898 results about "Exception handling" patented technology

The invention discloses a dynamic multi-modal data fusion and real-time analysis method, and particularly relates to the technical field of data analysis. Carrying out standardization and time alignment on the multi-modal data, evaluating the stability of each modal in the current scene, and calculating a stability score; extracting features by using a deep neural network, fusing the features, detecting abnormal synchronization offset between modes in real time, and calculating collaborative offset intensity; based on the combined feedback of the stability score and the offset intensity, soft shielding processing is executed on the suspected abnormal mode; the future fusion risk is further evaluated through a time sequence prediction algorithm, and the model execution process is interrupted and an exception handling mechanism is triggered when necessary, so that the reliability and the safety of the multi-modal system in a complex environment are remarkably improved.

The invention relates to the technical field of information, and discloses a ciphertext execution exception handling method and system for a privacy computing algorithm, which quantifies the data query sensitivity of each tenant by collecting differential privacy query parameters of a plurality of tenants, and performs time sequence analysis on the sensitivity quantification result by using an ARIMA model to predict the future trend of the tenant, so as to improve the accuracy of the ciphertext execution exception handling of the privacy computing algorithm. The residual error between the actual sensitivity and the predicted sensitivity is calculated to identify abnormal tenants; a sliding window method is adopted to extract statistical features of the abnormal tenants to construct a feature set, and clustering analysis is carried out on the feature set to identify abnormal modes and abnormal fingerprint feature clusters corresponding to the abnormal tenants; and training a model based on the positive and abnormal fingerprint feature clusters of the abnormal tenants, and quantifying the influence degree of each abnormal mode on the privacy protection level of the abnormal tenants by using the model, so as to adjust and execute the privacy budget allocation proportion of each abnormal tenant according to the obtained quantitative analysis result. And therefore, the exception in the ciphertext execution of the privacy calculation algorithm can be effectively processed.

The invention relates to the technical field of process management, and discloses a lightweight process management method, which comprises the following steps of: defining a business process and analyzing the business process through a process engine; a graphical process design interface is provided, and a user can dynamically configure business process nodes through dragging and clicking operations; the business process is disassembled into a plurality of task nodes, and each task node is packaged into an independent function module through modularization design; the task nodes are combined according to needs and executed in parallel on different processors, and efficient execution of the task nodes is ensured; the execution state of the business process is monitored in real time, and potential problems are found and processed in time through log recording and an exception processing mechanism. According to the technical scheme of supporting dynamic adjustment of the task nodes and optimizing parameters based on the historical abnormal mode, execution parameters of the task nodes are dynamically corrected, association logic between the nodes is adjusted in real time, and the effects that process management is flexible and adapts to complex service scenes, and operation stability is improved are achieved.

The invention provides a large model agent collaborative scheduling method and system oriented to complex tasks, relates to the technical field of artificial intelligence, and comprises the steps of task decomposition, feature extraction, agent matching, dynamic scoring, scheduling scheme generation and optimization, execution monitoring, exception handling and the like to realize efficient collaboration of large model agents. According to the method, accurate matching can be carried out according to task characteristics and intelligent agent capabilities, the task completion efficiency and quality are improved, meanwhile, the dynamic adjustment capability is achieved, abnormal conditions in the execution process are effectively handled, and the system robustness is enhanced.

The invention discloses a mechanical arm execution method for dynamic task tree traversal. The mechanical arm execution method comprises the following steps that S1, nodes of an analyzed tree topology structure task sequence are received; s2, traversing task nodes in an iteration mode, determining task types, and dynamically adjusting task type priorities of the nodes by utilizing a machine learning model; s3, on the basis of the adjusted priorities, mechanical arm actions defined by the nodes are executed, and execution results are fed back in real time; s4, monitoring the execution state of the node, and triggering an exception handling mechanism for adjustment when deviation or exception is detected; s5, analyzing a behavior pattern of an operator through an intention recognition technology, predicting a next operation intention, and dynamically adjusting subsequent nodes of the task tree; and S6, after the execution of the nodes is completed, an execution result is fed back to the machine learning model to optimize the decision until all the task nodes are executed. Through dynamic adjustment and real-time feedback, the task execution efficiency is improved, and the reliability of the mechanical arm in a complex and changeable task environment is improved.

