A new energy power battery maintenance operation monitoring method and system

Abstract

The present application relates to the technical field of battery maintenance, in particular to a new energy power battery maintenance operation supervision method and system. The present application builds a multi-dimensional verification model for maintenance and realizes millisecond-level synchronous deep integration of multi-source heterogeneous data, thereby uniformly mapping the process dynamic data with high space-time accuracy into the multi-dimensional compliance verification framework of the standardized process, forming an encrypted data packet by synchronously extracting visual key frames and operation parameters at the moment of violation judgment, and driving equipment closed-loop control through hierarchical early warning, thereby building a real-time closed-loop control mechanism for identification, early warning and forced intervention, which can immediately block when a violation action occurs or is about to occur. Based on the association storage of the encrypted data packet and the blockchain hash value, an unalterable evidence chain covering the process details is formed, realizing the improvement from result traceability to process credibility and detail certification, and ensuring the integrity of the maintenance whole-process data in the time, cause and responsibility dimensions.

Description

Technical Field

[0001] This invention relates to the field of battery repair technology, and in particular to a method and system for monitoring and supervising the repair operations of new energy power batteries. Background Technology

[0002] The new energy power battery maintenance operation supervision method refers to a comprehensive management system that uses technology, systems and management methods to standardize, monitor and trace the entire process of power battery operation, from fault diagnosis, disassembly, repair, replacement to quality testing, information recording and traceability. The core objectives are to ensure the safety of maintenance operations, ensure the reliability of maintenance quality, prevent safety risks, achieve full life cycle traceability, and comply with relevant national laws and standards.

[0003] Currently, the maintenance of new energy power batteries is monitored through the full life-cycle traceability technology. The principle is to assign a unique physical and digital identifier to each power battery, conforming to national standards, binding it with information from its production, use, and historical maintenance. This lays the core foundation for traceable maintenance operations. Then, relying on a unified traceability platform, the entire process data—including pre-maintenance qualification verification, in-maintenance operation documentation and parameter monitoring, and post-maintenance quality comparison—is encrypted, distributed, and verified according to rules, achieving violation warnings and closed-loop supervision across the entire chain. However, this full life-cycle traceability technology can only record the results before and after maintenance. For core operational stages such as unpacking for maintenance, cell replacement, disassembly, assembly, and tool use, there are no automated technologies to capture operational details and identify violations. It relies entirely on manual data entry, photos, and videos for evidence, failing to achieve technological supervision of the process. Summary of the Invention

[0004] The main objective of this invention is to provide a method for monitoring and managing the maintenance of new energy power batteries, aiming to solve the technical problems in the prior art.

[0005] This invention proposes a method for supervising the maintenance operations of new energy power batteries, including: To acquire battery production data, historical maintenance data, and standardized maintenance processes for new energy power batteries, a basic dataset for maintenance supervision is constructed, and a multi-dimensional verification model for maintenance is built based on the basic dataset for maintenance supervision. The system acquires visual recognition data, power battery BMS data, maintenance tool sensor data, and tooling equipment status data during the maintenance process of new energy power batteries, and performs millisecond-level time synchronization and data integration on the visual recognition data, power battery BMS data, maintenance tool sensor data, and tooling equipment status data to obtain integrated multi-source data. The integrated multi-source data is input into the maintenance multi-dimensional verification model to obtain the operation compliance judgment result, and visual keyframes, operation parameter data and early warning control records during the violation judgment process are collected simultaneously to generate multiple encrypted data packets; Based on the operational compliance determination results, a graded early warning signal is generated, and a closed-loop control command is sent to the maintenance tools and equipment to adjust the equipment operating status based on the graded early warning signal. A blockchain hash value is generated based on the encrypted data packet, and the encrypted data packet is associated with and stored with the blockchain hash value to construct an immutable chain of evidence for maintenance operations, thereby achieving deep coupling between maintenance process supervision and traceability system.

[0006] This application also provides a new energy power battery maintenance operation monitoring system, including: The module is used to acquire battery production data, historical maintenance data and standardized maintenance processes of new energy power batteries to build a basic dataset for maintenance supervision, and to build a multi-dimensional verification model for maintenance based on the basic dataset for maintenance supervision. The synchronization and integration module is used to acquire visual recognition data, power battery BMS data, maintenance tool sensor data and tooling equipment status data during the maintenance process of new energy power batteries, and to perform millisecond-level time synchronization and data integration on the visual recognition data, power battery BMS data, maintenance tool sensor data and tooling equipment status data to obtain integrated multi-source data. The input acquisition module is used to input the integrated multi-source data into the maintenance multi-dimensional verification model to obtain the operation compliance judgment result, and simultaneously collect visual keyframes, operation parameter data and early warning control records during the violation judgment process to generate multiple encrypted data packets; An adjustment generation module is used to generate graded early warning signals based on the operation compliance judgment results, and send closed-loop control commands to maintenance tools and tooling equipment based on the graded early warning signals to adjust the equipment operating status. The associated storage module is used to generate a blockchain hash value based on the encrypted data packet, and associate the encrypted data packet with the blockchain hash value for storage, thereby constructing an immutable chain of evidence for maintenance operations to achieve deep coupling between maintenance process supervision and traceability system.

[0007] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described new energy power battery maintenance operation monitoring method.

[0008] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described new energy power battery maintenance operation monitoring method.

[0009] The beneficial effects of this invention are as follows: By constructing a multi-dimensional verification model for maintenance and achieving deep integration of multi-source heterogeneous data with millisecond-level synchronization, this invention maps high-spatiotemporal precision process dynamic data uniformly to a multi-dimensional compliance verification framework of standardized processes. This enables automated, real-time, and objective identification and judgment of core maintenance operation details, significantly improving the accuracy and coverage of process violation detection. By simultaneously extracting visual keyframes and operation parameters to form encrypted data packets at the moment of violation judgment, and driving closed-loop control of equipment through hierarchical early warning, a real-time closed-loop control mechanism of identification, early warning, and forced intervention is constructed. This mechanism can immediately block violations when they occur or are about to occur, greatly reducing safety hazards and quality risks caused by improper operation. Based on the associated storage of encrypted data packets and blockchain hash values, an immutable evidence chain covering process details is formed. This is deeply coupled with a full lifecycle traceability platform, achieving an improvement from traceable results to trustworthy processes and verifiable details, ensuring the integrity of maintenance process data in terms of time, causality, and responsibility. Attached Figure Description

[0010] Figure 1 This is a schematic diagram of a method flow according to an embodiment of the present invention.

