37 results about "Repair - action" patented technology

The invention relates to the technical field of application state diagnosis, in particular to an application state dynamic diagnosis and automatic repair method and system.The method comprises the steps that multi-source heterogeneous logs are collected and standardized in real time, key fields are extracted, context information is injected, and structured log data are generated; inputting the structured log data into a dynamic anomaly detection model, constructing a dual-path detection mechanism based on LSTM time sequence analysis and a graph neural network, identifying an abnormal mode and positioning a fault root cause; according to the output of the anomaly detection model, a repair action is triggered in a grading manner through an intelligent repair strategy engine; in the repairing process, system state changes are stored and recorded through a pre-writing type redundancy log, automatic rollback during abnormity is achieved on the basis of check point information, and data consistency and system stability are guaranteed. The problem of service interruption or data inconsistency possibly caused by traditional automatic repair is avoided, and the reliability of automatic operation and maintenance is improved.

The invention discloses an abnormality diagnosis and repair method and device for an automatic process of a robot and a medium, and relates to the technical field of artificial intelligence. The anomaly diagnosis and repair method for the automatic process of the robot comprises the following steps: acquiring RPA anomaly processing data; matching the RPA exception processing data with a target causal knowledge base to obtain an exception scene repair action; and according to the abnormal scene repair action and the large language model, generating a process repair script, and according to the process repair script, performing abnormal scene process repair. According to the technical scheme provided by the embodiment of the invention, the handling effect of the PRA in the exception handling scene can be greatly improved on the premise of low maintenance cost.

The invention relates to the field of artificial intelligence, and particularly discloses a system repair and risk control dynamic balance method, which comprises the following steps: when a preset data acquisition condition is met, acquiring system operation state information, risk state information and repair historical data of a computer system, and vectorizing the information to obtain a state feature vector; defining a repair action space set comprising a plurality of repair actions; constructing a reward function; inputting the state feature vector into a reinforcement learning model, and outputting an optimal repair action; acquiring a real-time system operation state index of the computer system, and comparing the real-time system operation state index with a preset business red line threshold value; if the real-time system operation state index does not exceed the business red line threshold value, the optimal repair action is executed; if any real-time system operation state index exceeds a business red line threshold value, a replacement repair action is executed; and updating the state feature vector and the reward value of the computer system, and updating the parameters of the reinforcement learning model.

Systems and techniques are disclosed for automated damage reporting and repair assessment for automated teller machines (ATMs). An example technique may include detecting damage to an ATM using a sensor. Based on the detected damage, the example technique may include performing a damage assessment and generating a damage report including at least one of a level of damage, a type of damage, a recommended repair action, or an indication to dispatch a technician. The example technique may include outputting the damage report.

The invention discloses a configuration item repair-oriented Linux kernel boot test and configuration consistency intelligent repair method. The method comprises the following steps: collecting a compiling log, a boot log and a running state; performing segmentation, pattern matching and field extraction on the log, normalizing error information into structured error entries, and distinguishing a dependency missing class, a mutual exclusion conflict class and other compiling failure classes; performing preprocessing and grammar analysis on the Kconfig file, extracting configuration item attributes and constraint relationships, constructing a unified dependency graph and generating a configuration item metadata table; classifying and marking the configuration items based on the metadata, and calculating a key score; generating a repair action according to the structured error entry, the unified dependency graph and the key score, performing iteration execution, and performing recompilation and restart verification after each iteration until a reparated configuration file which can be compiled and started is obtained or a preset iteration upper limit is reached; according to the invention, the cross-version adaptation and problem closed-loop efficiency is improved.

The embodiment of the invention provides a data transmission method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining target transmission data of a transaction party, judging whether abnormal data exists in the target transmission data or not according to a preset semantic rule set, and if the abnormal data exists, sending the abnormal data to the transaction party; if the abnormal data is the repairable data, identifying whether the abnormal data is the repairable data according to a repair rule, and if the abnormal data is the repairable data, repairing the repairable data according to a preset repair action indicated in the repair rule to obtain the repaired transmission data, and transmitting the repaired transmission data and qualified data except the abnormal data in the target transmission data to a receiver.

