6271 results about "Test data" patented technology

The invention relates to the technical field of multi-modal data, provides a pilot competency dynamic evaluation method, system and device based on multi-modal data and a storage medium, and solves the problems of low accuracy of pilot competency evaluation and poor pertinence of training guidance. The method comprises the steps of collecting flight control data, physiological signal data, psychological assessment data, subjective scale data and international civil aviation organization core competency information, performing alignment fusion on multi-source data by adopting a time synchronization algorithm, and identifying an attention fixation mode based on a hidden Markov model. And constructing a workload index by fusing a subjective scale and a physiological entropy value, inputting the multi-modal features into a pre-trained competency assessment model, outputting three levels of psychological assessment indexes including a basic ability layer, a dynamic presentation layer and a risk early warning layer, and finally generating a personalized training report. According to the invention, the accuracy of pilot competency evaluation and the pertinence of training guidance are improved.

Systems and methods disclosed herein automatically evaluate, select, and coordinate artificial intelligence (AI)-based agents for collaborative distributed task execution based on dynamic, multi-attribute scoring and resource allocation models. The system obtains a task specification request defining a computational requirement set, a performance metric set, and an available resource set for one or more tasks to be executed by a network of AI-based agents. A first AI model set generates domain-specific test datasets and validates prospective agents by comparing agent-generated fingerprints against predetermined hash values stored on a distributed or federated ledger. A second AI model set constructs a multi-dimensional scoring data structure for each agent by using historical performance metrics to compute weighted composite scores. The system selects a subset of AI-based agents, ranks the agents, and allocates resources proportional to each agent's composite score. A third AI model set coordinates and executes distributed computer-executable workflows across the selected agents.

The present invention relates to the technical field of deep learning and computer vision. Disclosed is an unmanned aerial vehicle image-based small object detection method for target areas. The present invention crops images of obvious small objects in certain target areas, and annotates the small objects of different categories to form a raw training and testing dataset, so as to ensure the accuracy of data required in the early stage of the algorithm and further ensure the scientificity of the algorithm; uses the computing capability of an improved YOLOv7 detection model to collect image features of different degrees in the dataset, the improved YOLOv7 detection model using YOLOv7 as a basic model and adding to a neck network an MS-CET module, which is constituted by an improved self-attention mechanism and convolution module SPPCSP, and a BHC-FB module, which is constituted by bidirectional mixed convolution modules NConv and RPConv connected in parallel; and finally fuses different feature layers as a final judgment basis of an unmanned aerial vehicle for small object detection in the target areas, to further check the accuracy of the algorithm and criteria for dataset selection, thereby improving recognition accuracy.

The invention discloses an automatic anchor point searching and processing method for grid-connected test data of a photovoltaic inverter, and belongs to the technical field of automatic test of a power system. According to the method, a three-phase voltage and current signal output by a power grid simulator and an inverter power instruction signal are aligned through a high-precision time synchronization device; wavelet transform multi-scale noise reduction and moving average filtering combined preprocessing is adopted to improve the signal-to-noise ratio; identifying a voltage zero crossing point candidate set, a drop starting point candidate set and a recovery termination point candidate set based on a self-adaptive dynamic threshold value; transient energy characteristic verification is introduced for a voltage drop starting point; effective anchor points are confirmed through time window association of power instruction step changes. According to the method, the problems of low efficiency of manual key event point identification, misjudgment caused by noise interference, grid event and inverter response time sequence correlation missing and the like are solved, and the automation degree of test data analysis, anchor point positioning precision and control response time sequence analysis reliability are remarkably improved.

The invention discloses a fatigue life simulation evaluation method for a lightweight aluminum alloy material of a new energy automobile, and relates to the technical field of material life evaluation. A microstructure image is collected, coupling features are extracted through machine learning, and heterogeneous data fusion and enhancement are completed; generating a topological optimization structure based on a GAN, introducing a VPSC model to describe anisotropy according to a stress gradient dynamic grid, and constructing a dynamic finite element model; fusing vehicle driving data, predicting a load by using LSTM, performing VMD decomposition and environment correction, and realizing space-time correlation load spectrum reconstruction; a phase field model is used in a microcosmic mode, cracks are tracked in a macroscopic mode through XFEM, damage parameters are transmitted in a bidirectional coupling mode, and multi-physics field coupling simulation is carried out; fusing simulation and test data by adopting Bayesian reasoning, calculating life probability distribution, and correcting parameters when errors exceed the limit; according to the method, the fatigue life prediction error is finally reduced, the time consumption of single simulation is reduced, full-life-cycle evaluation and visual early warning are realized, an efficient scheme is provided for lightweight design, and industrial technology upgrading is promoted.