The invention discloses an agent optimization method, device and system based on reinforcement learning, and a storage medium. The method comprises the steps of obtaining an initial task instruction input by a user; performing instruction deepening and expansion on the initial task instruction; optimizing and adjusting a task instruction and a data collection strategy by using a reinforcement learning algorithm; the task instruction is converted into a specific execution plan, and data collection, data cleaning and analysis, API calling, database retrieval and exception handling tasks are executed; evaluating whether the data collected by the intelligent agent is enough to support the analysis requirement of the current task; and generating an industry analysis result based on optimization data obtained by instruction deepening and expansion, and by adopting the technical scheme of the invention, the intelligent agent can adaptively generate, adjust and optimize the task instruction in a complex task environment, so that the intelligence, flexibility and efficiency of task processing are improved.

According to the cross-platform kernel adaptation method provided by the embodiment of the invention, by accurately understanding kernel interface semantics, establishing a flexible parameter conversion mechanism and ensuring stable operation of a system through an efficient scheduling strategy, and by introducing deep learning and natural language processing technologies, the accuracy of interface matching is improved, better cross-platform compatibility is realized, and the performance of the system is improved. Meanwhile, monitoring and exception handling during operation are enhanced, and the reliability and safety of the system are improved.

The invention relates to the field of intelligent control and software management, and discloses a carrier rocket software control method and decision making device based on SAC, and the method comprises the following steps: obtaining the state information of a carrier rocket software system, and constructing a unified expression; an optimal action is generated based on an SAC algorithm; executing the action and collecting feedback information after the action is executed; utilizing feedback information to optimize an actor network and a commentator network in the SAC model, and continuously updating a decision strategy; based on the optimized strategy, the task priority, the running state of the software module and the abnormal recovery operation are dynamically adjusted, and the system state is collected again to optimize the model in a closed-loop mode. The system operation state can be adjusted in real time, the task scheduling efficiency, the resource utilization rate and the exception handling capacity are improved, intelligent management of a carrier rocket software system is achieved, and the method is suitable for complex task environments and high-dynamic operation scenes.

The invention relates to the technical field of equipment abnormity control, and discloses an equipment abnormity self-closed-loop control system. An anomaly detection engine of the system collects operation state data of target equipment in real time, and performs preliminary anomaly marking according to a preset anomaly judgment rule; the dynamic analysis unit receives the preliminary anomaly marks, generates anomaly feature vectors in combination with historical operation data, and inputs the anomaly feature vectors into a multi-dimensional evaluation model to divide anomaly levels; a strategy generator is matched with a preset repair strategy library according to the abnormal grade division result, and a corresponding repair instruction sequence is generated; the execution control module analyzes the repair instruction sequence, and dispatches hardware resources to execute repair operation according to a preset execution priority; and the feedback verification unit collects the running state data again after restoration, and compares the running state data with the initial abnormal mark to generate a restoration verification result. The system realizes full-process automatic processing of equipment exceptions, and improves exception processing efficiency and equipment operation stability.

The invention provides a software code defect detection method and system based on program code feature fusion, and the method comprises the steps: carrying out the grammar and structure inspection of a code through a static analysis tool, and extracting defect features related to control flow chaos, exception handling and resource leakage; running a preset test case through a dynamic analysis tool, capturing an execution path and an abnormal behavior in a runtime state, and obtaining defect features which are not covered by static analysis; if the output result of the decision tree model is that defects exist, grouping comprehensive defect feature vectors by adopting a density-based clustering algorithm, determining the types and distribution ranges of the defects, and selecting the density clustering algorithm to process defect features in non-spherical distribution; and according to a clustering result, generating a defect repairing suggestion, combining code logic and execution path information, and determining a specific position and an influence range of the defect. According to the method, the stealth defect in the code can be effectively identified and positioned, the software quality and reliability are improved, and timely and accurate defect repairing guidance is provided for developers.