The invention discloses a local adjustment method and system for electric transmission line tower arrangement, belongs to the technical field of electric transmission line design, and aims to solve the problems of local violation, high global optimization cost, lack of a self-adaptive repair strategy and dependence on experience decision in traditional manual arrangement. The method comprises the following steps: firstly, carrying out conductor sag fitting and comprehensive checking of sag, ground clearance, crossing and windage yaw on a tower arrangement scheme; recognizing a violation tower position according to a result, and determining a local repair section; then generating candidate repairing actions such as tower moving, tower height increasing and tower inserting; then constructing a multi-objective optimization model, quantifying the repair cost and sequencing the actions; and finally, executing the action according to the priority, and iteratively checking until the hard constraint is met. The method focuses on local repair, reduces the cost, improves the efficiency, adapts to multiple voltage grades and complex terrains, and provides intelligent support for pole tower arrangement design.

The invention discloses a power transmission line dynamic first-aid repair scheduling and path planning method and system under disasters, and relates to the technical field of power emergency first-aid repair, and the method comprises the steps: sensing the topological dynamic change of a power transmission line under the disasters in real time, fusing the geographic position priority of important users and the distribution information of first-aid repair resources, and carrying out the first-aid repair scheduling and path planning. Constructing a dynamic resource scheduling model under various constraint conditions; and generating a first-aid repair path based on a fastest path algorithm, establishing a recoverability evaluation system including dimensions such as resource matching degree and topological connectivity, and triggering a first-aid repair action when an evaluation value reaches 80% or above. According to the method, the problems that the traditional emergency repair scheduling response is lagged, the path planning is separated from the disaster real-time scene, and the resource configuration and the recovery demand are not matched are solved, the accuracy, the dynamic property and the high efficiency of the emergency repair of the power transmission line under the disaster condition are realized, and the post-disaster recovery efficiency of the power grid is remarkably improved.

The present application relates to the technical field of metal plate intelligent repair, and specifically discloses a stainless steel cathode plate self-adaptive repair method and system simulating manual operation. The method comprises: obtaining current three-dimensional topography data of the cathode plate to be repaired; inputting the data into a first machine learning model trained based on multi-modal expert data to generate an initial repair action sequence; controlling a repair execution unit to execute the sequence, and online obtaining intermediate three-dimensional topography of the cathode plate as a state during the repair process; driving a reinforcement learning algorithm based on the difference between the state and a target state to dynamically adjust subsequent repair actions; and data of successful cases will be used to update a training database and retrain the first machine learning model. The method can solidify expert experience through imitation learning, and realize online adaptive decision-making in combination with reinforcement learning, thereby achieving individualized high-precision repair of various complex deformations, and the flatness after repair can reach below 0.5 mm, and the system has continuous optimization capability.

The invention provides a method for autonomously repairing and controlling instability of a bogie and a railway vehicle. Wheel out-of-roundness problems such as wheel equivalent taper, radial run-out and wheel high-order polygon are autonomously repaired by detecting the wheel equivalent taper, the radial run-out and the wheel high-order polygon and grinding a wheel tread; releasing wheel equivalent taper, radial run-out and wheel high-order polygon alarms, automatically repairing high-pressure intermittence, low-pressure intermittence and low-pressure continuity through the wheel equivalent taper and radial run-out, and comparing a high-pressure continuity mode with a non-grinding sub-mode to obtain the relationship between the repairing action mileage and the wheel tread modification efficiency; the method is used for restraining rapid growth of wheel out-of-roundness problems such as wheel equivalent conicity, radial run-out and wheel high-order polygons. The relation between the millstone different-mode mileage and the wheel equivalent conicity radial run-out growth rate is obtained.

The invention discloses a restoration method and device based on dynamic call ticket abnormity and electronic equipment, and the method comprises the steps: firstly obtaining original call ticket data, then carrying out the preprocessing of the original call ticket data, obtaining structured data, carrying out the data fusion based on the structured data, obtaining multi-source data, carrying out the abnormity detection of the multi-source data through a first preset mode, and obtaining a dynamic call ticket abnormity result. The method comprises the steps of obtaining first abnormal information, classifying the first abnormal information through a first preset model to obtain a classification result, processing the classification result based on a first preset strategy to obtain a content-level repair scheme, and finally processing the content-level repair scheme based on a second preset model to obtain a repair action sequence. And the abnormal ticket is automatically repaired based on the repairing action sequence, so that waste of human resources is avoided.

System and methods for automated responses to M-Plane fault management in a Radio Access Network.

