A method and system for identifying and dynamically balancing the operating conditions of charging piles

By constructing a method for identifying and dynamically balancing charging pile operating conditions, and utilizing weekly operating condition feature values ​​and historical data mapping resources, the problems of false alarms in small samples and false alarms in large samples are solved. This enables accurate identification and efficient scheduling in the operation and maintenance of charging piles, improving operation and maintenance efficiency and decision-making adaptability.

CN123070936APending Publication Date: 2026-11-10SHANDONG LUNENG SOFTWARE TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611441036.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-16
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

In the operation and maintenance of electric vehicle charging stations, existing technologies struggle to balance the suppression of false alarms in small samples and the identification of false alarms in large samples under large-scale, low-staff operation and maintenance scenarios. Furthermore, they cannot provide lightweight and interpretable decision-making support, resulting in low operation and maintenance efficiency.

Method used

By constructing a method for identifying and dynamically balancing the operating conditions of charging piles, progressive statistical quantification is performed using weekly operating condition characteristic values, recent operating condition values, and operating condition drift benchmark values. A mapping relationship between the level of operating condition deviation and operational resources is established by combining historical data, resource matching and dynamic balancing decisions are made, a governance task list is generated, and parameters are optimized and adjusted through closed-loop feedback.

Benefits of technology

It enables accurate identification and efficient scheduling of charging pile operating conditions, ensuring that high-value tasks are prioritized, personnel workload is reasonable, resource utilization is optimized, and operation and maintenance efficiency and decision-making adaptability are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123070936A_ABST
    Figure CN123070936A_ABST
Patent Text Reader

Abstract

This application provides a method and system for identifying and dynamically balancing charging pile operating conditions, belonging to the field of industrial big data processing and operational resource optimization scheduling technology. The method involves: acquiring weekly operating data, on-site operational feedback data, and spare parts data of charging piles; calculating weekly operating condition characteristic values ​​and recent operating condition values; then calculating operating condition drift benchmark values ​​and operating condition deviation levels to form an operating condition deviation set; extracting historical operational resource records based on the operating condition deviation levels to establish a mapping between operating condition deviation levels and operational resources; sorting by operating condition deviation levels and matching operational personnel and spare parts based on the comprehensive ability characteristics of operators and spare parts inventory; calculating task utility scores and performing dynamic balancing decisions based on workload saturation to generate a governance task list; collecting execution feedback data and comparing it with preset targets to adjust configuration parameters and complete closed-loop optimization. This application can improve the accuracy of operation and maintenance and resource utilization efficiency, and is suitable for large-scale, low-staff operation and maintenance scenarios.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application belongs to the field of industrial big data processing and operation resource optimization scheduling technology, specifically involving a method and system for identifying and dynamically balancing the operating conditions of charging piles. Background Technology

[0002] With the explosive growth of electric vehicle charging networks, charging pile coverage has expanded from core business districts to remote areas. The operational conditions of charging equipment are becoming increasingly diverse. Meanwhile, the growth of professional maintenance personnel has lagged behind. The contradiction between the surge in the number of charging piles and the insufficient supply of maintenance personnel is becoming increasingly prominent. Furthermore, the massive amounts of operational data are multi-source, heterogeneous, and high-dimensional, making traditional manual processing methods insufficient to meet the demands. As the scale expands, the growth of maintenance personnel lags behind the growth in equipment, leading to a surge in maintenance workload per person. The gap between on-site maintenance workload and manpower allocation continues to widen, urgently requiring efficient data-driven decision-making methods.

[0003] Currently, the most similar existing technical solutions in the charging pile operation and maintenance field mainly fall into two categories. One is a rule-based monitoring method based on fixed thresholds, which sets fixed alarm thresholds for key indicators such as charging success rate and failure rate. When an indicator exceeds the limit, an operation and maintenance work order is automatically generated. This approach has two problems in practical applications: First, for low-frequency, small-sample charging piles, the small statistical base means that a single random failure can cause significant fluctuations in the success rate and trigger false alarms. Second, for high-frequency, large-sample charging piles, the slow increase in the number of failures is barely reflected in the success rate indicator, making it difficult to trigger threshold alarms, leading to the accumulation and expansion of failures. The other approach is a data-driven prediction method based on deep learning, which collects time-series data such as charging pile voltage and current, and trains a fault classifier using models such as LSTM and Transformer to predict and assess the future operating status of the equipment. The application of this method is limited by the specific characteristics of charging pile operation and maintenance scenarios. The distribution of charging pile operation data varies significantly across different regions and different commissioning times, and the model's generalization ability on newly commissioned or low-frequency used equipment is insufficient, making it difficult to guarantee prediction accuracy.

[0004] In summary, existing solutions struggle to balance small-sample false alarm suppression and large-sample false negative identification in large-scale, low-staff operation and maintenance scenarios. They also fail to provide interpretable decision-making support under limited computing power. Therefore, a lightweight, interpretable, and adaptive method for identifying operational drift and balancing resource scheduling is urgently needed. Summary of the Invention

[0005] In a first aspect, embodiments of this application provide a method for identifying and dynamically balancing the operating conditions of a charging pile, comprising the following steps: S1. Obtain the charging pile weekly operation dataset, charging pile on-site operation feedback dataset, and spare parts dataset as input data; S2. Preprocess the input data, calculate the weekly operating condition characteristic value of a single charging pile, calculate the recent operating condition value of a single charging pile based on the weekly operating condition characteristic value, calculate the operating condition drift benchmark value based on the recent operating condition value, and calculate the operating condition deviation level of each charging pile based on the recent operating condition value and the operating condition drift benchmark value to form an operating condition deviation set. S3. Based on the deviation levels of each working condition in the working condition deviation set, extract the historical operation resource records corresponding to each working condition deviation level from the charging pile on-site operation feedback dataset, establish the mapping relationship between each working condition deviation level and operation resources, and use it as the working condition drift identification result. S4. Based on the working condition drift identification results, the charging piles are sorted from high to low according to the working condition deviation level. The charging piles ranked first are marked as priority charging piles. Based on the comprehensive working ability characteristic value of each operator and the inventory status of each spare part, the corresponding operators and spare parts are matched for the priority charging piles. S5. Based on the matching results and the preset work constraint rules, calculate the utility score of each charging pile work task to be processed, and make dynamic equilibrium decisions based on the work saturation of each worker to generate a governance task list. S6. Collect the execution feedback data of the governance task list, compare it with the preset target, and adjust the configuration parameters of at least one processing link in steps S3 to S5 in reverse according to the comparison results to complete the closed-loop optimization.

[0006] Secondly, embodiments of this application also provide a charging pile operating condition drift identification and dynamic equilibrium decision-making system, including: The data acquisition module is used to acquire charging pile weekly operation dataset, charging pile on-site operation feedback dataset, and spare parts dataset as input data; The data preprocessing module is used to preprocess the input data, calculate the weekly operating condition characteristic value of a single charging pile, calculate the recent operating condition value of a single charging pile based on the weekly operating condition characteristic value, calculate the operating condition drift benchmark value based on the recent operating condition value, and calculate the operating condition deviation level of each charging pile based on the recent operating condition value and the operating condition drift benchmark value, forming an operating condition deviation set. The working condition drift identification module is used to extract historical operation resource records corresponding to each working condition deviation level from the charging pile on-site operation feedback dataset based on each working condition deviation level in the working condition deviation set, and establish a mapping relationship between each working condition deviation level and operation resources as the working condition drift identification result. The resource matching module is used to sort the charging piles from high to low according to the working condition drift identification results, mark the charging piles at the top of the list as priority charging piles, and match the corresponding operators and spare parts to the priority charging piles according to the comprehensive working ability characteristics of each operator and the inventory status of each spare part. The dynamic equilibrium decision module is used to calculate the utility score of each charging pile operation task to be processed based on the matching results and the preset operation constraint rules, and to make dynamic equilibrium decisions based on the workload of each operator to generate a governance task list. The closed-loop feedback optimization module is used to collect execution feedback data from the governance task list, compare it with preset targets, and adjust the configuration parameters of at least one of the following modules—the operating condition drift identification module, the resource matching module, and the dynamic equilibrium decision module—based on the comparison results to complete the closed-loop optimization.