The invention discloses an intelligent tightening tool capacity allocation system and method based on deep learning, and relates to the technical field of tightening tools, multi-modal data is captured in real time, the data weight is dynamically adjusted by using an attention mechanism, the capacity is predicted in combination with a deep learning algorithm, and dynamic production requirements and sudden tasks are quickly responded; in the aspect of exception handling, dual indexes of a variational auto-encoder VAE model, reconstruction error and distribution deviation are introduced, an exception mode is captured, a probability distribution model of normal operation data is established in combination with unsupervised learning, and the false alarm rate and the missing report rate are remarkably reduced; abnormal data are fed back to the productivity prediction module in real time through a closed-loop mechanism, and deep learning model parameters are dynamically optimized, so that the system can adapt to changes of production tasks and attenuation of tool performance; and the allocation module is adopted to allocate tool types and optimize paths through task priorities, so that the resource utilization efficiency is improved, and tool conflicts and station waiting time are reduced.

The invention discloses a dynamic script compiling and hot updating system based on Grouping and an implementation method of the dynamic script compiling and hot updating system. The system comprises a script source code management module, a dynamic compiling engine, an intelligent cache manager, an exception handling and recovery system and a hot update controller. And the script source code management module receives and preprocesses the Grouping script to generate a unique identifier for subsequent management. And the dynamic compilation engine executes compilation operation according to the identifier to ensure that a compilation result accords with a specification. And the intelligent cache manager efficiently stores and retrieves the compiling result, so that the repeated compiling cost is reduced. And the exception handling and recovery system monitors the execution state in real time, and automatically starts a recovery process in case of exception to guarantee stable operation of the system. The hot update controller realizes seamless update and version switching of scripts, and ensures service continuity. Through cooperative work of the modules, high-reliability, high-performance and zero-shutdown deployment is realized.

The invention relates to the field of industrial automation testing, discloses a multifunctional testing tool and method for an EUROMAP interface of an injection molding machine, and achieves the crossing from single function verification to system-level compliance evaluation. A test network is constructed through a controller, and deep compliance verification, coverage data interaction, exception handling and cross-layer response consistency are realized. The test tool integrates the functions of protocol analysis, signal simulation, fault injection and the like, supports dynamic generation of boundary value test scenes and abnormal injection of signal delay, CRC errors and the like, and verifies the time sequence / logic consistency of a protocol instruction and an execution mechanism response through a hardware-level timestamp alignment technology. According to the method, an evaluation report including compliance rating, fault diagnosis and optimization suggestions is automatically generated, the problems of single coverage dimension, insufficient abnormal simulation, cross-layer verification deficiency and the like of a traditional test are solved, the intelligence and standardization level of the injection molding machine interface test is remarkably improved, and the equipment interconnection and intercommunication performance is guaranteed.

The invention relates to the technical field of storage, in particular to a test method, device, equipment, system, medium and product, and the method comprises the steps: obtaining a test management command of an input and output management command representing error injection in a real working environment of a disk array controller chip; the command is analyzed to obtain the error injection test type and the corresponding related parameters, and the error processing mode is changed according to the error injection test type, so that the exception is only reported and not processed when encountering the exception in the test, the interference of exception processing on the test process is avoided, and the test can be smoothly carried out according to the preset error injection scene; and according to the error injection test type and the related parameters, constructing analog input and output and injecting the analog input and output into the target hardware unit to perform an execution behavior test, thereby realizing accurate test of the specific hardware unit in a specific error scene in an environment of performing input and output test based on a read-write input and output test program, so that the test is more comprehensive, and the test efficiency is improved. And the test precision is higher.

The invention relates to the field of automatic crane control, in particular to a gantry crane remote control platform access system and method, and the system comprises a task prediction module, a task distribution module, a task execution module and an exception handling module. The task prediction module generates task prediction data by collecting the task pool and the equipment state data; the task allocation module dynamically generates a task allocation scheme by using a reinforcement learning optimization algorithm and a simulated annealing algorithm; the task execution module generates control path data through real-time data acquisition and multi-modal data fusion, and controls equipment to execute tasks; and the exception handling module predicts a fault through a time sequence model and implements a task relay and dynamic recovery mechanism. According to the method, the efficiency of task scheduling, the precision of a control path and the stability of system operation are improved, and the execution efficiency of a remote control task is remarkably improved.