[0011] Figure 2 This is a schematic diagram of the system structure according to an embodiment of the present invention.

[0012] Figure 3 This is a schematic diagram of the internal structure of a computer device according to an embodiment of this application.

[0013] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0014] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0015] like Figure 1 As shown, this application provides a method for supervising the maintenance operations of new energy power batteries, including: S1. Obtain battery production data, historical maintenance data, and standardized maintenance processes for new energy power batteries to construct a basic dataset for maintenance supervision, and construct a multi-dimensional verification model for maintenance based on the basic dataset for maintenance supervision. S2. Acquire visual recognition data, power battery BMS data, maintenance tool sensor data, and tooling equipment status data of the new energy power battery during the maintenance process, and perform millisecond-level time synchronization and data integration on the visual recognition data, power battery BMS data, maintenance tool sensor data, and tooling equipment status data to obtain integrated multi-source data. S3. Input the integrated multi-source data into the maintenance multi-dimensional verification model to obtain the operation compliance judgment result, and simultaneously collect visual keyframes, operation parameter data and early warning control records during the violation judgment process to generate multiple encrypted data packets; S4. Generate a graded early warning signal based on the operation compliance judgment result, and send closed-loop control instructions to maintenance tools and tooling equipment based on the graded early warning signal to adjust the equipment operating status. S5. Generate a blockchain hash value based on the encrypted data packet, and associate and store the encrypted data packet with the blockchain hash value to construct an immutable chain of evidence for maintenance operations, so as to achieve deep coupling between maintenance process supervision and traceability system.

[0016] As described in steps S1-S5 above, the steps of obtaining battery production data, historical maintenance data and standardized maintenance process of new energy power battery are as follows: obtain the unique code and corresponding model parameters of new energy power battery, and retrieve battery production data, historical maintenance data and standardized maintenance process from the full life cycle traceability platform according to the unique code and model parameters. The steps for acquiring visual recognition data, battery BMS data, repair tool sensor data, and tooling status data during the maintenance of new energy power batteries are as follows: Real-time image streams and motion video streams of the maintenance operation scene are acquired through distributed visual acquisition devices. Visual recognition data is obtained by parsing image frames, including personnel hand motion coordinates, tool posture data, battery appearance images, and assembly alignment images. Battery BMS data is obtained by reading core battery parameters in real time through the battery BMS interface, including cell voltage, total voltage, real-time temperature, SOC value, SOH value, and charge / discharge cycle status. Repair tool sensor data is obtained by collecting tool operation data through the built-in sensor modules of the repair tools, including tool output torque, speed, impact force, operation time, and operating current. Tooling status data is obtained by collecting operating parameters through the tooling status monitoring unit, including tooling positioning accuracy, clamping force, operating noise, and fault alarm information. This invention organically integrates production batch characteristic parameters, initial cell consistency indicators, historical maintenance damage patterns, and standard operating sequences into a structured multi-dimensional dataset. Based on this dataset, a verification model incorporating multi-dimensional constraints such as spatial, temporal, and parameter thresholds is trained / constructed. Essentially, this creates a personalized and dynamic digital twin benchmark for operational compliance tailored to specific maintenance objects. By employing millisecond-level cross-sensor timestamp alignment, unified mapping of heterogeneous data protocols, and spatiotemporal fusion techniques, it achieves true time-aligned multimodal data streams of visual data, BMS data, sensor data, and equipment status. This overcomes the problems of lost process details and difficulties in causal tracing caused by data asynchrony and modal silos in existing technologies, providing a solid foundation for subsequent model input. It provides a complete and computable digital image of the process, significantly improving the spatiotemporal accuracy and reliability of evidence in process violation identification. By inputting multi-source spatiotemporally aligned data into a pre-constructed multi-dimensional verification model, it achieves real-time quantitative comparison and violation pattern matching of multi-dimensional features such as operation sequence, action posture, tool trajectory, mechanical parameters, and BMS abnormal response. In the judgment link, it automatically extracts visual keyframes (i.e., high information density images at the moment of violation), corresponding multi-source parameter slices, and intermediate inference records of the model to form structured and verifiable evidence units, which are then encapsulated in the form of encrypted packages. This realizes the transformation from manual post-event corroboration to automatic evidence generation and solidification by the system, solving the core technical problems of missing, unreliable, and incomplete process evidence in existing traceability technologies. By generating multi-level early warning signals based on the quantitative compliance classification of the model output, and realizing active closed-loop control of actuators such as power tool speed, torque, tool clamping force, and power supply on / off through the equipment-side protocol interface, and by converting the process compliance judgment results into an executable control command chain, a complete closed loop of perception, judgment, and intervention is constructed. This significantly reduces technical safety issues such as secondary damage, thermal runaway risk, and cell consistency damage caused by violations, and achieves a fundamental improvement in the maintenance process from passive recording to active control. By calculating irreversible hashes for each violation and key evidence package and anchoring them on the chain, while maintaining a strong correlation between the original encrypted package and the on-chain hash, a time-ordered, tamper-proof, distributed consensus maintenance process evidence chain is formed, providing a complete process proof that is technically irrefutable. This invention achieves full-process automated supervision of the core operation links of power battery maintenance through the organic coupling of personalized multi-dimensional verification model construction, millisecond-level multi-source spatiotemporal fusion, automated process compliance judgment and automatic evidence generation, real-time closed-loop physical intervention, and blockchain-level tamper-proof evidence chain solidification.