[0007] As can be seen from the above technical solutions, this application has the following advantages: The charging pile operating condition drift identification and dynamic equilibrium decision-making method and system provided in this application solves the problems of the fixed threshold method being sensitive to small sample random fluctuations and insensitive to large sample slow deterioration by constructing weekly operating condition characteristic values, recent operating condition values, operating condition drift benchmark values, and operating condition deviation levels for progressive statistical quantification. By automatically establishing the mapping relationship between operating condition deviation levels and work resources based on historical data, and combining personnel comprehensive ability assessment and spare parts inventory constraints, the system achieves accurate allocation of manpower and materials to charging piles with high deviation levels and high efficiency utilization. Through dynamic equilibrium decision-making that links task utility scores with work saturation, the system ensures that high-value tasks are prioritized, low-value tasks are proactively released, and personnel load is maintained within a reasonable range. Through weekly closed-loop parameter iteration driven by execution feedback, the system's adaptive optimization capability is enabled, achieving accurate identification and efficient scheduling in large-scale, low-management operation and maintenance scenarios. Attached Figure Description

[0008] To more clearly illustrate the technical solution of this application, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic diagram of the overall process of the charging pile operating condition drift identification and dynamic equilibrium decision-making method of the present invention.

[0010] Figure 2 This is a schematic diagram illustrating the specific process of the charging pile operating condition drift identification and dynamic equilibrium decision-making method of the present invention.

[0011] Figure 3 This is a schematic diagram of the charging pile operating condition drift identification and dynamic equilibrium decision system of the present invention. Detailed Implementation

[0012] The various embodiments of this disclosure will be described more fully in the following detailed description of the specific steps of the charging pile operating condition drift identification and dynamic equilibrium decision-making method. This disclosure may have various embodiments, and adjustments and changes may be made therein. However, it should be understood that there is no intention to limit the various embodiments of this disclosure to the specific embodiments disclosed herein, but rather this disclosure should be understood to cover all adjustments, equivalents, and / or alternatives falling within the spirit and scope of the various embodiments of this disclosure.

[0013] This embodiment provides a method for identifying and dynamically balancing the operating conditions of charging piles. By statistically quantifying the operating condition drift and matching the resource gradient, it solves the problems of false alarms due to random fluctuations in small samples and unrepaired problems in large samples, and achieves accurate deployment of limited resources in high-value scenarios.

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] Please see Figure 1 The diagram shows a flowchart of a charging pile operating condition drift identification and dynamic equilibrium decision-making method in a specific embodiment. The method includes the following steps: S1. Obtain the charging pile weekly operation dataset, charging pile on-site operation feedback dataset, and spare parts dataset as input data; It should be noted that this step, by acquiring weekly operation data of charging piles, on-site operation feedback data, and spare parts data, has built a data foundation that combines operation status monitoring, maintenance resource guarantee, and historical handling experience, providing a basis for operation and maintenance decisions. S2. Preprocess the input data, calculate the weekly operating condition characteristic value of a single charging pile, calculate the recent operating condition value of a single charging pile based on the weekly operating condition characteristic value, calculate the operating condition drift benchmark value based on the recent operating condition value, and calculate the operating condition deviation level of each charging pile based on the recent operating condition value and the operating condition drift benchmark value to form an operating condition deviation set. It should be noted that this step calculates the weekly operating condition characteristic value and recent operating condition value of a single charging pile, and determines the operating condition deviation level of each charging pile by combining the operating condition drift benchmark value. This can effectively identify abnormal trends in the equipment operating status. It solves the problem of false alarms under small sample data and the problem of slow fault omissions under large sample data, and realizes accurate classification and priority ranking of the operating status of massive charging piles. S3. Based on the deviation levels of each working condition in the working condition deviation set, extract the historical operation resource records corresponding to each working condition deviation level from the charging pile on-site operation feedback dataset, establish the mapping relationship between each working condition deviation level and operation resources, and use it as the working condition drift identification result. It should be noted that this step is based on the deviation level of the working condition. The system extracts the corresponding level of work resource records from the historical work feedback data, establishes the mapping relationship between the working condition level and the work resources, and can automatically match the spare parts, handling methods and working hours required for a specific fault level based on historical maintenance experience, thus realizing the advance preparation and recommendation of maintenance resources. S4. Based on the working condition drift identification results, the charging piles are sorted from high to low according to the working condition deviation level. The charging piles ranked first are marked as priority charging piles. Based on the comprehensive working ability characteristic value of each operator and the inventory status of each spare part, the corresponding operators and spare parts are matched for the priority charging piles. It should be noted that charging piles are prioritized for treatment based on the order of deviation level from high to low. Combined with the comprehensive ability characteristics of the operators and the status of spare parts inventory, the coordination and matching of personnel and materials are achieved. This ensures that highly skilled personnel handle high-priority tasks, and that task allocation is consistent with resource availability, avoiding duplicate work or work interruption caused by resource mismatch. S5. Based on the matching results and the preset work constraint rules, calculate the utility score of each charging pile work task to be processed, and make dynamic equilibrium decisions based on the work saturation of each worker to generate a governance task list. It should be noted that by calculating the utility score of the tasks to be processed and combining it with the workload of the operators, a governance task list is generated. This comprehensively considers the task value, processing time and personnel load, achieving objective allocation of tasks and load balancing of human resources, thereby improving overall operation and maintenance efficiency. S6. Collect the execution feedback data of the governance task list, compare it with the preset target, and adjust the configuration parameters of at least one processing link in steps S3 to S5 in reverse according to the comparison results to complete the closed-loop optimization. It should be noted that by collecting execution feedback data and comparing it with preset targets, the configuration parameters of the working condition identification, resource matching and decision-making links can be adjusted in reverse to form a closed-loop optimization, so that the decision model can adaptively adjust according to the actual execution effect, ensuring the continuous effectiveness of the operation and maintenance strategy.

[0016] This embodiment solves the problems of false alarms due to random fluctuations in small samples and uncorrected problems in large samples by statistical quantification of operating condition drift and resource gradient matching, and realizes the accurate allocation of limited resources in high-value scenarios; combined with dynamic equilibrium decision-making that links utility scores and saturation, an executable governance list is generated.