Methods and apparatus consistent with the present disclosure may be performed by a Cloud computing device may use instrumentation code that remains transparent to an application program that the instrumentation code has been injected into, may perform deep packet inspection (DPI) on computer data, or identify a content rating associated with computer data. In certain instances, data sets that include executable code may be received via packetized communications or be received via other means, such as, receiving a file from a data store. The present technique allows one or more processors executing instrumentation code to monitor actions performed by the program code included in a received data set. Malware can be detected using exception handling to track memory allocations of the program code included in the received data set. Furthermore, access to content associated with malware, potential malware, or with inappropriate content ratings may be blocked.

The invention relates to a comprehensive management process design system and method based on AI code interpretation, and belongs to the technical field of computer networks. The system comprises a process design console, an AI code interpreter and a business logic execution framework, the flow design console is used for providing a Web interface, so that a user can design and manage a flow in an interactive form, including interactive natural language programming, real-time feedback and preview, flow deployment and management, and historical record and version control; the AI code interpreter is used for converting business requirements and processes expressed by a user in a natural language form into executable codes, and comprises a natural language understanding sub-module, a code generation sub-module and a security verification sub-module; the business logic execution framework comprises a code execution sub-module, a data access sub-module and an exception handling sub-module. According to the method, the technical threshold of process design is lowered, personnel can easily master, and organic combination of intelligence and humanization is achieved.

The invention discloses a scheduling instruction real-time monitoring system for streaming data processing, which relates to the technical field of streaming data processing and comprises a data access module, an instruction analysis module, a dynamic scheduling strategy module, a real-time monitoring module, a scheduling execution module, a visual monitoring interface, an exception handling module and an extension module. The data access module is used for receiving a multi-source heterogeneous data stream in real time and carrying out protocol analysis and priority marking on input data, and the instruction analysis module is connected with the data access module and is used for extracting a scheduling instruction in the data stream and converting the scheduling instruction into executable operation; and the dynamic scheduling strategy module generates a dynamic scheduling strategy based on the real-time data traffic, the system load and a preset rule. Through the dynamic scheduling strategy module and real-time monitoring feedback, the system can automatically adjust resource allocation according to real-time load prediction, and dynamic balance of high resource utilization rate and low delay is achieved.

The invention discloses a game controller dynamic configuration system, and relates to the technical field of computers. The system comprises a controller end and a host end. The controller end adopts a heterogeneous equipment hybrid communication protocol stack to realize bidirectional data synchronization between the host end and the controller end; the controller end comprises a trigger, a selector, an interceptor, a switch assembly, a dynamic resource configuration unit and a firmware upgrading interface; the host end comprises an equipment dynamic identification and registration module, a game semantic analysis module, a dynamic mapping strategy generation model, a user adaptive learning unit, a multi-modal feedback fusion unit and a security verification and exception handling unit. According to the method, the mapping relation library of the controller hardware features and the game scene parameters is established, the physical state data of the controller is collected in real time through the sensor group, and the dynamic matching rule is generated in combination with the role action threshold in the game scene, so that the controller configuration difficulty is reduced, and the controller configuration efficiency is improved.

The invention discloses an intelligent timed task configuration and multi-mode feedback system based on natural language interaction, and relates to the field of industrial automation control, enterprise-level task scheduling management and instant messaging. The system comprises a natural language interaction module, a user natural language instruction is converted into structured semantic information through a BERT + BiLSTM + CRF mixed architecture, and lightweight processing is achieved in combination with TinyBERT distillation; the dynamic task arrangement module completes task creation, modification and the like based on structured information, completes missing parameters through historical task mode matching, and supports multi-session real-time synchronization; the rich media report generation module generates a multi-mode report containing charts, tables and the like, and the RAG technology and large model interpretation are combined; and the multi-channel pushing adaptation module adapts multiple platforms by adopting a strategy and factory mode. According to the system, the task configuration efficiency and flexibility are improved, the feedback form is enriched, the information pushing accuracy is guaranteed, and the exception handling capacity is enhanced.