[0017] In one embodiment, step S1, which involves constructing a multi-dimensional verification model for maintenance based on the maintenance supervision dataset, includes: S11. Obtain the battery model, cell parameters and factory inspection threshold based on the battery production data, and obtain the fault type, corresponding repair plan, compliance judgment case and violation judgment case based on the historical repair data. S12. Based on the standardized maintenance process, obtain multiple key operation steps, step parameter ranges and safety specifications, and establish a battery attribute-maintenance requirement mapping relationship in combination with battery model and cell parameters; S13. Based on the compliance judgment cases and violation judgment cases, the historical maintenance data is labeled to obtain the labeled dataset, and the verification judgment benchmark is determined by combining the factory inspection threshold and the step parameter range. S14. Construct a pre-defined multi-dimensional verification framework for operation process compliance, parameter threshold matching, and fault repair adaptation. Configure weight coefficients for each dimension in the pre-defined multi-dimensional verification framework according to the battery attribute-repair requirement mapping relationship and integrate a fusion rule engine and a lightweight machine learning model to obtain a multi-dimensional hybrid architecture. S15. Input the labeled dataset into the multi-dimensional hybrid architecture, and train the model parameters according to the verification judgment benchmark to obtain the maintenance multi-dimensional verification model.

[0018] As described in steps S11-S15 above, this invention extracts battery model, key cell parameters (such as capacity, internal resistance, positive and negative electrode material systems, design voltage windows, etc.) and factory inspection thresholds from battery production data in a structured manner. It then correlates and matches these parameters with historical repair data, including the types of faults that occurred, the actual repair solutions used, and specific cases determined to be compliant or non-compliant. This forms the first layer of objective mapping between individual battery characteristics and historical repair behavior patterns, breaking through the limitations of coarse-grained data in existing traceability technologies. It provides a priori knowledge benchmark sensitive to individual battery differences for subsequent dimensional verification, through key operational steps extracted based on standardized repair processes. The system establishes a dynamic maintenance constraint knowledge base driven by battery attributes by defining the sequence, allowable parameter ranges for each step (such as torque range, temperature range, voltage measurement timing, protection level requirements, etc.), and mandatory safety specifications. It also establishes an explicit many-to-many mapping relationship between these specifications and battery models and cell parameters. This transforms the unstructured maintenance requirements that were originally scattered in standard documents, manuals, and specifications into a set of calculable and matchable structured constraint rules. Furthermore, it enables the system to adaptively adjust constraint boundaries for individual battery attributes, giving it battery-specific maintenance compliance templates. This significantly improves the system's ability to objectively judge core operations such as unpacking depth, cell selection and matching, and disassembly and assembly process parameters. By constructing a high-quality labeled dataset through supervised annotation of historical maintenance records using confirmed compliant or non-compliant cases, and by fusing factory inspection thresholds with key step parameter ranges to form a composite judgment benchmark with multi-source cross-validation, a gold standard judgment threshold that can be directly used for model training and inference is generated. This overcomes the fundamental defects of existing technologies, such as highly subjective judgment criteria, difficulty in reproduction, and inability to be applied at scale. It realizes the transformation from subjective judgment to data-driven and rule-based judgment, providing highly reliable and traceable correct labels for subsequent automated verification, and significantly improving the objectivity, consistency, and auditability of process violation identification. By constructing a multi-dimensional verification framework covering three orthogonal dimensions—operational process compliance, parameter threshold matching degree, and fault-repair solution adaptability—and scientifically configuring weights for each dimension based on the battery attribute-maintenance requirement mapping relationship, the rule engine and lightweight machine learning model are finally integrated at the architectural level to form a multi-dimensional hybrid verification architecture. This achieves a multi-dimensional, orthogonal decomposition, and heterogeneous fusion verification architecture for the power battery maintenance process, overcoming the dual defects of excessive rigidity of a single rule system and the black box and poor generalization of pure machine learning models. By dynamically configuring weights to reflect individual battery differences, and by combining rules with lightweight machine learning fusion to ensure interpretability and adaptive learning capabilities, a technical judgment center with multi-perspective and confidence-integrated process details, which is completely absent in existing traceability platforms, has been formed. This provides architectural support for truly achieving automated process supervision without human intervention. Through high-quality labeled data, domain-prior weights, and closed-loop training with a rule-learning dual-driven architecture, the model possesses both traceability to specific dimensions and rules and strong generalization capabilities (such as the ability to identify violations that are not explicitly listed but have similar patterns). This solves the core technical problem of existing traceability technologies having zero automated process judgment capabilities, enabling the system to perform machine-readable, objective, and real-time compliance quantification scoring and violation warnings for the process details of key actions such as unpacking, cell replacement, disassembly and assembly, and tool use for the first time. The above steps, from individualized data and structured knowledge extraction, objective judgment benchmark construction, multi-dimensional heterogeneous fusion architecture design, and domain knowledge-driven model training closed loop, jointly construct a multi-dimensional intelligent verification closed loop system driven by individual battery attributes and focused on process operation details. This achieves automated, objective, and interpretable technical supervision of core operation links in the maintenance process, significantly eliminating the strong reliance on manual input and unstructured evidence, and greatly improving the real-time capture rate of violations, the ability to control false alarms and omissions, and the traceability of the entire chain.

[0019] In one embodiment, step S2, which involves performing millisecond-level time synchronization and data integration on the visual recognition data, power battery BMS data, maintenance tool sensor data, and tooling equipment status data to obtain integrated multi-source data, includes: S21. Acquire the acquisition features and offset features corresponding to the visual recognition data, power battery BMS data, maintenance tool sensor data and tooling equipment status data respectively. The acquisition features include the acquisition timestamp and the original acquisition frequency, and the offset features include the transmission delay value and the timestamp drift deviation. S22. Construct a unified millisecond-level time reference axis based on multiple acquisition timestamps and the original acquisition frequency, and obtain the corresponding time calibration coefficient based on each transmission delay value and timestamp drift deviation; S23. Based on the unified millisecond-level time reference axis and each time calibration coefficient, calibrate and align each acquisition timestamp of the visual recognition data, power battery BMS data, maintenance tool sensor data and tooling equipment status data to obtain time-synchronized single-source data. S24. Obtain the data type, data range, and data format of each time-synchronized single-source data, and determine the standardized mapping rules based on the data type and data range, and formulate a unified data coding standard based on the data format; S25. Perform range normalization processing on the corresponding time-synchronized single-source data according to each of the standardized mapping rules to obtain normalized single-source data, and perform format uniform conversion on the corresponding normalized single-source data according to each of the unified data encoding specifications to obtain standardized single-source data. S26. Obtain the abnormal data judgment threshold in each of the standardized single-source data, and remove invalid abnormal values ​​in the corresponding standardized single-source data according to each judgment threshold to obtain valid single-source data; S27. Based on a unified millisecond-level time reference axis, perform correlation matching on multiple valid single-source data to obtain multiple associated multi-source data, and then integrate the multiple associated multi-source data in an orderly manner according to the time series to obtain integrated multi-source data.