[0017] Furthermore, as a refinement and extension of the specific implementation methods of the above embodiments, in order to fully illustrate the specific implementation process in this embodiment, as follows: Figure 2 As shown, another method for identifying and dynamically balancing charging pile operating conditions is provided, which includes the following steps: S1. Obtain the charging pile weekly operation dataset, charging pile on-site operation feedback dataset, and spare parts dataset as input data; The specific steps of step S1 are as follows: S11. Obtain the charging pile weekly operation dataset SJYX, which includes the following fields: week date SJ1YX, charging pile province / city location SJ2YX, total number of charging times SJ3YX, charging amount SJ4YX, total number of component failures SJ5YX, component failure name SJ6YX, number of component failures SJ7YX, total number of reasons for abnormal shutdown at the charging pile SJ8YX, reasons for shutdown at the charging pile SJ9YX, number of reasons for shutdown at the charging pile SJ10YX, and number of user IDs for abnormally ended charging SJ11YX. S12. Obtain the charging pile on-site operation feedback dataset SJFK, which includes the following fields: personnel ID SJ1FK, operation hours SJ2FK, week and date SJ3FK, component name and model SJ4FK, quantity SJ5FK, fault category SJ6FK, handling method SJ7FK, and operation type SJ8FK; the operation type includes hardware replacement, software upgrade, configuration maintenance, and testing and troubleshooting. S13. Obtain the spare parts dataset SJBJ, which includes the following fields: province and city location of spare parts SJ1BJ, name and model of spare parts SJ2BJ, inventory quantity of spare parts SJ3BJ, and circulation status of spare parts SJ4BJ. S14. Use the charging pile weekly operation dataset SJYX, the charging pile on-site operation feedback dataset SJFK, and the spare parts dataset SJBJ as input data; S2. Preprocess the input data, calculate the weekly operating condition characteristic value of a single charging pile, calculate the recent operating condition value of a single charging pile based on the weekly operating condition characteristic value, calculate the operating condition drift benchmark value based on the recent operating condition value, and calculate the operating condition deviation level of each charging pile based on the recent operating condition value and the operating condition drift benchmark value to form an operating condition deviation set. In step S2, the weekly operating condition characteristic value is used to characterize the multi-dimensional operating status of a single charging pile within a single statistical period; the recent operating condition value is used to eliminate the influence of a single random fluctuation and characterize the stable operating condition trend of the charging pile over a continuous period of time; the operating condition drift benchmark value is used to establish a statistical reference standard for normal operating conditions in each dimension; the operating condition deviation level is used to quantify the degree of deviation of each charging pile from the benchmark value, providing a grading basis for subsequent resource matching; in step S2, the weekly operating condition characteristic value of a single charging pile is calculated, and the specific steps are as follows: For a single charging pile dx, calculate the weekly operating condition characteristic value T=(TLH,TWX,TWD,TJS,TJR) based on the charging pile weekly operation dataset SJYX. Among them, TLH represents the degree of aging, TWX represents the degree of danger, TWD represents the degree of stability, TJS represents the degree of damage, and TJR represents the charging compatibility. Among them, the aging degree (TLH) is used to characterize the impact of accumulated component failures on the effective operating time of the charging pile; the danger degree (TWX) is used to characterize the severity level of the current failure; the stability degree (TWD) is used to characterize the continuity level of the charging pile operation; the cumulative damage degree (TJS) is used to characterize the cumulative damage of shutdown-type failures; and the charging compatibility degree (TJR) is used to characterize the impact of abnormal user behavior on the charging success rate. The above five dimensions describe the charging pile operating conditions from three levels: equipment status, operating performance, and user experience, respectively, providing a quantitative basis for multi-dimensional drift identification. The aging degree TLH is calculated by multiplying the total number of component failures SJ5YX by the preset standard repair time for a single failure, and then dividing by the standard operating time of one week. Let the standard repair time for a charging station shutdown caused by a single component failure be . (For example, take) =15 minutes), the standard weekly runtime is (Specifically =7 days × 24 hours / day × 60 minutes / hour = 10080 minutes), then the aging degree TLH is calculated according to the following formula:

[0018] SJ5YX represents the total number of component failures within the current statistical period (i.e., one week); this ratio represents the proportion of downtime caused by component failures to the total weekly operating time, and the larger the value, the more severe the aging of the equipment. In practical applications, and It can be adjusted according to the actual operation and maintenance scenario; This ratio represents the proportion of downtime caused by component failures to the total weekly operating time; a higher value indicates a more severe degree of equipment aging. In practical applications, it represents the standard repair time for a single failure. and a standard week's runtime These values ​​can be adjusted according to different operation and maintenance scenarios and are not limited to the specific values ​​mentioned above. The danger level TWX calculates according to the following logic: When the component fault name SJ6YX matches the preset dangerous keyword list and the corresponding component fault count SJ7YX≥SJ7YX_MIN, TWX=SJ7YX; otherwise, TWX=0. Among them, SJ7YX_MIN is the minimum number of component failures; The list of hazardous keywords includes, but is not limited to, fault keywords related to personal safety or core equipment functions, such as insulation and guidance. It can be configured and updated according to the type of charging pile equipment and operation and maintenance specifications. The stability factor TWD is calculated using the following formula:

[0019] Among them, SJ8YX represents the total number of reasons for abnormal shutdown at the pile end, and SJ3YX represents the total number of charging times; The degree of accumulated loss (TJS) is calculated according to the following logic: When the cause of the pile end shutdown failure SJ9YX matches the preset list of dangerous keywords, TJS=SJ10YX; otherwise, TJS=0, where SJ10YX is the number of times the cause of the pile end shutdown failure has been detected. TJR calculates charging compatibility using the following formula:

[0020]

[0021]

[0022] in, SJ11YX represents the number of user IDs that abnormally ended charging, SJ3YX represents the total number of charging times, and SJ8YX represents the total number of reasons for abnormal shutdown at the charging station. SJ4YX represents the average charge per charge, while SJ4YX represents the charge amount. In step S2, the recent operating condition value of a single charging pile is calculated based on the weekly operating condition characteristic value. The specific steps are as follows: For a single charging pile dx, the recent operating condition value L=(LLH,LWX,LWD,LJS,LJR) is calculated based on the weekly operating condition characteristic value T. Among them, LLH is the recent value of aging level, LWX is the recent value of danger level, LWD is the recent value of stability level, LJS is the recent value of accumulated damage level, and LJR is the recent value of charging compatibility. Specifically, for each historical week going back from this week, the aging degree TLH, the danger degree TWX, the stability degree TWD, the accumulated damage degree TJS, and the charging compatibility TJR are calculated week by week according to the method of calculating the weekly operating condition characteristic value T, and the calculated values ​​of each week are stored in the corresponding dataset. For each dataset, backtracking truncation is performed according to the following rules, and recent operating condition values ​​are calculated: LLH = mean(LLH dataset) + std(LLH dataset); LLH dataset = {TLH1, TLH2, ..., TLHy}; The LLH dataset is a backtracking dataset that starts from the total number of component failures SJ5YX1 of this week and goes back y weeks. Only values ​​that continuously satisfy the condition of total number of component failures SJ5YX≥SJ5YX_MIN are retained. Once a week does not meet this condition, the backtracking is terminated immediately. Where SJ5YX_MIN is the minimum number of failures in the total number of component failures, mean is the mean function, and std is the standard deviation function; LWX = mean(LWX dataset) + std(LWX dataset); LWX dataset = {TWX1,TWX2,…,TWX} y}; The LWX dataset is a backtracking dataset that starts from the number of component failures SJ7YX1 in this week and goes back y weeks. It retains only the values ​​that continuously satisfy the condition that the component failure name SJ6YX hits the list of dangerous keywords and the corresponding number of component failures SJ7YX ≥ SJ7YX_MIN. Once a week does not meet the condition, the backtracking is terminated immediately. Wherein, SJ7YX_MIN is the minimum dangerous number of component failures, mean is the mean function, and std is the standard deviation function; LWD = mean(LWD dataset) + std(LWD dataset); LWD dataset = {TWD1, TWD2, ..., TWD} y}; The LWD dataset is generated by backtracking y weeks from the total number of charging times SJ3YX1 and the total number of abnormal shutdown reasons at the charging piles SJ8YX1 this week. Only values ​​that continuously satisfy the coefficient of variation CV≥SJ3YX_MIN are retained. Once a week that does not meet the condition is found, the backtracking is terminated immediately. Among them, SJ3YX_MIN is the minimum coefficient of variation of the cause of abnormal shutdown at the pile end, mean is the mean function, and std is the standard deviation function; It should be noted that the coefficient of variation (CV) is calculated according to the standard definition in statistics, which is the ratio of the standard deviation to the mean. LJS = mean(LJS dataset) + std(LJS dataset); LJS dataset = {TJS1, TJS2, ..., TJS} y}; The LJS dataset is a backtracking dataset that starts from the number of times the pile end shutdown failures occur in this week (SJ10YX1) and goes back y weeks. Only the values ​​that continuously meet the condition that the pile end shutdown failure cause (SJ9YX) hits the list of dangerous keywords and the corresponding number of times the pile end shutdown failure cause (SJ10YX) occurs are retained. If a week does not meet the condition, the backtracking is terminated immediately. Wherein, SJ10YX_MIN is the minimum number of causes of pile end shutdown failure, mean is the mean function, and std is the standard deviation function; LJR = mean(LJR dataset) + std(LJR dataset) LJR dataset = {TJR1,TJR2,…,TJR} y} The LJR dataset starts from the number of abnormally ended charging user IDs SJ11YX1 and the charging amount SJ4YX1 of this week and traces back y weeks. Only values ​​that continuously satisfy the condition of number of abnormally ended charging user IDs SJ11YX > SJ11YX_MIN and charging amount SJ4YX ≥ SJ4YX_MIN are retained. Once a week that does not meet the conditions is found, the backtracking is terminated immediately. Wherein, SJ4YX_MIN is the minimum amount of charging power at the charging pile, SJ11YX_MIN is the minimum number of user IDs that abnormally ended charging, mean is the mean function, and std is the standard deviation function. In step S2, the operating condition drift reference value is calculated based on the recent operating condition values. The specific steps are as follows: Calculate the baseline drift value K for each drift category based on the recent operating condition value L. , , , , ); in, , , , , These correspond to the deviation levels of operating conditions for hardware system, component performance, software function, system parameter status, and scenario adaptability, respectively. A hardware system class drift dataset LHK is constructed. From the recently calculated aging values ​​(LLH) of each charging pile, the LLH values ​​of all charging piles are extracted to form the dataset LHK={d1·LLH1,d2·LLH2,…,d…}. i •LLH i …,d n •LLH n}, Where n is the total number of charging piles, LLH i Let d be the recent value of the aging degree of the i-th charging pile. i The charging station numbered i; Calculate the hardware system class drift baseline value using the following formula. :