This application relates to a throttle control method, a vehicle, and a storage medium. The method includes: acquiring throttle jamming determination parameters; and sending a shutdown request to a power controller when the jamming determination parameters exceed a preset parameter threshold and the duration exceeds a preset time. The power controller responds to the shutdown request by controlling the engine to shut down, and after the engine shuts down, it performs a throttle repair action. By performing the throttle repair action while the engine is shut down, the repair drive action is decoupled from the engine's real-time operating conditions, eliminating the conflict between the repair drive action and the engine's torque demand at the source, avoiding safety hazards caused by torque mismatch, effectively relieving temporary throttle jamming and quickly restoring normal control, significantly improving the overall vehicle driving experience and ride smoothness.

The GPU graphics processor fault repair method and system provided in this application first determines the connectivity fault tolerance threshold based on the node communication topology of the currently running training service. This threshold represents the actual node communication dependencies of the training service. Then, it enters the fault batch division stage. Using the connectivity fault tolerance threshold as a constraint, the batch division is completed in combination with the error information reported by the faulty nodes. This ensures that the repair scale of each batch does not exceed the safety limit that the service can bear, avoiding the risk of service communication interruption and training task crash caused by repair actions. At the same time, it can maximize the number of repaired nodes in a single batch within the safety boundary. This solves the problems of extremely low efficiency of single-node rolling repair and lack of safety guarantee of experience-based batch repair in the existing technology. Batch fault repair can be completed without interrupting the training task during the GPU repair process.

The aim is to improve the convenience of remote device repair. [Solution] The network device in the system of the present invention includes means for transmitting operational information to a device management system, means for executing a remote repair process in response to instructions from the device management system, means for receiving information on candidate automatic repair actions from the device management system, and means for executing processing for automatic repair based on the information on candidate automatic repair actions without executing the remote repair process, with the user's consent.

The invention belongs to the field of memory control, and provides a memory control method and a storage device.The method comprises the steps that a maintenance strategy setting command from a host system is received, and the command is used for configuring maintenance strategy data for a specified logic area; updating a maintenance record in a key data maintenance table according to the command, wherein the maintenance record comprises maintenance strategy data and dynamic metadata of a logic region; dynamically selecting a target logic area from the logic areas associated with the key data maintenance table, and executing background health examination on the target logic area; and according to a health examination result, triggering a grading repair action on the target logic region. According to the method, the reliability strategy is flexibly defined by the host, closed-loop management of health monitoring and hierarchical repair is autonomously executed by the storage device, the data degradation trend can be actively found and processed, and the reliability and maintainability of the storage device are effectively improved.

The invention discloses a computer intelligent service management system and method based on network big data, and belongs to the technical field of operation and maintenance management. Comprising a data stream generation module used for forming a context-aware data stream carrying a management intention; the knowledge network generation module is used for forming a knowledge network for business intention query; the candidate set generation module is used for generating a root cause candidate set with priority; and the action execution module is used for executing corresponding automatic repair actions. According to the method, the technical effects of intelligent association analysis of operation and maintenance data, accurate root cause positioning and system self-healing capability are achieved by constructing the context sensing data flow injected with service semantics, establishing a multi-dimensional context index knowledge network and realizing an automatic repair mechanism based on priority evaluation; the problems of weak operation and maintenance data relevance, low fault positioning efficiency and low manual response speed in the prior art are solved.

The invention is applicable to the technical field of vulnerability repair, and provides a self-adaptive vulnerability repair method, device and equipment based on gray sandbox verification, the production environment is prevented from being directly exposed to the risk by simulating and executing a repair action in a gray sandbox and recognizing the compatibility problem in advance, and then through real-time index monitoring and self-adaptive threshold decision, the vulnerability repair efficiency is improved. Pushing or rollback is automatically judged, manual intervention hysteresis is reduced, finally, a knowledge feedback mechanism is used for continuously optimizing the model, a closed-loop self-learning system is formed, and therefore preposed risk control, intelligent decision acceleration and long-term self-adaptive evolution of the repairing process are achieved on the whole, the operation and maintenance burden is effectively reduced, and the system stability is guaranteed; and the reliability and efficiency of vulnerability repair are greatly improved.