[0020] As described in steps S21-S27 above, this invention, by simultaneously extracting timestamps, original sampling frequencies, transmission delays, and timestamp drift deviations from four typical heterogeneous data sources—visual recognition data, BMS data, maintenance tool sensor data, and tooling equipment status data—decouples the inherent attributes of the acquisition end from the dynamic damage of the transmission link and explicitly models them. This provides a complete and quantified basis for deviation compensation for subsequent millisecond-level alignment, avoiding sub-second to second-level time misalignments caused by different sensors, buses, acquisition clock sources, different network hop counts, and protocol stack delays. This gives a comparable temporal semantic basis to various process operation signals that were originally severely discrete in physical time and could not be directly correlated, solving the fundamental deficiency of existing traceability technologies that lack a time reference for maintenance process operation details. This technology, based on the construction of a unified millisecond-level time reference axis using multi-source acquisition timestamp sequences and their respective original sampling frequencies, breaks through the limitations of using a single master device clock or external time source as the absolute reference in traditional industrial scenarios. It achieves a relatively consistent time skeleton formed by collective negotiation of the historical sampling patterns of all participating data streams. At the same time, it independently calculates the transmission delay value and drift deviation for each data link and generates dedicated time calibration coefficients, avoiding the over-compensation or under-compensation problem of a certain type of data caused by using a unified calibration model. By establishing an adaptive time reference and personalized compensation system that adapts to heterogeneous acquisition cycles and link nondeterministic delays, it significantly improves the alignment accuracy of multi-source data at the millisecond scale, laying a reliable time coordinate system for subsequent causal correlation analysis of process actions. By remapping and aligning the timestamps of each data record one by one using a unified millisecond-level time reference axis and dedicated calibration coefficients for each source, multiple signals that were originally misaligned by hundreds of milliseconds can be accurately placed into the same millisecond-level time window. This enables the simultaneous observation of a multi-dimensional chain of evidence for the same instant of operation, overcoming the technical problem of existing technologies being unable to automatically determine the details of maintenance processes due to the inability to correlate them in time. By extracting the data type, physical range, and original encoding format of the single-source data after time synchronization, and formulating standardized mapping rules and unified data encoding specifications in a two-stage standardization path, the problems of dimensional conflicts, numerical scale differences that overwhelm features, and format incompatibility that prevent unified storage and analysis caused by simply splicing the original heterogeneous data are avoided. This achieves the unification of numerical space and semantic space while preserving the physical meaning of each source, enabling subsequent anomaly removal, temporal correlation, and rule judgment to be performed in the same feature space. By establishing an anomaly cleaning mechanism that is adaptive to data sources and semantically aware of scenarios, the high fidelity and representativeness of the data entering subsequent correlation stages are ensured. This avoids misjudgment of the entire time series due to glitches or packet loss in individual channels, providing a clean and reliable process snapshot sequence for subsequent identification of violations based on multi-source evidence chains. This significantly improves the false alarm and false alarm control capabilities of automated supervision. By performing precise correlation matching and ordered temporal integration based on the time reference axis on effective multi-source data, multi-dimensional process time series multi-source fusion data is formed. This enables previously isolated information such as actions seen visually, forces applied by tools, voltage, current, temperature transients observed by the BMS, clamping of tooling, and positioning status to form a mutually corroborating evidence chain in the millisecond time dimension. Thus, at the technical level, it achieves automated, objective, and machine-readable complete reconstruction and representation of the specific process details of core operation links such as unpacking, cell replacement, disassembly and assembly, and tool use. The above steps, through a complete closed-loop path of heterogeneous deviation modeling, adaptive millisecond-level time benchmark construction, record-by-record precise alignment, dual standardization of physical meaning and format, adaptive anomaly cleaning, and time-driven integration of multi-source evidence chains, achieve millisecond-level temporal semantic unification and high-fidelity fusion of multi-source heterogeneous process data from vision, electrical, mechanical, and state perspectives. This enables every critical operation moment in the maintenance process to have a multi-channel, objective, traceable, and computable digital image, providing a solid technical foundation for subsequent rule-based and model-based automated identification of violations, real-time process quality assessment, and precise tracing of responsibility.

[0021] In one embodiment, step S3, which involves inputting the integrated multi-source data into the maintenance multi-dimensional verification model to obtain the operational compliance determination result, includes: S31. Extract operation process features, parameter threshold features and fault adaptation features from the integrated multi-source data, and prioritize the operation process features, parameter threshold features and fault adaptation features according to the dimension weight coefficients of the maintenance multi-dimensional verification model to obtain a weighted feature set. S32. Input the weighted feature set into the rule engine module of the maintenance multi-dimensional verification model to obtain the threshold determination result; S33. Input the weighted feature set into the lightweight machine learning module of the maintenance multi-dimensional verification model to obtain the correlation analysis results, and obtain the comprehensive compliance score based on the correlation analysis results and the threshold judgment results; S34. Determine whether the overall compliance score is less than a preset compliance threshold; If the overall compliance score is not less than the preset compliance threshold, the operation is deemed compliant. If the overall compliance score is determined to be less than the preset compliance threshold, it is determined to be an operational violation.