[0023] in, Let LHK be the population mean of the dataset. Let LHK be the population standard deviation. This is the deviation adjustment factor; Construct a component performance drift dataset LCP. Extract the LWX values ​​of all charging piles from the calculated recent danger levels (LWX) of each charging pile, forming the dataset LCP={LWX1,LWX2,…,LWX}. i …,LWX n}; Where n is the total number of charging piles, LWX i This represents the recent value of the danger level of the i-th charging pile; The component performance drift baseline value is calculated using the following formula. :

[0024] in, Let LCP be the population mean of the dataset. The population standard deviation of the LCP dataset; A software function class drift dataset LSF is constructed. From the calculated recent stability values ​​(LWD) of each charging station, the LWD values ​​of all charging stations are extracted to form the dataset LSF={LWD1,LWD2,…,LWD}. i …,LWD n}; Where n is the total number of charging piles, LWD i Let be the recent value of the stability of the i-th charging pile; The software function class drift baseline value is calculated using the following formula. :

[0025] in, The population mean of the dataset LSF Let LSF be the population standard deviation of the dataset. A system parameter state drift dataset LPS is constructed. From the recently calculated LJS values ​​of each charging pile's accumulated damage level, the LJS values ​​of all charging piles are extracted to form the dataset LPS={LJS1,LJS2,…,LJS}. i …,LJS n}; Where n is the total number of charging piles, LJS i This represents the recent value of the accumulated damage level of the i-th charging pile; Calculate the system parameter state-class drift baseline value using the following formula. :

[0026] in, The population mean of the dataset LPS. The population standard deviation of the LPS dataset; Construct a scenario-adaptive class drift dataset LSA. Extract the LJR values ​​of all charging piles from the calculated recent charging compatibility values ​​LJR of each charging pile to form the dataset LSA={LJR1,LJR2,…,LJR…} i …,LJR n}; Where n is the total number of charging piles, LJR i Let be the recent compatibility value of the i-th charging pile; The scene adaptability drift baseline value is calculated using the following formula. :

[0027] in, The population mean of the LSA dataset. The population standard deviation of the LSA dataset; In step S2, the operating condition deviation level of each charging pile is calculated based on the recent operating condition values ​​and the operating condition drift benchmark value. The steps are as follows: Based on the recent operating condition value L and the operating condition drift baseline value K, the five types of operating condition deviation levels for each charging pile are calculated, forming an operating condition deviation set G={g1,g2,…,g i ,…,g n}; Where g i ={ , , , , }, , , , , The deviation levels of the operating conditions correspond to the hardware system category, component performance category, software function category, system parameter status category, and scenario adaptability category of the i-th charging pile, respectively. The deviation level determination function DF is calculated using the following formula:

[0028] Output ; in, For operating conditions deviating from the standard, The input is the recent operating condition value. for The drift reference value for the corresponding drift category under operating conditions. To achieve balanced granularity and meet the requirements , The preset total number of levels, For deviation from adjustment factor, This represents the overall standard deviation of the corresponding drift category dataset; Balanced granularity Used to control the fineness of the deviation level classification of operating conditions; The smaller the value, the more detailed the classification, which is suitable for high-value sites that require refined management; The larger the value, the coarser the classification, which is suitable for routine scenarios that require rapid screening; The value of satisfies It can be dynamically adjusted according to the urgency of maintenance; For each charging pile, the five recent operating condition values ​​LLH, LWX, LWD, LJS, and LJR are substituted into the operating condition deviation level judgment function DF to obtain the five types of operating condition deviation levels for each charging pile. , , , , ; By backtracking from this week and retaining only weeks that continuously meet the minimum threshold condition, occasional random fluctuations can be effectively filtered out, ensuring that only anomalies that show a continuous deterioration trend are included in recent operating condition statistics; once a week that does not meet the condition is found, the backtracking is terminated immediately, avoiding interference from outdated data on the current operating condition assessment. S3. Based on the deviation levels of each working condition in the working condition deviation set, extract the historical operation resource records corresponding to each working condition deviation level from the charging pile on-site operation feedback dataset, establish the mapping relationship between each working condition deviation level and operation resources, and use it as the working condition drift identification result. The specific steps of step S3 are as follows: S31. Based on the set of operating condition deviations G, calculate the drift reference value for each operating condition deviation level k. :

[0029] in, This represents the overall mean of the corresponding dataset. To achieve balanced granularity; S32. For each operating condition deviation level k, based on the drift reference value at that level. Extract historical operation resource records corresponding to this level from the charging pile on-site operation feedback dataset SJFK; the operation resource r = (component name type SJ4FK, processing method SJ7FK, operation type SJ8FK, quantity SJ5FK, operation time SJ2FK). S33. Establish the mapping RMP between operating condition deviation levels and operating resources according to the following formula:

[0030] in, Same level The following is a list of charging pile operation resources. This is a mapping function used to map the operational deviation level k to the corresponding set of operational resources; The RMap function is a data-driven map builder that takes as input a deviation level k of the operating condition and a list of historical job resources at the same level. The output is a mapping table between this level and the corresponding resource set; the establishment of this mapping relationship does not rely on expert experience or manual rules, but directly extracts the actual operational resources used under each level from historical operational data. S34. Calculate the mode matching resource for each level according to the following formula:

[0031] in, It is the mode function. This is the set of job resources corresponding to level k. Return the resource with the most repetitions at level k, and use it as the standard matching resource for level k; S4. Based on the working condition drift identification results, the charging piles are sorted from high to low according to the working condition deviation level. The charging piles ranked first are marked as priority charging piles. Based on the comprehensive working ability characteristic value of each operator and the inventory status of each spare part, the corresponding operators and spare parts are matched for the priority charging piles. The specific steps of step S4 are as follows: S41. Based on the charging pile on-site operation feedback dataset SJFK, extract the operation records of each operator and calculate the comprehensive operation capability feature value RS of each operator according to the following formula:

