the power grid connection relationship based on the equipment number list, tracing the power source points and load distribution, and determining the power flow redistribution path; and retrieving historical fault tripping probabilities based on the power flow redistribution path, combining them with real-time load data, calculating the node overload probability, and generating a cascading fault prediction set that includes faulty equipment and the outage range. The risk classification of the predicted cascading faults is performed, cable damage information is integrated to distinguish deviations from regulations, transformer anomaly data is extracted, and risk level classification results are generated. The process involves extracting a high-risk subset from the risk level classification results, assessing the probability distribution of the repair process, adjusting the priority of more than one device or cross-regional operations, and generating an optimized repair path. This includes: filtering high-risk scenarios from the risk level classification results, extracting device numbers, fault types, repair durations, and location coordinates, and establishing a repair task list; querying historical repair records based on the repair task list, constructing the time probability distribution of each repair stage, and calculating the task completion time interval; sorting the number of users affected by each device based on the time probability distribution, adjusting the timing of cross-regional tasks, optimizing the task order and personnel allocation, and generating an optimized repair path containing task sequences and resource configurations; extracting device fault codes and repair durations from the high-risk subset, querying a human resources database, and establishing a task list; identifying parallel task groups based on the task list, calculating the distance between work locations, and recording personnel conflict states; extracting device hierarchical locations based on the personnel conflict states, determining the power outage and restoration order, calculating the number of users during power outages, and generating an optimized repair path containing task sequences and personnel allocations. Based on the differences between the optimized repair path and the initial maintenance operation process, the optimized repair path is integrated into the real-time operation dataset to generate an updated operation guidance set. Based on the updated set of work instructions, the exemption conditions that deviate from the specifications are identified, and a compliance determination result is generated.

2. The method for supervising power technology service guarantees according to claim 1, characterized in that, The data collection includes the model of the work tools, the level of protection equipment, and the maintenance procedures. It integrates fault handling information such as cable damage or transformer malfunction to generate a real-time work dataset, including: The operation frequency of the data acquisition tool is matched with the standard procedure database to extract the insulation level of protective gloves and the safety helmet level. The installation sequence of grounding wires and the operation sequence of circuit breakers are recorded to generate the initial work record. Based on the initial work record, detect the wear depth of the cable sheath, extract the ambient humidity and temperature gradient, or read the transformer vibration amplitude, compare the fault handling plan and protection level, and determine the compliance of the work specifications. Based on the compliance with the work specifications, a real-time work dataset containing full-process monitoring is generated.

3. The method for supervising power technology service guarantees according to claim 1, characterized in that, The process involves identifying the matching degree between tool models and protection equipment levels based on the real-time operation dataset, incorporating compliance deviations from unplanned repairs, and determining operational rationality assessment criteria, including: Obtain the insulation class, withstand voltage value and applicable voltage range of the tool, compare it with the standard tool library, obtain the insulation resistance and level of the protective equipment, and generate a preliminary matching score; Based on the initial matching score being lower than the standard, the operation sequence is extracted, the time nodes and behaviors that deviate from the procedure are identified, and a reasonableness judgment value is generated by combining environmental constraints and the urgency of the task. Based on the aforementioned reasonableness judgment value, a comprehensive evaluation result is calculated, and an operational reasonableness evaluation criterion containing judgment thresholds and handling rules is generated.

4. The method for supervising power technology service guarantees according to claim 1, characterized in that, The process involves identifying the matching degree between tool specifications and protection levels based on the real-time job dataset, incorporating compliance deviations from unplanned repairs, and determining operational rationality assessment criteria, including: Obtain the tool number, compare it with the standard tool list, and calculate the specification deviation of non-standard tools; Based on the specification deviation, query the protection configuration database, calculate the protection parameter margin, and determine the unavailability of standard tools by combining the tool inventory status and the urgency of the task. Based on the unavailability of the standard tools, the dimensions of the workspace and the distance between equipment are extracted, environmental adaptability indicators are calculated, and adaptation risks and safety risks are combined to generate operational rationality assessment criteria for the use of non-standard tools.

5. The method for supervising power technology service guarantees according to claim 1, characterized in that, The process involves risk classification of the predicted cascading failures, integrating cable damage information to differentiate deviations from regulations, extracting transformer anomaly data, and generating risk level classification results, including: The number of users, power outage time, and equipment damage level are extracted from the cascading failure prediction set to calculate a comprehensive risk index. Based on the comprehensive risk index, the cable insulation resistance and crack depth are retrieved to identify transformer oil temperature exceeding the limit and gas concentration, and to distinguish deviations from the regulations. Based on the comprehensive risk index and abnormal signs, risk levels are divided, and a risk level classification result containing level identifiers and risk-causing factors is generated.

6. The method for supervising power technology service guarantees according to claim 1, characterized in that, The step of integrating the optimized repair path into the real-time job dataset based on the differences between the optimized repair path and the initial maintenance operation process, and generating an updated job guidance set, includes: Compare the optimized repair path with the initial process, record the task order and time differences, and match the emergency response steps; Based on the emergency response steps, replace the original process operations, merge and optimize the paths and abnormal handling requirements, and generate an updated set of work instructions.

7. The method for supervising power technology service guarantees according to claim 1, characterized in that, The step of confirming the exemption conditions for deviations from the specifications based on the updated work instruction set and generating a compliance determination result includes: Extract task nodes and safety measures from the updated job instruction set to generate a flowchart; According to the flowchart, compare the operation sequence with the safety procedures, check the power outage operation and protective equipment, and record the deviation type. Based on the deviation type, the exemption conditions are queried, and the compliance degree and reasonableness of the deviation are considered to generate a compliance judgment result.

Images