The invention relates to the technical field of deep learning, in particular to an AI chip test parameter adaptive optimization method based on deep learning, which comprises the following steps: acquiring historical test data of an AI chip, and calculating correlation strength among different failure modes based on the historical test data; identifying a failure coupling matrix according to the edge weight, and converting a preset static detection parameter constraint boundary into a dynamic constraint space changing along with a failure detection state; a multi-level optimization framework is constructed, the upper layer executes failure type correlation analysis and generates constraint propagation information, the middle layer optimizes a parameter cluster based on the constraint propagation information, and the lower layer adjusts a single detection parameter and outputs a parameter optimization result; establishing a neural network mapping model, and obtaining a nonlinear mapping relationship between the detection parameters and the failure types; based on the physical state parameters, the nonlinear mapping relation is adjusted, the dynamic constraint space is updated, parameter optimization is executed again, a parameter optimization result is output, and an optimal test parameter combination is output.

The invention provides a mutual inductor test data cloud edge cooperative processing method and device, and relates to the technical field of data processing, and the method comprises the steps: carrying out the time domain and frequency domain combined feature extraction of the original measurement data of a mutual inductor test through a multi-mode decoupling preprocessing model, and forming a data feature vector set, further generating a confidence label flow through state recognition and Bayesian inference, evaluating a prediction error, and realizing data classification screening and priority queue construction; the data are uploaded to a cloud end in a semantic compression and multi-resolution representation vector mode, so that the transmission efficiency is improved; carrying out deep modeling and model performance monitoring at the cloud, and if degradation is detected, returning edge original data to update the model; and finally, generating a scheduling weight factor according to the prediction error distribution graph, and dynamically optimizing an edge cloud task allocation proportion. According to the invention, the problem of low resource allocation efficiency caused by lack of a dynamic scheduling mechanism based on prediction errors and confidence driving in existing mutual inductor test data edge cloud cooperative processing can be solved.

The invention relates to the field of electric power material quality detection, and discloses a test scheduling system for detection task collaboration and data acquisition, which comprises a task construction module, a state collaboration module, a graph reasoning module and a data acquisition module. And the task construction module generates a standardized test task packet including a task identifier, a project code, a target equipment identifier, an environment requirement parameter and a two-dimensional code according to the test rule base and the resource configuration state, and pushes the standardized test task packet to corresponding test equipment through a Web Service interface. And the state collaboration module receives an equipment state feedback event, constructs an event time sequence flow graph based on the task identifier and generates a task state sequence with a timestamp. The atlas reasoning module takes the state sequence and the environmental parameters as input, constructs a test atlas structure and generates an optimization execution path. And the data acquisition module controls the test equipment to complete a detection task according to the path, acquires test data and environmental parameters, and encapsulates the test data and the environmental parameters to form a structured task data packet, thereby realizing data collection and task tracing.

The invention relates to the technical field of artificial intelligence and databases, in particular to an NL2SQL method and system based on reinforcement learning, and the method comprises the following steps: natural language input preprocessing, semantic analysis and abstract representation construction, mode linking and candidate set determination, SQL generation, query execution and feedback optimization, and multi-round interaction processing. The method has the beneficial effects that complex scenes such as nested sub-query, multi-table JOIN and aggregation function combination are supported, and the SQL generation accuracy is improved by more than 30% compared with that of a traditional method (test data is from a spider data set); a mode link mechanism based on reinforcement learning can quickly adapt to a new database mode, and in a cross-domain test (such as switching from an e-commerce database to a medical database), the accuracy rate decrease amplitude is less than 15%.

The invention discloses a railway vehicle test data management and analysis system, and the system comprises a multi-modal data collection module which collects the heterogeneous data of tests such as airtightness and weighing in real time; the block chain credible evidence storage module is used for performing time-space stamp marking and hash encryption on the data and verifying the integrity; the space-time atlas analysis module is used for constructing a space-time heterogeneous atlas and generating vehicle-level feature vectors through graph convolutional network fusion data; the predictive maintenance decision module is used for predicting subsystem performance degradation and fault risks and generating a hierarchical maintenance strategy; and the adaptive visual platform integrates a large screen and a mobile terminal to display data and decisions. According to the system, efficient integration and credible evidence storage of multi-source data are realized, the data utilization efficiency and analysis depth are improved, accurate maintenance decision is supported, and safe operation of railway vehicles is guaranteed.