[0032] in, Let i be the number of assignments corresponding to the i-th skill. The total number of skills mastered by the workers. For the number of years the operator has worked, The maximum number of years of service among all workers is represented by w1, w2, and w3, which are weighting coefficients and satisfy w1+w2+w3=1. The overall performance characteristic value RS is derived from skill proficiency. Breadth of skills and years of experience The three dimensions comprehensively evaluate the operator's ability to avoid the one-sidedness of evaluating based on only a single dimension; w1, w2, and w3 are the weight coefficients of the three dimensions, and the relative importance of the three dimensions can be adjusted according to the operation and maintenance management strategy; under the default configuration, w1, w2, and w3 are 0.4, 0.3, and 0.3 respectively, indicating that skill proficiency contributes the most to the overall ability; S42. Establish a mapping relationship between each operator and the manageable deviation level of the working condition according to the following formula. :

[0033] in, For the workers, For the workers The work conditions that can be performed deviate from the grade. This refers to the set of operational resources for the corresponding level. S43. Sort the charging piles from high to low according to the deviation level k of the working condition, and mark the charging piles at the top of the list as priority charging piles for treatment. S44. Based on the deviation level k of the charging piles in priority for management, query the mapping relationship. Match operators with corresponding processing capabilities, and match spare parts that meet the inventory requirements based on the inventory quantity SJ3BJ and circulation status SJ4BJ of each spare part in the spare parts dataset SJBJ. S5. Based on the matching results and the preset work constraint rules, calculate the utility score of each charging pile work task to be processed, and make dynamic equilibrium decisions based on the work saturation of each worker to generate a governance task list. The specific steps of step S5 are as follows: S51. For each charging pile dx to be processed, the task requirement a = {operating condition deviation level k, required resources r, required comprehensive capability RS, required time limit s} is calculated according to the following formula:

[0034] in, Let be the number of resources corresponding to task requirement 'a'. , , These are the utility components of processing method, job type, and job hours, respectively. For any dimension X (X∈{processing method SJ7FK, job type SJ8FK, job time SJ2FK}), the corresponding utility component V(X) is calculated according to the following minimum catch-all model:

[0035] in, and These represent the frequency and quantity of dimension X under the same working condition k this week. When X=SJ7FK This represents the frequency of different processing methods SJ7FK under the same working condition k this week. The corresponding number of occurrences; When X=SJ8FK This represents the frequency of different work types SJ8FK under the same working condition k this week. The corresponding number of occurrences; When X = SJ2FK This represents the frequency of SJ2FK occurrences under different working hours of the same working condition k this week. The corresponding number of occurrences; It should be noted that "same type of working condition k" means that the working conditions have the same deviation level, that is, charging piles with the same deviation level are considered to belong to the same type of working condition. and They are respectively and The value after normalization according to the following formula: , ; The term "same type of working condition k" refers to working conditions with the same level of deviation. The minimum catch-all model uses a structure consisting of the sum of the minimum value and a discount term equal to 0.3 times the maximum value, where... and The values ​​are normalized for frequency and quantity, respectively. The smaller of the two values ​​is taken as the base utility, and the larger value is multiplied by the advantage discount factor of 0.3 and then summed. This design allows the high-frequency or large-volume operation dimension to obtain a higher utility contribution, while the discount factor of 0.3 prevents the advantage of a single dimension from becoming too dominant, maintaining a relative balance among the three dimensions. S52. Calculate the workload saturation Ssat of each worker according to the following formula:

[0036] in, Total time spent by a single person Actual available working hours; Total time spent by a single person in step S52 Calculate as follows:

[0037] in, For on-site work hours, according to calculate; For travel time; Deducting working hours for traveling at night; The travel time Determined according to the following job constraint rules: Within the same city =0, inter-city within the province =4 hours, cross-province =8 hours; Working hours will be deducted for traveling at night. Determined according to the following rules: When traveling at night =2 hours and the number of nighttime travels per week does not exceed 2 times, when not traveling =0; S53. Set the comprehensive evaluation range for workload saturation (e.g., set it to 60%≤Ssat≤85%), and set the task utility equilibrium value PPat>mean(PP)+std(PP); Where mean(PP) is the mean of the utility scores of all pending charging pile tasks this week, std(PP) is the standard deviation of the utility scores of all pending charging pile tasks this week, and PP is the task utility score. Work saturation is used to quantify the utilization of workers' time resources; a value that is too low indicates idleness, while a value that is too high indicates overwork; total time spent per person =60%-85% is the comprehensive evaluation range for workload saturation. When it is below 60%, the personnel utilization rate can be improved by increasing the maintenance width. When it is above 85%, the personnel load can be reduced by decreasing the maintenance granularity, so as to ensure that the workload of each operator is maintained at a reasonable level. The upper and lower limits of the comprehensive evaluation range for workload saturation can be adjusted according to the actual carrying capacity of the team, and are not limited to the above specific values. S54. Based on the task utility score of each task and the workload of each worker, make decisions on task allocation and release to generate a charging pile governance set. ; in, For the xth assigned worker, For the charging piles to be processed and assigned to the operator, Charging piles awaiting processing The task utility score, where x is the total number of assigned tasks; Based on the charging pile management set ZZ, personnel dispatch instructions and spare parts allocation instructions are generated; Step S54, which involves task allocation and release decisions and the generation of the governance set, specifically includes: S541. For each worker hx, determine whether the task utility score of each task in the set of tasks to be executed is greater than the task utility equilibrium value PPat; If not, the corresponding task will be released; If so, the corresponding task will be added to the task set of this worker hx; S542. Calculate the workload saturation Ssat of each worker hx performing the task set; Determine if the workload saturation (Ssat) is less than the lower limit of the overall workload saturation evaluation range (e.g., 60%): If so, adjust the number N of deviations from the working condition ranking to increase the maintenance width; S543. Determine whether the workload saturation Ssat is greater than the upper limit of the comprehensive evaluation range of workload saturation (e.g., 85%). If so, adjust the granularity e to reduce the maintenance granularity; S544. If the workload saturation Ssat is between the lower and upper limits (e.g., between 60% and 85%), then generate a charging pile management set. Based on the governance set ZZ, personnel dispatch instructions and spare parts allocation instructions are generated; S6. Collect the execution feedback data of the governance task list, compare it with the preset target, and adjust the configuration parameters of at least one processing link in steps S3 to S5 in reverse according to the comparison results to complete the closed-loop optimization. The specific steps of step S6 are as follows: S61. Collect execution feedback data from the governance task list, wherein the execution feedback data includes fault handling results, changes in equipment indicators, resource consumption, and personnel operation feedback; S62. Compare the execution feedback data with the preset target: If the equipment indicators meet the preset constraints and resource consumption is controlled within the preset reasonable range, then the current decision-making logic and configuration parameters are solidified. If the equipment indicators fail to meet the preset constraints or the resource consumption exceeds the preset reasonable range, the feedback will be sent back to at least one of the processing steps in steps S3 to S5. S63. Based on the feedback results, adjust at least one of the following configuration parameters: operating condition drift identification threshold (including SJ5YX_MIN, SJ7YX_MIN, SJ3YX_MIN, SJ10YX_MIN, SJ4YX_MIN, SJ11YX_MIN), deviation adjustment factor β, balance granularity e, weight coefficients w1, w2, w3, and comprehensive evaluation range of work saturation; The deviation adjustment factor β is used to adjust the sensitivity of the drift baseline value to the dispersion of the dataset. The larger the β value, the more lenient the baseline value, and the lower the overall deviation level of the operating condition. The smaller the β value, the more stringent the baseline value, and the higher the overall deviation level of the operating condition. The value of β can be adjusted according to the site operation and maintenance strategy. For example, a smaller value can be used for sites with strict operation and maintenance standards to improve sensitivity, and a larger value can be used for remote areas to reduce the false alarm rate. It should be noted that the β in the above five drift baseline value formulas is the same deviation adjustment factor, which is globally uniform to ensure that all drift categories are judged under the same sensitivity scale. In step S6, the adjustment of configuration parameters is not random, but rather based on the comparison between execution feedback data and preset targets, the corresponding configuration parameters are located and adjusted in a targeted manner, specifically including: When the equipment availability or charging success rate is lower than the preset target, it indicates that the current operating condition drift identification sensitivity is insufficient. At least one of the following parameters needs to be adjusted: reduce the deviation adjustment factor β to improve the overall identification sensitivity; or reduce the minimum threshold corresponding to each drift category (SJ5YX_MIN, SJ7YX_MIN, SJ10YX_MIN) to lower the alarm threshold; or reduce the balance granularity e to refine the division granularity of the operating condition deviation level, so that more charging piles in the critical state can be included in the management scope. When the workload saturation Ssat of maintenance personnel is consistently below the lower limit of the comprehensive evaluation range, it indicates that the number of charging piles currently included in the governance is insufficient to fully utilize manpower. It is necessary to increase the maintenance depth level parameter N to expand the governance coverage and include more charging piles in this governance. When Ssat is consistently above the upper limit of the comprehensive evaluation range, it indicates that the current maintenance granularity is too fine, resulting in excessive processing time for a single pile. It is necessary to increase the equilibrium granularity e to coarsen the level division and reduce unnecessary fine maintenance operations. When the processing efficiency of high-deviation-level charging piles is lower than the preset standard, it indicates that the personnel capability assessment weights do not match the actual situation, and the relative proportions of the weight coefficients w1, w2, and w3 need to be adjusted: if skill proficiency has a greater impact on processing efficiency, then increase w1; if skill breadth has a greater impact, then increase w2; if years of work experience (i.e., years of experience) have a greater impact, then increase w3. When the deviation between the actual maintenance man-hours and the estimated man-hours in S52 continues to exceed the preset range, it indicates that there is a deviation between the man-hour parameters in the RMP mapping table and the actual operation and maintenance data. The man-hour parameters in the standard matching resources under the corresponding working condition deviation level need to be updated with the average actual working man-hours SJ2FK of the current period. When low-value tasks frequently consume operation and maintenance resources and high-value tasks are not processed first, it indicates that the utility score threshold is set too low. The task utility equilibrium value PPat needs to be appropriately increased to release low-value tasks and ensure that high-value tasks are executed first. S64. Complete closed-loop optimization once a week, and use the optimized configuration parameters for the next cycle of operating condition drift identification and dynamic equilibrium decision-making; Through weekly parameter iterations, the thresholds, weights, and adjustment coefficients continuously converge to the actual operational needs, avoiding the drawbacks of traditional fixed-rule systems that gradually fail due to their inability to adapt to dynamic changes.