[0022] As described in steps S31-S34 above, the operation compliance judgment result includes operation compliance and operation violation; the threshold judgment result refers to whether the threshold characteristics of the judgment parameter meet the preset range of the standardized maintenance process; the correlation analysis result refers to the analysis of the temporal rationality of the operation process characteristics and the matching degree of the fault adaptation characteristics; the operation process characteristics reflect the behavioral trajectory of operation sequence, timing, integrity, etc.; the parameter threshold characteristics reflect the real-time deviation of key process parameters; and the fault adaptation characteristics reflect whether the operation under the current fault mode matches the preset safety and standard path. This invention extracts operational process features, parameter threshold features, and fault adaptation features from integrated multi-source process data. It then introduces pre-trained or calibrated dimensional weight coefficients from a multi-dimensional maintenance verification model to prioritize and weight these three heterogeneous features, resulting in a unified weighted feature set. The model's intrinsic dimensional weight coefficients enable adaptive dynamic adjustment of feature importance across different scenarios, avoiding the failure or poor generalization ability of fixed thresholds or manual experience rules in complex maintenance scenarios. The formation of this weighted feature set allows subsequent decision-making modules to base their decisions on a unified and prioritized quantization. The representation-based decision-making significantly reduces misjudgments / missed judgments caused by feature dimension conflicts, information redundancy, or imbalance of importance. The rule engine performs deterministic, transparent, and traceable threshold logic judgments on weighted features, overcoming the fundamental defect of existing traceability platforms that can only process discrete result data before and after maintenance and cannot process continuous process feature sequences. Since the input is weighted features, the rule conditions can directly focus on high-priority violation patterns, such as serious exceedance of key parameters, missing core operations, and serious time sequence disorder, which significantly improves the detection rate and real-time performance of hard violations. By applying lightweight machine learning to the complex correlation mining of multi-source heterogeneous features in the maintenance process, implicit violation combination patterns that are difficult for rule engines to express were captured, such as parameters that do not exceed the threshold but have abnormal timing, and subtle deviations such as mismatch between operation paths and faults. This significantly improves the ability to identify hidden and new violations. By simultaneously driving the rule engine and the lightweight machine learning module with the same batch of weighted features, complementary multi-paradigm fusion judgment was achieved, overcoming two types of technical problems: incomplete coverage of a single rule system and poor interpretability and sensitivity to small samples of a single model. The continuous quantitative output of the comprehensive compliance score is more refined than discrete level judgment, and can support the quantitative basis for risk classification and early warning, flexible setting of in-process intervention thresholds, and post-event responsibility tracing. The above steps form a structured extraction and adaptive feature extraction from multi-source data and multi-dimensional features of the process. This fully automated compliance verification technology chain, encompassing weighted averages, rule-based hard judgment, lightweight learning-based latent association mining, multi-paradigm fusion continuous scoring, and threshold decision-making, enables automated, objective, and real-time capture and compliance assessment of core maintenance operational details. It eliminates technical blind spots in process supervision. Through a progressive, multi-paradigm collaborative judgment framework using feature adaptive weighting, rule engine hard constraints, lightweight machine learning modules for soft association, and fusion scoring, it simultaneously addresses multiple requirements such as high interpretability, high detection rate, sensitivity to concealed and novel violations, real-time performance, and edge deployability. This overcomes the limitations of single-technology approaches, providing quantifiable, traceable, and auditable technical compliance judgment criteria for the entire power battery maintenance process, significantly improving the objectivity, consistency, timeliness, and closed-loop management capabilities of supervision.

[0023] In one embodiment, step S4, which sends closed-loop control commands to maintenance tools and equipment to adjust the equipment operating status based on the graded early warning signal, includes: S41. Obtain a preset control instruction library and establish a signal-instruction mapping relationship based on the graded early warning signals and the preset control instruction library. The signal-instruction mapping relationship includes a first-level early warning signal corresponding to a tool parameter prompt adjustment instruction, a second-level early warning signal corresponding to a tooling equipment speed reduction operation instruction, and a third-level early warning signal corresponding to a maintenance tool shutdown instruction and a tooling equipment locking instruction. S42. Select the corresponding control command according to the graded early warning signal, and send the control command to the control module of the maintenance tool and tooling equipment through the industrial bus; S43. Collect instruction execution feedback data of maintenance tools and equipment in real time, and determine whether the maintenance tools and equipment adjust their operating status according to the control instructions based on the instruction execution feedback data; S44. If the maintenance tools and equipment fail to complete the instruction execution, the control instruction is resent and the instruction priority is increased until the status of the maintenance tools and equipment meets the control requirements, thus forming a closed-loop control.

[0024] As described in steps S41-S44 above, this invention achieves tiered matching between the severity of warnings and the physical control strength of equipment by pre-constructing a signal-command mapping relationship that strictly corresponds to multi-level warning signals and specific control commands. This enables the system to automatically trigger technical interventions of varying intensities at different stages of violation risk, rather than merely remaining at the level of information recording or alarm prompts. By using an industrial bus as the channel for issuing control commands, commands are directly sent to the control modules of maintenance tools and equipment, achieving real-time, reliable, and deterministic communication between the warning signals and the underlying controller of the equipment. Real-time collection of command execution feedback data, such as actual tool speed, torque feedback, actual operating mode status of the tooling, actuator positioning signals, current, and voltage feedback, is also achieved. This feedback data determines whether the equipment has truly completed the status adjustment according to the issued instructions, realizing a single closed loop of control instruction issuance, equipment physical response, and status verification. When the single instruction execution fails to achieve the expected state, the system automatically reissues the control instruction and gradually increases the instruction priority, such as from ordinary priority to the highest preemptive priority or from prompt-type instructions to forced shutdown / lock-type instructions, until the equipment status meets the control requirements. This forms a dynamic closed-loop control strategy of multiple retries + priority increment, which enables the system to force the equipment to return to a compliant state even in the event of human resistance on site by continuously strengthening the control intensity. This greatly enhances the robustness and anti-interference capability of the monitoring system and achieves strong closed-loop, strong intervention, and strong convergence for violations in the maintenance process.