The invention discloses an exception handling method and device for an embedded system, equipment and a storage medium, and relates to the technical field of embedding. The method comprises the steps of determining a current operation mode of a target processor according to a program state register of the target processor under the condition that abnormal codes appear in the operation process of the target processor in a target embedded system; according to the program state register, switching the current operation mode into a target operation mode, and obtaining a stack pointer value of a stack pointer register of the target processor in the target operation mode; and determining a target function address of a target function to which the abnormal code belongs and a call chain of the target function according to the stack pointer value. According to the technical scheme, the target function address is determined according to the stack pointer register, and before the target function address is determined, the target processor is switched to the target operation module and the interrupt is closed, so that interruption in the exception handling process can be effectively avoided, and the exception handling reliability can be improved.

The invention relates to an operating system startup exception processing method and device, equipment, a medium and a program. The method comprises the steps that when an operating system is started, a loading program is guided to load an external initramfs file into a memory, and an initial address of the external initramfs file is transmitted to a kernel of the operating system; performing Hash verification on the external initramfs file by executing the Hash verification function to obtain a Hash value of the external initramfs file; the operating system kernel decompresses the external initramfs file into a root file system according to the initial address; and when it is determined that the external initramfs file cannot be normally decompressed, determining a decompression failure reason according to a comparison result of the hash value of the external initramfs file and an expected hash value. And reporting the decompression failure problem to the corresponding processing object according to the decompression failure reason. By adopting the method, the decompression failure reason can be quickly and accurately positioned, the decompression failure problem can be pointedly processed, and resource waste is avoided.

The embodiment of the invention provides an application log exception processing method and device, equipment and a storage medium, and the method comprises the steps: extracting exception information corresponding to an application identifier from original log data, and determining a target keyword and a target time range based on a preset alarm rule and the exception information; determining candidate log data from the original log data according to the target time range and the target keyword, and filtering the candidate log data to obtain target log data; inputting the target log data and a set prompt text into the target network model to obtain a problem description text and a solution description text; and determining a manual processing description text associated with the problem description text from a stored manual processing record corresponding to the application identifier, fusing the solution description text and the manual processing description text to obtain a target scheme description text, and feeding back the target scheme description text to the background server. According to the scheme, the log exception processing efficiency can be improved.

Disclosed in the present invention are a method and apparatus for performing full-link reconciliation on the basis of a snowflake algorithm, and a device and a medium. The method comprises: responding to a cross-border business reconciliation request, and acquiring bill source data from a business terminal; parsing the acquired bill source data, and converting same into target bill data; calling a reconciliation rule engine to acquire a corresponding reconciliation rule and select a corresponding reconciliation algorithm, and executing a full-link reconciliation task for the target bill data, so as to obtain a reconciliation result; transmitting reconciliation imbalance data to a corresponding exception handling system for checking processing of abnormal data; and on the basis of an optimized snowflake algorithm, generating a unique target ID to associate a corresponding bill with the reconciliation result, automatically penetrating full-link data of an information flow and a fund flow on the basis of the corresponding unique target ID in the reconciliation result, and on the basis of the unique target ID, tracing the whole process of the circulation of a corresponding bill between the fund flow and the information flow, so as to achieve a complete reconciliation closed loop. In this way, the reconciliation working efficiency and data accuracy of the full-link reconciliation of a cross-border business are improved.

The invention provides an RPA-based information mining method and an artificial intelligence system, relates to the technical field of artificial intelligence, and realizes accurate evaluation and processing of abnormal risks of scheduled nodes in a process execution process. Firstly, a to-be-detected process and a process execution path diagram and a scheduling path diagram of the scheduled nodes of the to-be-detected process are acquired; and performing anomaly diagnosis by using the optimized target RPA anomaly diagnosis model to generate an anomaly risk index of each node. Then, the nodes are arranged in sequence through the risk indexes, the risk weights are generated, and then the corresponding exception handling knowledge base is matched for the to-be-detected process according to the risk weights, so that the accuracy and efficiency of exception detection are improved, accurate matching of the exception handling knowledge is realized, the stability and reliability of the RPA process are remarkably enhanced, and the reliability of the RPA process is improved. The operation and maintenance cost is reduced, and the automation level and the response speed of business processing are improved.