[0038] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0039] like Figure 3 As shown, the following are embodiments of the charging pile operating condition drift identification and dynamic equilibrium decision system provided in this disclosure. This system and the charging pile operating condition drift identification and dynamic equilibrium decision method in the above embodiments belong to the same inventive concept. For details not described in detail in the embodiments of the charging pile operating condition drift identification and dynamic equilibrium decision system, please refer to the embodiments of the above charging pile operating condition drift identification and dynamic equilibrium decision method.

[0040] The system includes: The data acquisition module is used to acquire charging pile weekly operation dataset, charging pile on-site operation feedback dataset, and spare parts dataset as input data; The data preprocessing module is used to preprocess the input data, calculate the weekly operating condition characteristic value of a single charging pile, calculate the recent operating condition value of a single charging pile based on the weekly operating condition characteristic value, calculate the operating condition drift benchmark value based on the recent operating condition value, and calculate the operating condition deviation level of each charging pile based on the recent operating condition value and the operating condition drift benchmark value, forming an operating condition deviation set. The working condition drift identification module is used to extract historical operation resource records corresponding to each working condition deviation level from the charging pile on-site operation feedback dataset based on each working condition deviation level in the working condition deviation set, and establish a mapping relationship between each working condition deviation level and operation resources as the working condition drift identification result. The resource matching module is used to sort the charging piles from high to low according to the working condition drift identification results, mark the charging piles at the top of the list as priority charging piles, and match the corresponding operators and spare parts to the priority charging piles according to the comprehensive working ability characteristics of each operator and the inventory status of each spare part. The dynamic equilibrium decision module is used to calculate the utility score of each charging pile operation task to be processed based on the matching results and the preset operation constraint rules, and to make dynamic equilibrium decisions based on the workload of each operator to generate a governance task list. The closed-loop feedback optimization module is used to collect execution feedback data from the governance task list, compare it with preset targets, and adjust the configuration parameters of at least one of the following modules—the operating condition drift identification module, the resource matching module, and the dynamic equilibrium decision module—based on the comparison results to complete the closed-loop optimization.

[0041] This embodiment achieves precise operation and maintenance and efficient resource scheduling of charging piles through the interactive collaboration of a data acquisition module, a data preprocessing module, a working condition drift identification module, a resource matching module, a dynamic equilibrium decision-making module, and a closed-loop feedback optimization module.

[0042] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying drift in charging pile operating conditions and making dynamic equilibrium decisions, characterized in that, Includes the following steps: S1. Obtain the charging pile weekly operation dataset, charging pile on-site operation feedback dataset, and spare parts dataset as input data; S2. Preprocess the input data, calculate the weekly operating condition characteristic value of a single charging pile, calculate the recent operating condition value of a single charging pile based on the weekly operating condition characteristic value, calculate the operating condition drift benchmark value based on the recent operating condition value, and calculate the operating condition deviation level of each charging pile based on the recent operating condition value and the operating condition drift benchmark value to form an operating condition deviation set. S3. Based on the deviation levels of each working condition in the working condition deviation set, extract the historical operation resource records corresponding to each working condition deviation level from the charging pile on-site operation feedback dataset, establish the mapping relationship between each working condition deviation level and operation resources, and use it as the working condition drift identification result. S4. Based on the working condition drift identification results, the charging piles are sorted from high to low according to the working condition deviation level. The charging piles ranked first are marked as priority charging piles. Based on the comprehensive working ability characteristic value of each operator and the inventory status of each spare part, the corresponding operators and spare parts are matched for the priority charging piles. S5. Based on the matching results and the preset work constraint rules, calculate the utility score of each charging pile work task to be processed, and make dynamic equilibrium decisions based on the work saturation of each worker to generate a governance task list. S6. Collect the execution feedback data of the governance task list, compare it with the preset target, and adjust the configuration parameters of at least one processing link in steps S3 to S5 in reverse according to the comparison results to complete the closed-loop optimization.

2. The charging pile operating condition drift identification and dynamic equilibrium decision-making method according to claim 1, characterized in that, The specific steps of step S1 are as follows: S11. Obtain the charging pile weekly operation dataset SJYX, which includes the following fields: week date SJ1YX, charging pile province / city location SJ2YX, total number of charging times SJ3YX, charging amount SJ4YX, total number of component failures SJ5YX, component failure name SJ6YX, number of component failures SJ7YX, total number of reasons for abnormal shutdown at the charging pile SJ8YX, reasons for shutdown at the charging pile SJ9YX, number of reasons for shutdown at the charging pile SJ10YX, and number of user IDs for abnormally ended charging SJ11YX. S12. Obtain the charging pile on-site operation feedback dataset SJFK, which includes the following fields: personnel ID SJ1FK, operation hours SJ2FK, week date SJ3FK, component name and model SJ4FK, quantity SJ5FK, fault category SJ6FK, handling method SJ7FK, and operation type SJ8FK; S13. Obtain the spare parts dataset SJBJ, which includes the following fields: province and city location of spare parts SJ1BJ, name and model of spare parts SJ2BJ, inventory quantity of spare parts SJ3BJ, and circulation status of spare parts SJ4BJ. S14. Use the charging pile weekly operation dataset SJYX, the charging pile on-site operation feedback dataset SJFK, and the spare parts dataset SJBJ as input data.