[0025] In one embodiment, step S5, which generates a blockchain hash value based on the encrypted data packet and associates and stores the encrypted data packet with the blockchain hash value to construct an immutable chain of evidence for maintenance operations, includes: S51. Obtain the violation event identifier, visual keyframe summary, core data of operation parameters, and early warning control record timestamp based on the encrypted data packet; S52. Obtain the maintenance supervision safety level of new energy power batteries to determine the appropriate hash algorithm; S53. Perform a one-way hash operation on the violation event identifier, visual keyframe summary, core data of operation parameters and timestamp according to the adaptive hash algorithm to generate a unique corresponding blockchain hash value. S54. Extract the digital signature and data integrity verification code of the encrypted data packet; S55. Establish an index mapping relationship between encrypted data packets and blockchain hash values ​​based on the violation event identifier, and encapsulate the encrypted data packets, blockchain hash values, digital signatures and data integrity verification codes to obtain associated data units. S56. Obtain maintenance supervision and traceability requirements, write related data units into the consortium blockchain node according to the maintenance supervision and traceability requirements, and synchronously record the node writing time, supervision node signature and data access permission identifier to realize distributed storage of related data. S57. Based on the immutability of the consortium blockchain, sort multiple related data units according to the time sequence of the violation events and integrate them to form a multi-level evidence chain structure. S58. The integrity of the encrypted data packet is verified by reverse verification of the blockchain hash value. If the verification passes, the evidence chain is confirmed to be valid. Finally, an immutable and traceable maintenance operation evidence chain is constructed, realizing the binding of maintenance process data with the traceability system.

[0026] As described in steps S51-S58 above, the multi-level evidence chain structure includes data originality verification (hash value), data integrity verification (check code), and operation tracing basis (core data). This invention extracts four highly refined and complementary types of process feature data from encrypted data packets: violation event identifiers, visual keyframe summaries, core operational parameter data, and early warning control record timestamps. These data serve as the input for subsequent hash calculations, enabling automated, standardized, and machine-readable representation of dynamic, high-risk operational details during maintenance processes such as unpacking, disassembly, cell replacement, assembly, and tool use. This avoids the shortcomings of relying on subjective human descriptions, scattered photos, or complete videos, providing an objective and computable technical data source for process-level supervision. It significantly improves the automation and reliability of violation identification and evidence preservation. By dynamically selecting and adapting hash algorithms according to the specific maintenance and supervision safety level of new energy power batteries, the limitations of using a unified and static hash method in existing technologies are overcome. By introducing a safety level-algorithm adaptation mechanism, hash algorithms with higher collision resistance, stronger resistance to quantum risks, or higher computational safety margins are used for high-safety-level maintenance processes, while algorithms with higher computational efficiency are used for low-risk scenarios. This achieves a targeted and adaptive technical balance among safety, computational overhead, and anti-attack capabilities, and solves the technical problems of hash collision risk or performance issues in existing traceability systems under high-risk maintenance scenarios for power batteries. By using a specific combination of violation event identifiers, visual keyframe summaries, core operational parameter data, and timestamps as the complete input for a one-way hash operation, a unique blockchain hash value highly bound to the maintenance process is generated. This ensures that even minor alterations to key details of the maintenance process will result in a completely different hash value, thus achieving strong integrity protection for unobservable details of the maintenance process at the technical level. This significantly enhances the process-level anti-tampering capabilities lacking in existing technologies. While generating the blockchain hash, the digital signature and data integrity verification code extracted from the original encrypted data packet are retained and associated, forming a multi-layered protection chain. This extends the end-to-end integrity chain to the moment before blockchain storage, preventing data from being tampered with before entering distributed storage without being traceable. By establishing an index mapping relationship between encrypted data packets and blockchain hash values ​​with violation event identifiers as the core, and encapsulating the data packet body, hash, signature, and verification code into associated data units, a logically unified and physically separated evidence organization method is achieved. By establishing fast-retrieval evidence atomic units at the granularity of violation events, regulators can efficiently locate the complete evidence set of a specific maintenance violation process in massive distributed data, providing a structured and scalable technical foundation for rapid response and accurate accountability. Based on the needs of maintenance supervision and traceability, relevant data units are selectively written into the consortium blockchain, and the node writing time, supervisory node signature, and data access permission identifier are recorded simultaneously. This achieves a distributed storage mechanism with on-demand on-chain data upload and fine-grained access control, avoiding the problems of irrelevant data occupying on-chain resources and excessive exposure of sensitive process details. At the same time, the verifiability of public power intervention is strengthened through supervisory node signatures, providing a credible and controllable distributed evidence storage foundation for multi-party collaborative supervision scenarios. By utilizing the inherent immutability and temporal characteristics of the consortium blockchain, multiple relevant data units are automatically sorted and integrated into multi-level evidence chains according to the time sequence of violation events, forming a time-ordered and logically progressive process evidence chain structure. This automatically constructs a chain evidence structure that reflects the complete timeline and causal relationship of maintenance operations from originally discrete single violation records, solving the technical defects of existing technologies that can only provide isolated result snapshots and cannot reconstruct the evolution path of violations. This provides a basis for accident cause analysis and liability ratio determination. This provides objective and complete temporal technical support for complex regulatory scenarios such as classification and continuous violation identification. It uses blockchain hash values ​​to reverse verify the integrity of the original encrypted data packets, confirming the validity of the entire evidence chain only when the verification passes. This establishes a closed-loop self-verification mechanism. Through close technical coupling with the previously described steps, the validity of the evidence chain no longer depends on any centralized institution's declaration, but is jointly guaranteed by cryptography and distributed consensus. This achieves a deep and irreversible binding of maintenance process data and the full lifecycle traceability system at the technical trust level. The above steps form a complete technical closed loop, from automated extraction of process details, adaptive security level protection, process fingerprint cryptographic solidification, multi-layered credential encapsulation, event granularity indexing, on-demand and controllable on-chaining, construction of temporal evidence chains, and closed-loop hash verification. This enables automated capture, objective representation, strong tamper-proofing, temporal restoration, and cryptographic-level verification of core maintenance operations, significantly improving the accuracy of violation warnings.