The invention provides a control method, device and equipment for an FMS production line warehouse system based on fault warning and elimination verification and a storage medium. The control method comprises the steps that material attribute information is acquired, a classification model based on machine learning is adopted to conduct classification and recognition on materials, a unique material identification code is generated and bound with an electronic tag, and a real-time material tracking system is established; a storage position distribution model based on multi-objective optimization is constructed, material relevance, use frequency and warehouse constraint conditions are considered, a material storage position distribution scheme is generated, and a corresponding carrying equipment control instruction is generated; a real-time monitoring and exception handling mechanism is established, and when it is detected that the actual state of the material does not accord with the system record, a fault warning process is triggered, an exception handling control instruction is generated, and a correction action is executed; the material demand prediction is carried out based on the neural network model, the replenishment control instruction is generated according to the prediction result, the accuracy of the execution result is ensured through a secondary verification mechanism, and the reliability of the warehouse system is improved.

The invention discloses a verification method for a second-level cache interface protocol, which belongs to the technical field of multi-core verification, and comprises the following steps of: setting verification targets, including function correctness, performance standard, compatibility and exception handling capacity; establishing a verification environment, wherein the verification environment comprises a hardware platform and a test tool; function verification, including read-write operation verification, cache line state transition verification and consistency verification; performance verification including delay testing and bandwidth testing; the compatibility verification comprises the compatibility with a first-level cache and the compatibility with a processor core; the abnormal condition verification comprises interrupt processing verification and error injection verification; and analyzing a test result. According to the verification method for the second-level cache interface protocol, the second-level cache interface protocol can be comprehensively and deeply verified, the function correctness, the performance standard reaching performance and the compatibility with other components of the second-level cache interface protocol are ensured, and a solid guarantee is provided for stable operation of a high-performance computer system.

The invention provides a time series prediction method and system based on exception handling, and the method comprises the steps: detecting an abnormal fluctuation point through employing the abnormal fluctuation point, calculating the weight of the abnormal fluctuation point according to the distance between the abnormal fluctuation point and other data points, carrying out the weight adjustment of other data points according to the weight of the abnormal fluctuation point, and carrying out the prediction of the time series. The influence of abnormal fluctuation points on subsequent data feature extraction can be effectively weakened, so that the feature extraction error is reduced, and the overall prediction precision is improved; a ConvLSTM network is utilized to perform feature extraction, and the ConvLSTM network can analyze a spatial relationship among multivariate variables on the basis of extracting a time relationship of time sequence data, so that the problem that data feature information extracted only from a time relationship level is incomplete can be remarkably improved, and prediction precision is improved; a double-attention network is provided, the influence weights of different time sequences are balanced and calculated through self-attention mechanism adaptive learning, and the prediction precision is improved.

The invention discloses a data processing method and system. The method comprises the steps that in response to a data collection instruction of a processing system PS end, collected data are stored in a target cache according to the data type of the collected data, and the target cache belongs to a programmable logic PL end and corresponds to the data type; under the condition that the target cache is full, the collected data are stored in a target memory area according to the data type, and the target memory area belongs to the memory space mounted by the PS end and corresponds to the data type; and in response to the received analysis data, processing is carried out by adopting the target model based on the analysis data to obtain an exception identification result, the analysis data is obtained by analyzing the collected data by the PS end, and the PS end is used for generating an exception processing strategy based on the exception identification result. The technical problem that the data processing efficiency is not ideal in the prior art is solved.

The invention discloses an agent assisted testing method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring demand information of a Web system to be tested; analyzing the demand information and generating a test case set; performing hierarchical labeling on the test case set to form a hierarchical case set containing the basic operation type case, the data verification type case and the complex logic type case; distributing the basic operation type use case to an execution module of the agent system, distributing the data verification type use case to an interface automation test module, and distributing the complex logic type use case to an artificial interaction terminal; and when the execution module or the interface automatic test module monitors that the case execution fails or the state is abnormal, an exception processing strategy is matched according to an exception type and an exception source, corresponding processing operation is executed, and feedback information used for optimizing the layered case set is generated. According to the method, the test tasks can be intelligently allocated and accurately executed, and automatic adaptation and optimization can be performed when an exception occurs, so that an efficient man-machine collaborative test is formed.