The invention discloses a cross-chain smart contract vulnerability detection method and system based on multi-feature fusion learning. The method comprises the following steps: collecting a cross-chain smart contract vulnerability data set for cleaning and labeling; feature extraction is carried out from the source code and the byte code, an abstract syntax tree (AST) is extracted from the cleaned source code, a basic control flow graph (CFG) is extracted from the byte code, and a cross-chain control flow graph (xCFG) is constructed; carrying out feature representation on AST and xCFG, generating a graph vector through a graph neural network (GNN), generating a semantic vector through CodeBert, and fusing the semantic vector into a feature fusion vector; performing model training and detection, taking the generated vectors as training data and test data, obtaining a cross-chain smart contract vulnerability detection model by adopting Transform-FC model training data, and finally evaluating model performance through accuracy, recall rate, precision rate and F1 value. According to the method, the structural features and semantic features of the codes can be effectively fused, potential vulnerability information in the codes can be fully mined, the recognition capability of the model for cross-chain vulnerabilities can be enhanced, and the accuracy and reliability of the cross-chain vulnerability detection model can be improved, so that the security of a block chain system can be more efficiently guaranteed.

The invention belongs to the technical field of machine learning and pre-stress, and discloses a pre-stress release loss value prediction method and system based on machine learning, and the method comprises the steps: carrying out the multi-working-condition modeling and simulation of a pre-stress beam through finite element numerical software, and extracting the working parameters and design parameters of the pre-stress beam, carrying out data preprocessing, distribution check and feature importance analysis to obtain an initial data set; dividing the initial data set into an initial training data set and an initial test data set, and processing the initial training data set and the initial test data set to obtain a processed training data set and a processed test data set; constructing a full-connection multi-layer perceptron neural network model, defining training, verification and monitoring functions, training the full-connection multi-layer perceptron neural network model by using the processed training data set, and testing the trained model by using the processed test data set to obtain a prediction model; real parameters of the prestressed beam are obtained, the prediction model is used for predicting the prestress release loss value, and a prediction result is obtained.

The invention provides a single pile bearing capacity prediction method based on an XGBoost machine learning algorithm, comprising the following steps: (1) acquiring and preprocessing test data including soil layer input parameters, pile parameters and construction parameters, and dividing the preprocessed test data into a test set and a training set; (2) constructing a machine learning model in a pile foundation design stage according to the soil layer input parameters and the pile parameters, constructing a machine learning model in a pile foundation construction stage according to the soil layer input parameters, the pile parameters and the construction parameters, and respectively optimizing the two machine learning models by adopting a Bayesian optimization algorithm; and (3) the two optimized machine learning models are used for calculating the single-pile bearing capacity in the pile foundation design stage and the single-pile bearing capacity in the construction stage according to needs. According to the method, influence factors of all stages are comprehensively considered, algorithm learning is carried out on the influence factors, the bearing capacity of the reinforced concrete prefabricated pipe pile can be rapidly and effectively predicted, and the method can be used for optimizing the pile length design, reducing the pile material cost and improving the construction quality.

The invention provides an engineering test method and system for a storage chip and a medium, and belongs to the technical field of semiconductor packaging. The method comprises the following steps: acquiring read-write operation data, temperature data and voltage data of a storage chip in real time through a distributed sensor network, performing dimension reduction and segmentation processing on test data by using a multi-dimensional feature extraction algorithm, extracting statistical features and correlation features, inputting a fault prediction model constructed based on a deep learning algorithm, and performing fault prediction. Fault type and probability prediction is realized; and judging whether the chip has a fault by combining a preset threshold value, if the chip has the fault, accurately positioning a fault area by adopting a dynamic probe test technology, determining a fault time point and a trigger condition through time sequence analysis, and finally generating a fault diagnosis report and providing a repair suggestion. According to the invention, the accuracy and efficiency of fault detection of the storage chip can be effectively improved.

The invention relates to the technical field of hydrogen energy automobiles, in particular to a method for establishing a hydrogen leakage prediction model of a hydrogen energy automobile. According to the method, training, verification and test data sets are obtained through multi-modal data fusion and screening based on physical significance on source data fusing CFD simulation data and actually measured leakage data, multi-dimensional information is extracted from training data, time domain, frequency domain and physical field feature matrixes are obtained, a time sequence coding layer and a physical decision layer are constructed on the basis, and the time sequence coding layer and the physical decision layer are constructed. Therefore, fluid mechanics is integrated into a model structure, the model can explain and make decisions according to physical laws, the problem that fault attribution analysis conforming to the physical laws cannot be provided in the prior art is solved, a hybrid architecture model integrating data driving and physical constraint is constructed based on a two-layer structure, data set optimization is verified, and a test data set is adjusted. And finally, a hydrogen leakage monitoring model with high-precision prediction capability and reliable physical interpretability is established.