[0027] like Figure 2 As shown, this application also provides a new energy power battery maintenance operation monitoring system, including: The module is used to acquire battery production data, historical maintenance data and standardized maintenance processes of new energy power batteries to build a basic dataset for maintenance supervision, and to build a multi-dimensional verification model for maintenance based on the basic dataset for maintenance supervision. The synchronization and integration module is used to acquire visual recognition data, power battery BMS data, maintenance tool sensor data and tooling equipment status data during the maintenance process of new energy power batteries, and to perform millisecond-level time synchronization and data integration on the visual recognition data, power battery BMS data, maintenance tool sensor data and tooling equipment status data to obtain integrated multi-source data. The input acquisition module is used to input the integrated multi-source data into the maintenance multi-dimensional verification model to obtain the operation compliance judgment result, and simultaneously collect visual keyframes, operation parameter data and early warning control records during the violation judgment process to generate multiple encrypted data packets; An adjustment generation module is used to generate graded early warning signals based on the operation compliance judgment results, and send closed-loop control commands to maintenance tools and tooling equipment based on the graded early warning signals to adjust the equipment operating status. The associated storage module is used to generate a blockchain hash value based on the encrypted data packet, and associate the encrypted data packet with the blockchain hash value for storage, thereby constructing an immutable chain of evidence for maintenance operations to achieve deep coupling between maintenance process supervision and traceability system.

[0028] In one embodiment, the adjustment generation module includes: A setup unit is used to acquire a preset control instruction library and establish a signal-instruction mapping relationship based on the graded early warning signals and the preset control instruction library. The signal-instruction mapping relationship includes a first-level early warning signal corresponding to a tool parameter prompt adjustment instruction, a second-level early warning signal corresponding to a tooling equipment speed reduction operation instruction, and a third-level early warning signal corresponding to a maintenance tool shutdown instruction and a tooling equipment locking instruction. The sending unit is used to select the corresponding control command according to the graded early warning signal, and send the control command to the control module of the maintenance tool and tooling equipment through the industrial bus. The data acquisition unit is used to acquire real-time instruction execution feedback data of maintenance tools and tooling equipment, and to determine whether the maintenance tools and tooling equipment have adjusted their operating status according to the control instructions based on the instruction execution feedback data; The adjustment unit is used to resend control commands and increase command priority if the maintenance tools and equipment have not completed the execution of the instructions, until the status of the maintenance tools and equipment meets the control requirements, thus forming a closed-loop control.

[0029] It should be noted that each module and unit in the new energy power battery maintenance operation supervision system corresponds one-to-one with the steps in the new energy power battery maintenance operation supervision method.

[0030] like Figure 3 As shown, this application also provides a computer device, which can be a server, and its internal structure can be as follows: Figure 3As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores all data required for the process of monitoring and managing the maintenance operations of new energy power batteries. The network interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the monitoring and management methods for the maintenance operations of new energy power batteries.

[0031] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer equipment on which the present application is applied.

[0032] An embodiment of this application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements any one of the above-described new energy power battery maintenance operation monitoring methods.

[0033] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in this application and in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0034] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0035] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for supervising the maintenance and operation of new energy power batteries, characterized in that, include: To acquire battery production data, historical maintenance data, and standardized maintenance processes for new energy power batteries, a basic dataset for maintenance supervision is constructed, and a multi-dimensional verification model for maintenance is built based on the basic dataset for maintenance supervision. The system acquires visual recognition data, power battery BMS data, maintenance tool sensor data, and tooling equipment status data during the maintenance process of new energy power batteries, and performs millisecond-level time synchronization and data integration on the visual recognition data, power battery BMS data, maintenance tool sensor data, and tooling equipment status data to obtain integrated multi-source data. The integrated multi-source data is input into the maintenance multi-dimensional verification model to obtain the operation compliance judgment result, and visual keyframes, operation parameter data and early warning control records during the violation judgment process are collected simultaneously to generate multiple encrypted data packets; Based on the operational compliance determination results, a graded early warning signal is generated, and a closed-loop control command is sent to the maintenance tools and equipment to adjust the equipment operating status based on the graded early warning signal. A blockchain hash value is generated based on the encrypted data packet, and the encrypted data packet is associated with and stored with the blockchain hash value to construct an immutable chain of evidence for maintenance operations, thereby achieving deep coupling between maintenance process supervision and traceability system.

2. The method for supervising the maintenance and operation of new energy power batteries according to claim 1, characterized in that, The step of constructing a multi-dimensional verification model for maintenance based on the maintenance supervision basic dataset includes: Based on the battery production data, obtain the battery model, cell parameters, and factory testing thresholds; based on the historical repair data, obtain compliance judgment cases and violation judgment cases. Based on the standardized maintenance process, multiple key operation steps, step parameter ranges, and safety specifications are obtained, and a battery attribute-maintenance requirement mapping relationship is established by combining the battery model and cell parameters. Based on the aforementioned compliance and violation cases, historical maintenance data are annotated, and verification criteria are determined by combining factory inspection thresholds and step parameter ranges. Based on the battery attribute-maintenance requirement mapping relationship, weight coefficients are configured for each dimension in the preset multi-dimensional verification framework, and a multi-dimensional hybrid architecture is obtained by integrating a fusion rule engine and a lightweight machine learning model. The labeled dataset is input into a multi-dimensional hybrid architecture, and the model parameters are trained according to the verification judgment benchmark to obtain a maintenance multi-dimensional verification model.

3. The method for supervising the maintenance and operation of new energy power batteries according to claim 1, characterized in that, The step of performing millisecond-level time synchronization and data integration on the visual recognition data, power battery BMS data, maintenance tool sensor data, and tooling equipment status data to obtain integrated multi-source data includes: Acquire the acquisition features and offset features corresponding to the visual recognition data, power battery BMS data, maintenance tool sensor data, and tooling equipment status data respectively, and obtain a unified millisecond-level time reference axis and time calibration coefficient according to the acquisition features and offset features respectively; Based on the unified millisecond-level time reference axis and time calibration coefficient, the visual recognition data, power battery BMS data, maintenance tool sensor data and tooling equipment status data are calibrated and aligned to obtain time-synchronized single-source data. The data type, data range, and data format of each time-synchronized single-source data are obtained to determine the standardized mapping rules and unified data coding specifications. The time-synchronized single-source data are then normalized and transformed according to the standardized mapping rules and unified data coding specifications to obtain standardized single-source data. Multiple standardized single-source data are correlated and matched based on a unified millisecond-level time reference axis to obtain multiple correlated multi-source data. The multiple correlated multi-source data are then integrated in an orderly manner according to the time series to obtain integrated multi-source data.