3. The charging pile operating condition drift identification and dynamic equilibrium decision-making method according to claim 1, characterized in that, In step S2, the weekly operating characteristic value of a single charging pile is calculated. The specific steps are as follows: For a single charging pile dx, calculate the weekly operating condition characteristic value T=(TLH,TWX,TWD,TJS,TJR) based on the charging pile weekly operation dataset SJYX. Among them, TLH represents the degree of aging, TWX represents the degree of danger, TWD represents the degree of stability, TJS represents the degree of damage, and TJR represents the charging compatibility. The aging degree TLH is calculated by multiplying the total number of component failures SJ5YX by the preset standard repair time for a single failure, and then dividing by the standard operating time of one week. The danger level TWX calculates according to the following logic: When the component fault name SJ6YX matches the preset dangerous keyword list and the corresponding component fault count SJ7YX≥SJ7YX_MIN, TWX=SJ7YX; otherwise, TWX=0. Among them, SJ7YX_MIN is the minimum number of component failures; The stability factor TWD is calculated using the following formula: Among them, SJ8YX represents the total number of reasons for abnormal shutdown at the pile end, and SJ3YX represents the total number of charging times; The degree of accumulated loss (TJS) is calculated according to the following logic: When the cause of the pile end shutdown failure SJ9YX matches the preset list of dangerous keywords, TJS=SJ10YX; otherwise, TJS=0, where SJ10YX is the number of times the cause of the pile end shutdown failure has been detected. TJR calculates charging compatibility using the following formula: in, SJ11YX represents the number of user IDs that abnormally ended charging, SJ3YX represents the total number of charging times, and SJ8YX represents the total number of reasons for abnormal shutdown at the charging station. SJ4YX represents the average charge per charge, while SJ4YX represents the charge amount.

4. The charging pile operating condition drift identification and dynamic equilibrium decision-making method according to claim 3, characterized in that, In step S2, the recent operating condition value of a single charging pile is calculated based on the weekly operating condition characteristic value. The specific steps are as follows: For a single charging pile dx, the recent operating condition value L=(LLH,LWX,LWD,LJS,LJR) is calculated based on the weekly operating condition characteristic value T. Among them, LLH is the recent value of aging level, LWX is the recent value of danger level, LWD is the recent value of stability level, LJS is the recent value of accumulated damage level, and LJR is the recent value of charging compatibility. Specifically, for each historical week going back from this week, the aging degree TLH, the danger degree TWX, the stability degree TWD, the accumulated damage degree TJS, and the charging compatibility TJR are calculated week by week according to the method of calculating the weekly operating condition characteristic value T, and the calculated values ​​of each week are stored in the corresponding dataset. For each dataset, backtracking truncation is performed according to the following rules, and recent operating condition values ​​are calculated: LLH = mean(LLH dataset) + std(LLH dataset); LLH dataset = {TLH1, TLH2, ..., TLHy}; The LLH dataset is a backtracking dataset that starts from the total number of component failures in this week and goes back y weeks. Only values ​​that continuously satisfy the condition SJ5YX≥SJ5YX_MIN are retained. Once a week does not meet this condition, the backtracking is terminated immediately. Where SJ5YX_MIN is the minimum number of failures in the total number of component failures, mean is the mean function, and std is the standard deviation function; LWX = mean(LWX dataset) + std(LWX dataset); LWX dataset = {TWX1,TWX2,…,TWX} y }; The LWX dataset is a backtracking dataset that starts from the number of component failures in this week and goes back y weeks. It retains only the values ​​that continuously satisfy the condition that the component failure name SJ6YX hits the list of dangerous keywords and the corresponding number of component failures SJ7YX ≥ SJ7YX_MIN. Once a week that does not meet the condition is found, the backtracking is terminated immediately. Wherein, SJ7YX_MIN is the minimum dangerous number of component failures, mean is the mean function, and std is the standard deviation function; LWD = mean(LWD dataset) + std(LWD dataset); LWD dataset = {TWD1, TWD2, ..., TWD} y }; The LWD dataset is generated by backtracking y weeks from the total number of charging times and the total number of reasons for abnormal shutdowns at charging piles this week. Only values ​​that continuously satisfy the coefficient of variation CV≥SJ3YX_MIN are retained. Once a week that does not meet the condition is found, the backtracking is terminated immediately. Among them, SJ3YX_MIN is the minimum coefficient of variation of the cause of abnormal shutdown at the pile end, mean is the mean function, and std is the standard deviation function; LJS = mean(LJS dataset) + std(LJS dataset); LJS dataset = {TJS1, TJS2, ..., TJS} y }; The LJS dataset is a backtracking dataset that starts from the number of times the pile end shutdown failure causes occur in this week and goes back y weeks. Only the values ​​that continuously meet the condition that the pile end shutdown failure cause SJ9YX continuously hits the list of dangerous keywords and the corresponding number of times the pile end shutdown failure cause SJ10YX is greater than or equal to SJ10YX_MIN are retained. Once a week does not meet the condition, the backtracking is terminated immediately. Wherein, SJ10YX_MIN is the minimum number of causes of pile end shutdown failure, mean is the mean function, and std is the standard deviation function; LJR = mean(LJR dataset) + std(LJR dataset) LJR dataset = {TJR1,TJR2,…,TJR} y } The LJR dataset starts from the number of user IDs that abnormally ended charging and the charging amount in this week and traces back y weeks. Only values ​​that continuously satisfy the condition of the number of user IDs that abnormally ended charging SJ11YX>SJ11YX_MIN and the charging amount SJ4YX≥SJ4YX_MIN are retained. Once a week that does not meet the conditions is found, the backtracking is terminated immediately. Wherein, SJ4YX_MIN is the minimum amount of charging power at the charging pile, SJ11YX_MIN is the minimum number of user IDs that abnormally ended charging, mean is the mean function, and std is the standard deviation function.

5. The charging pile operating condition drift identification and dynamic equilibrium decision-making method according to claim 4, characterized in that, In step S2, the operating condition drift reference value is calculated based on the recent operating condition values. The specific steps are as follows: Calculate the baseline drift value K for each drift category based on the recent operating condition value L. , , , , ); in, , , , , These correspond to the deviation levels of operating conditions for hardware system, component performance, software function, system parameter status, and scenario adaptability, respectively. A hardware system class drift dataset LHK is constructed. From the recently calculated aging values ​​(LLH) of each charging pile, the LLH values ​​of all charging piles are extracted to form the dataset LHK={d1·LLH1,d2·LLH2,…,d…}. i •LLH i …,d n •LLH n }, Where n is the total number of charging piles, LLH i Let d be the recent value of the aging degree of the i-th charging pile. i The charging station numbered i; Calculate the hardware system class drift baseline value using the following formula. : in, Let LHK be the population mean of the dataset. Let LHK be the population standard deviation. This is the deviation adjustment factor; Construct a component performance drift dataset LCP. Extract the LWX values ​​of all charging piles from the calculated recent danger levels (LWX) of each charging pile, forming the dataset LCP={LWX1,LWX2,…,LWX}. i …,LWX n }; Where n is the total number of charging piles, LWX i This represents the recent value of the danger level of the i-th charging pile; The component performance drift baseline value is calculated using the following formula. : in, Let LCP be the population mean of the dataset. The population standard deviation of the LCP dataset; A software function class drift dataset LSF is constructed. From the calculated recent stability values ​​(LWD) of each charging station, the LWD values ​​of all charging stations are extracted to form the dataset LSF={LWD1,LWD2,…,LWD}. i …,LWD n }; Where n is the total number of charging piles, LWD i Let be the recent value of the stability of the i-th charging pile; The software function class drift baseline value is calculated using the following formula. : in, The population mean of the dataset LSF Let LSF be the population standard deviation of the dataset. A system parameter state drift dataset LPS is constructed. From the recently calculated LJS values ​​of each charging pile's accumulated damage level, the LJS values ​​of all charging piles are extracted to form the dataset LPS={LJS1,LJS2,…,LJS}. i …,LJS n }; Where n is the total number of charging piles, LJS i This represents the recent value of the accumulated damage level of the i-th charging pile; Calculate the system parameter state-class drift baseline value using the following formula. : in, The population mean of the dataset LPS. The population standard deviation of the LPS dataset; Construct a scenario-adaptive class drift dataset LSA. Extract the LJR values ​​of all charging piles from the calculated recent charging compatibility values ​​LJR of each charging pile to form the dataset LSA={LJR1,LJR2,…,LJR…} i …,LJR n }; Where n is the total number of charging piles, LJR i Let be the recent compatibility value of the i-th charging pile; The scene adaptability drift baseline value is calculated using the following formula. : in, The population mean of the LSA dataset. The population standard deviation of the LSA dataset; In step S2, the operating condition deviation level of each charging pile is calculated based on the recent operating condition values ​​and the operating condition drift benchmark value. The steps are as follows: Based on the recent operating condition value L and the operating condition drift baseline value K, the five types of operating condition deviation levels for each charging pile are calculated, forming an operating condition deviation set G={g1,g2,…,g i ,…,g n }; Where g i ={ , , , , }, , , , , The deviation levels of the operating conditions correspond to the hardware system category, component performance category, software function category, system parameter status category, and scenario adaptability category of the i-th charging pile, respectively. The deviation level determination function DF is calculated using the following formula: Output ; in, For operating conditions deviating from the standard, The input is the recent operating condition value. for The drift reference value for the corresponding drift category under operating conditions. To achieve balanced granularity and meet the requirements , The preset total number of levels, For deviation from adjustment factor, This represents the overall standard deviation of the corresponding drift category dataset; For each charging pile, the five recent operating condition values ​​LLH, LWX, LWD, LJS, and LJR are substituted into the operating condition deviation level judgment function DF to obtain the five types of operating condition deviation levels for each charging pile. , , , , .