This invention provides a method and system for operating system fault self-healing based on reinforcement learning, comprising: real-time acquisition of multi-source heterogeneous observation data of a preset observation object, constructing a dynamic fault propagation graph, and calculating the operating system health; responding to the operating system health falling below a preset threshold, selecting fault domains and nodes based on a reinforcement learning-based localization strategy, and performing active probing to obtain new observation data; updating root cause beliefs based on the new observation data to obtain candidate fault locations; constructing a candidate repair action set based on the candidate fault locations, executing the optimal repair action, and determining whether fault self-healing is successful, continuing to execute repair actions, or executing a rollback action based on the execution result of the optimal repair action; recording the complete process of each fault handling, and iteratively updating the localization strategy and repair strategy. This invention can achieve rapid fault localization, high-confidence confirmation, low-risk repair, and continuous learning optimization without requiring historical fault samples.

A system applies an AI-based model pipeline for joint prediction of repair time and repair actions based on received fault reports related to manufacturing automation machines and devices. The model pipeline includes a first encoder configured as a Transformer neural network trained to transform fault report text into latent fault report representations that group semantically similar fault texts. Another Transformer encoder is trained to transform the latent fault report representations into latent repair report representations. A repair time prediction module with a feedforward neural network is trained to generate repair time predictions from the latent repair report representations against actual repair times. A repair action prediction module configured as a Transformer decoder neural network is trained to generate repair action predictions from the latent repair report representations against actual repair actions. End-to-end training of the pipeline is achieved by minimizing a combined loss function of the joint predictions.

The application belongs to the technical field of automatic operation and maintenance, and discloses a self-checking and self-repairing method and device for electronic equipment, equipment and a storage medium, wherein current configuration information and real-time running information of the electronic equipment are acquired, the real-time running information is denoised to obtain denoised running information, the current configuration information and the denoised running information are input into a trained fault prediction model to obtain a fault occurrence probability of the electronic equipment in a future time period, a target fault type is identified when the fault occurrence probability exceeds a preset threshold, the current configuration information, the denoised running information, the target fault type and the fault occurrence probability are input into a trained repair strategy model to obtain at least one configuration repair action, and the current configuration information of the electronic equipment is modified, so that the self-checking and self-repairing capabilities are realized, the dependence on manual intervention is reduced, and the automation level is improved.

Systems, methods, and techniques described herein relate to remediating vulnerabilities in data centers. In an aspect, a change in a configuration of a resource is automatically detected. Aggregated data representative of a resource inventory of the data center is generated. The resource inventory specifying a plurality of resources of the data center. A policy of the data center is determined. The policy specifies a rule applied to a resource of the plurality of resources. The resource is determined to fail to satisfy the rule based at least on the change in the configuration and / or the aggregated data. A remedial action is caused to be performed based at least on the resource failing to satisfy the rule. In an aspect, the remedial action comprises identifying a repair action specified in the policy and performing the repair action with respect to the resource.

A method is provided for repair of a mechanical or electromechanical system of a machine. An inference engine receives indication of a failure mode of the system, and measurements of operating conditions of the machine, and the inference engine defines a current problem including (a) the failure mode of the system, and (b) a pattern in the measurements. The inference engine searches a knowledge base with historical problems including (a) failure modes of systems of the machine, and (b) patterns in measurements of the operating conditions, and respective solutions with (c) repair actions performed to address the respective ones of the failure modes, for a respective solution to a historical problem most similar to the current problem. This is inferred as a solution to the current problem, and the inference engine generates an output display indicating the repair action of the solution to address the failure mode of the system.

This invention discloses a dual-arm robot and its control method for repairing broken strands in high-altitude cables, belonging to the field of power system operation and maintenance technology. The method includes: acquiring multi-source initial perception information of the cable to be repaired and generating a global environment model; driving the robot to move to the broken strand area based on this model, collecting detailed perception information to construct a dynamic three-dimensional operation model; performing semantic understanding and task decomposition on the repair task instructions to generate a logical sub-task sequence and action prototype information; generating low-level control instructions based on the action prototypes and executing the repair actions, integrating end-effector visual feedback and force feedback in real time during the process, and dynamically adjusting the control instructions; after the repair is completed, collecting scanned reconstruction information, comparing it with preset process standards, and autonomously deciding whether to perform supplementary repair or re-repair until it meets the standards; finally, withdrawing to a safe location. This method achieves fully autonomous perception, planning, execution, and quality verification for broken strand repair, significantly improving the safety, efficiency, and reliability of high-altitude operations.