4. The method for supervising the maintenance and operation of new energy power batteries according to claim 1, characterized in that, The step of inputting the integrated multi-source data into the maintenance multi-dimensional verification model to obtain the operation compliance judgment result includes: Based on the integrated multi-source data, operation process features, parameter threshold features, and fault adaptation features are extracted, and the operation process features, parameter threshold features, and fault adaptation features are prioritized and quantified according to the dimension weight coefficients of the maintenance multi-dimensional verification model to obtain a weighted feature set. The weighted feature set is input into the rule engine module of the maintenance multi-dimensional verification model to obtain the threshold determination result; The weighted feature set is input into the lightweight machine learning module of the maintenance multi-dimensional verification model to obtain the correlation analysis results, and the comprehensive compliance score is obtained based on the correlation analysis results and the threshold judgment results. Determine whether the overall compliance score is less than a preset compliance threshold; If the overall compliance score is not less than the preset compliance threshold, the operation is deemed compliant. If the overall compliance score is determined to be less than the preset compliance threshold, it is determined to be an operational violation.

5. The method for supervising the maintenance and operation of new energy power batteries according to claim 1, characterized in that, The step of sending closed-loop control commands to maintenance tools and equipment to adjust the equipment operating status based on the graded early warning signals includes: Obtain a preset control command library and establish a signal-command mapping relationship based on the graded early warning signals and the preset control command library. The signal-command mapping relationship includes a first-level early warning signal corresponding to a tool parameter prompt adjustment command, a second-level early warning signal corresponding to a tooling equipment speed reduction command, and a third-level early warning signal corresponding to a maintenance tool shutdown command and a tooling equipment locking command. Based on the graded early warning signal, the corresponding control command is selected, and the control command is sent to the control module of the maintenance tool and tooling equipment through the industrial bus; Real-time acquisition of instruction execution feedback data from maintenance tools and equipment, and determination of whether maintenance tools and equipment have adjusted their operating status according to control instructions based on the instruction execution feedback data; If the maintenance tools and equipment fail to complete the instruction execution, the control instruction is resent and the instruction priority is increased until the status of the maintenance tools and equipment meets the control requirements, thus forming a closed-loop control.

6. The method for supervising the maintenance and operation of new energy power batteries according to claim 1, characterized in that, The step of generating a blockchain hash value based on the encrypted data packet, and associating and storing the encrypted data packet with the blockchain hash value to construct an immutable chain of evidence for the maintenance operation includes: Based on the encrypted data packet, obtain the violation event identifier, visual keyframe summary, core data of operation parameters, and early warning control record timestamp; Obtain the maintenance and regulatory safety level of new energy power batteries to determine the appropriate hash algorithm; Based on the adaptive hash algorithm, a one-way hash operation is performed on the violation event identifier, visual keyframe summary, core data of operation parameters and timestamp to generate a unique corresponding blockchain hash value. Extract the digital signature and data integrity check code from the encrypted data packet; Based on the violation event identifier, an index mapping relationship between encrypted data packets and blockchain hash values ​​is established. The encrypted data packets, blockchain hash values, digital signatures, and data integrity verification codes are associated, encapsulated, and stored to construct an immutable chain of evidence for maintenance operations.

7. A new energy power battery maintenance operation monitoring system, characterized in that, include: The module is used to acquire battery production data, historical maintenance data and standardized maintenance processes of new energy power batteries to build a basic dataset for maintenance supervision, and to build a multi-dimensional verification model for maintenance based on the basic dataset for maintenance supervision. The synchronization and integration module is used to acquire visual recognition data, power battery BMS data, maintenance tool sensor data and tooling equipment status data during the maintenance process of new energy power batteries, and to perform millisecond-level time synchronization and data integration on the visual recognition data, power battery BMS data, maintenance tool sensor data and tooling equipment status data to obtain integrated multi-source data. The input acquisition module is used to input the integrated multi-source data into the maintenance multi-dimensional verification model to obtain the operation compliance judgment result, and simultaneously collect visual keyframes, operation parameter data and early warning control records during the violation judgment process to generate multiple encrypted data packets; An adjustment generation module is used to generate graded early warning signals based on the operation compliance judgment results, and send closed-loop control commands to maintenance tools and tooling equipment based on the graded early warning signals to adjust the equipment operating status. The associated storage module is used to generate a blockchain hash value based on the encrypted data packet, and associate the encrypted data packet with the blockchain hash value for storage, thereby constructing an immutable chain of evidence for maintenance operations to achieve deep coupling between maintenance process supervision and traceability system.

8. The new energy power battery maintenance operation monitoring system according to claim 7, characterized in that, The generation and adjustment module includes: A setup unit is used to acquire a preset control instruction library and establish a signal-instruction mapping relationship based on the graded early warning signals and the preset control instruction library. The signal-instruction mapping relationship includes a first-level early warning signal corresponding to a tool parameter prompt adjustment instruction, a second-level early warning signal corresponding to a tooling equipment speed reduction operation instruction, and a third-level early warning signal corresponding to a maintenance tool shutdown instruction and a tooling equipment locking instruction. The sending unit is used to select the corresponding control command according to the graded early warning signal, and send the control command to the control module of the maintenance tool and tooling equipment through the industrial bus. The data acquisition unit is used to acquire real-time instruction execution feedback data of maintenance tools and tooling equipment, and to determine whether the maintenance tools and tooling equipment have adjusted their operating status according to the control instructions based on the instruction execution feedback data; The adjustment unit is used to resend control commands and increase command priority if the maintenance tools and equipment have not completed the execution of the instructions, until the status of the maintenance tools and equipment meets the control requirements, thus forming a closed-loop control.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images