6. The charging pile operating condition drift identification and dynamic equilibrium decision-making method according to claim 5, characterized in that, The specific steps of step S3 are as follows: S31. Based on the set of operating condition deviations G, calculate the drift reference value for each operating condition deviation level k. : in, This represents the overall mean of the corresponding dataset. To achieve balanced granularity; S32. For each operating condition deviation level k, based on the drift reference value at that level. Extract historical operation resource records corresponding to this level from the charging pile on-site operation feedback dataset SJFK; the operation resource r = (component name type SJ4FK, processing method SJ7FK, operation type SJ8FK, quantity SJ5FK, operation time SJ2FK). S33. Establish the mapping RMP between operating condition deviation levels and operating resources according to the following formula: in, () are of the same level The following is a list of charging pile operation resources. This is a mapping function used to map the operational deviation level k to the corresponding set of operational resources; S34. Calculate the mode matching resource for each level according to the following formula: in, It is the mode function. This is the set of job resources corresponding to level k. Return the resource with the most repetitions at level k, and use it as the standard matching resource for level k.

7. The charging pile operating condition drift identification and dynamic equilibrium decision-making method according to claim 6, characterized in that, The specific steps of step S4 are as follows: S41. Based on the charging pile on-site operation feedback dataset SJFK, extract the operation records of each operator and calculate the comprehensive operation capability feature value RS of each operator according to the following formula: in, Let i be the number of assignments corresponding to the i-th skill. The total number of skills mastered by the workers. For the number of years the operator has worked, The maximum number of years of service among all workers is represented by w1, w2, and w3, which are weighting coefficients and satisfy w1+w2+w3=1. S42. Establish a mapping relationship between each operator and the manageable deviation level of the working condition according to the following formula. : in, For the workers, For the workers The work conditions that can be performed deviate from the grade. This refers to the set of operational resources for the corresponding level. S43. Sort the charging piles from high to low according to the deviation level k of the working condition, and mark the charging piles at the top of the list as priority charging piles for treatment. S44. Based on the deviation level k of the charging piles in priority for management, query the mapping relationship. It matches operators with corresponding processing capabilities, and at the same time matches spare parts that meet the inventory requirements based on the inventory quantity SJ3BJ and circulation status SJ4BJ of each spare part in the spare parts dataset SJBJ.

8. The charging pile operating condition drift identification and dynamic equilibrium decision-making method according to claim 7, characterized in that, The specific steps of step S5 are as follows: S51. For each charging pile dx to be processed, the task requirement a = {operating condition deviation level k, required resources r, required comprehensive capability RS, required time limit s} is calculated according to the following formula: in, Let be the number of resources corresponding to task requirement 'a'. , , These are the utility components of processing method, job type, and job hours, respectively. S52. Calculate the workload saturation Ssat of each worker according to the following formula: in, Total time spent by a single person Actual available working hours; S53. Set the comprehensive evaluation interval for workload saturation, and set the task utility equilibrium value PPat>mean(PP)+std(PP); Where mean(PP) is the mean of the utility scores of all pending charging pile tasks this week, std(PP) is the standard deviation of the utility scores of all pending charging pile tasks this week, and PP is the task utility score. S54. Based on the task utility score of each task and the workload of each worker, make decisions on task allocation and release to generate a charging pile governance set. ; in, For the xth assigned worker, For the charging piles to be processed and assigned to the operator, Charging piles awaiting processing The task utility score, where x is the total number of assigned tasks; Based on the charging pile management set ZZ, personnel dispatch instructions and spare parts allocation instructions are generated.

9. The charging pile operating condition drift identification and dynamic equilibrium decision-making method according to claim 8, characterized in that, The specific steps of step S6 are as follows: S61. Collect execution feedback data from the governance task list, wherein the execution feedback data includes fault handling results, changes in equipment indicators, resource consumption, and personnel operation feedback; S62. Compare the execution feedback data with the preset target: If the equipment indicators meet the preset constraints and resource consumption is controlled within the preset reasonable range, then the current decision-making logic and configuration parameters are solidified. If the equipment indicators fail to meet the preset constraints or the resource consumption exceeds the preset reasonable range, the feedback will be sent back to at least one of the processing steps in steps S3 to S5. S63. Based on the feedback results, adjust at least one of the following configuration parameters: operating condition drift identification threshold, deviation adjustment multiple β, balance granularity e, weight coefficients w1, w2, w3, and comprehensive evaluation range of work saturation; S64. Complete closed-loop optimization once a week, and use the optimized configuration parameters for the next cycle of operating condition drift identification and dynamic equilibrium decision-making.

10. A charging pile operating condition drift identification and dynamic equilibrium decision-making system, characterized in that, include: The data acquisition module is used to acquire charging pile weekly operation dataset, charging pile on-site operation feedback dataset, and spare parts dataset as input data; The data preprocessing module is used to preprocess the input data, calculate the weekly operating condition characteristic value of a single charging pile, calculate the recent operating condition value of a single charging pile based on the weekly operating condition characteristic value, calculate the operating condition drift benchmark value based on the recent operating condition value, and calculate the operating condition deviation level of each charging pile based on the recent operating condition value and the operating condition drift benchmark value, forming an operating condition deviation set. The working condition drift identification module is used to extract historical operation resource records corresponding to each working condition deviation level from the charging pile on-site operation feedback dataset based on each working condition deviation level in the working condition deviation set, and establish a mapping relationship between each working condition deviation level and operation resources as the working condition drift identification result. The resource matching module is used to sort the charging piles from high to low according to the working condition drift identification results, mark the charging piles at the top of the list as priority charging piles, and match the corresponding operators and spare parts to the priority charging piles according to the comprehensive working ability characteristics of each operator and the inventory status of each spare part. The dynamic equilibrium decision module is used to calculate the utility score of each charging pile operation task to be processed based on the matching results and the preset operation constraint rules, and to make dynamic equilibrium decisions based on the workload of each operator to generate a governance task list. The closed-loop feedback optimization module is used to collect execution feedback data from the governance task list, compare it with preset targets, and adjust the configuration parameters of at least one of the following modules—the operating condition drift identification module, the resource matching module, and the dynamic equilibrium decision module—based on the comparison results to complete the closed-loop optimization.