The invention relates to the technical field of financial management analysis, in particular to an intelligent reconciliation error repairing method in a financial management system, which comprises the following steps: obtaining and processing financial data to generate a standardized reconciliation node set, and providing a data basis for reconciliation error analysis; by establishing a time sequence enhanced reconciliation relation graph and identifying a suspected error node cluster, a node set which may have errors is locked; by carrying out feature extraction and root cause classification on the suspected error node cluster, an error root source is judged, and classification confidence is given; an optimal repair scheme is screened out by matching an optimal repair action sequence and virtually simulating a rehearsal consequence, and the actual repair risk is reduced; the execution sequence can be flexibly adjusted through dynamic priority rearrangement of the optimal repair action sequence; the repair effect is verified based on double indexes, and the repair effect can be evaluated from account checking accuracy and business rule conformity; and through the iterative optimization of the reinforcement learning mechanism, the reconciliation error repair capability and accuracy are continuously improved.

This invention relates to the field of anomaly detection technology, and proposes a method, system, and storage medium for anomaly diagnosis and repair of energy storage devices. The method includes: performing anomaly pattern recognition on real-time operating data to obtain initial anomaly event information; reconstructing the operating conditions of a local area to obtain a local state parameter set; optimizing a preset repair strategy mapping library to obtain a targeted repair scheme; capturing the feedback state data of the targeted repair scheme in real time, and evaluating the repair expectation of the feedback state data to obtain a repair action execution record; performing a full-state parameter synchronous re-examination of the target energy storage device to obtain a full-state verification comparison result; and archiving and storing the repair action execution record and the full-state verification comparison result to obtain a repair success record. This invention can improve the efficiency of anomaly diagnosis and repair of energy storage devices.

The invention provides a GPU server fault pre-diagnosis system and a self-healing method. The method comprises the following steps: periodically collecting operation data of a GPU, calculating a window average value for SM utilization rate and power consumption, calculating a window accumulated value for ECC error count, and calculating a power consumption change slope; constructing a state vector based on the operation data, and performing normalization processing to obtain an operation state vector; inputting the running state vector into a pre-trained auto-encoder model, and calculating a mean square error of the state vector and a reconstruction vector as a reconstruction error; adding the reconstruction error and the regular term to obtain a risk score; calculating a benefit score of each repair action based on the risk score, the operation data and the environment context information; and executing a selected repair action, the repair action including holding, underclocking, migration or resetting, and updating the node state flag according to an execution result.

The device includes a memory to store computer readable program code for a knowledge-based system including an inference engine and a knowledge base; and a processing circuit configured to access the memory and execute the code. The device is caused to receive at least a time series of measurements of operating conditions of a machine recorded during operation of the machine. The device is also caused to cluster the time series of measurements into one or more respective clusters and identify patterns across the clusters. The device is further caused to define a current state of the machine including the patterns across the clusters, access and search a historical case in the knowledge base describing a respective solution to a historical problem state similar to the current state, the respective solution including a repair action, and generate an output display indicating the repair action to resolve the current state.

The invention relates to a white-box switch fault self-repairing method and system based on a large model, and the method achieves the precise semantic understanding and root cause reasoning of a fault through the combination of the large model and real-time multi-mode operation data, and introduces a sandbox safety verification and positive and negative sample closed-loop learning mechanism. The defects that traditional artificial experience and static scripts are slow in positioning and high in repairing risk in a complex scene are fundamentally overcome, the full automation level and accuracy of fault processing are remarkably improved, it is ensured that the repairing action is safe and controllable, the fault recovery time is remarkably shortened, and the overall reliability and operation and maintenance efficiency of a white-box switch network are improved.

This application provides a method and apparatus for repairing semiconductor equipment, as well as a model training method and apparatus. The semiconductor equipment repair model training method includes the following steps: acquiring a training data set and an original model; classifying the repair action data according to the target of the repair action to obtain a multi-level repair action set; obtaining reinforcement learning rules based on the execution result of the repair action and the reward function, wherein the reward function calculates corresponding positive or negative reward values ​​based on whether the fault is eliminated, whether a new fault is introduced, or whether the situation worsens after the repair action; training the original model based on the training data set and the reinforcement learning rules, outputting the evaluation value corresponding to each repair action in the multi-level repair action set, and storing the training data using an experience playback mechanism to obtain a repair model. The technical solution of this application can quickly locate the cause of the fault, facilitating efficient repair of equipment faults.