The invention relates to the technical field of data processing, and discloses an in-vehicle testing method and system based on a small model. The method comprises the following steps: acquiring an in-vehicle machine screen image, and extracting interface elements through a lightweight recognition model to generate a recognition result; analyzing element function association to determine operation types to form a test operation sequence; an ADB instruction and a mechanical arm action are generated according to the operation sequence to construct a test script; executing the script to drive the vehicle machine to test and collecting multi-dimensional response data; and analyzing test data, calculating a passing rate and response time, and generating a verification report. According to the method, the technical problems of low interface element identification accuracy, poor test script adaptability and insufficient multi-modal data analysis capability in the traditional vehicle-mounted terminal test are solved, and the automation degree of the vehicle-mounted terminal test and the reliability of the test result are remarkably improved.

The invention discloses a software test automation report generation optimization method which comprises the following steps: automatically registering and authenticating a multi-source test data access endpoint, uniformly collecting and normalizing test data with a multi-dimensional attribute tag, and forming a high-quality grouped data set through structured modeling and automatic clustering; further extracting event and result feature nodes, and establishing an interpretable tree rule dependency relationship; according to the method, the minimum redundancy and reusable customized rule path is generated by applying a dependency tree traversal and optimization algorithm and automatic reasoning, and the continuous adaptive evolution of the rule base among multiple projects is realized by combining conflict detection and visual interaction, so that the efficiency and expandability of data processing in a complex test scene are improved, and the test efficiency is improved. And the flexibility and automation level of report customization rule configuration are enhanced.

The invention discloses a test evaluation method and device for an industrial park integrated management system. The method comprises the following steps: acquiring a performance index data set of the industrial park integrated management system; the performance index data set comprises a performance index test data subset of each subsystem; the performance index test data subset comprises a test data sequence of each performance index; pre-processing the performance index data set to obtain a to-be-evaluated data set; and performing test evaluation processing on the to-be-evaluated data set to obtain a performance evaluation result value of the industrial park integrated management system.

The invention discloses a vehicle test data feature extraction method and system, and belongs to the field of vehicle test data processing, and the method comprises the steps: sampling a preset feature extraction strategy to extract time-frequency features from intermediate test data, and constructing cross features according to the time-frequency features; the preset feature extraction strategy comprises the step of extracting time domain features from the intermediate test data by using a first sliding window, and the window size of the first sliding window is adaptively adjusted according to the extracted data; after the time-frequency features and the cross features are scored through multiple preset evaluation methods, weighted summation is carried out on all scoring results, and a comprehensive score of each feature is obtained; based on the scene to which the features belong, adaptively adjusting the weight during weighted summation; and removing low-score features of which the comprehensive scores are lower than a preset score threshold to obtain high-score features, and performing dimension reduction on the high-score features to obtain optimized features. The accuracy of feature extraction is improved by dynamically adjusting the feature extraction window and the scoring standard.

The invention discloses an intelligent aeration control system and method for a sewage treatment plant, and relates to the technical field of sewage treatment control. Comprising a data acquisition module, an aeration test module, a demand prediction module, a normal state analysis module and an aeration control module, wherein the data acquisition module is used for acquiring real-time working data of a primary sedimentation tank, test data of an aeration tank, historical working data and real-time working data of the aeration tank and preprocessing the data; the technical key points are as follows: an aeration tank is scientifically divided into multiple sections, professional monitoring equipment is deployed in each section, multi-dimensional data is acquired, the data of each section is independently acquired and tested, the operation condition and the processing effect of the aeration equipment in each section can be accurately mastered, and on the basis of the accurate data and in combination with an aeration efficiency evaluation model and a regulation and control strategy, the aeration efficiency of the aeration tank is evaluated. And fine adjustment can be performed according to actual requirements of each section, so that resource waste or poor treatment effect caused by one-step regulation and control is avoided, efficient and accurate operation of the aeration equipment is realized, and the aeration equipment has a good use prospect.

The invention relates to a test case generation method and system based on an industrial agent. The method comprises the steps that an industrial data set is acquired, and a system dependency model is constructed; generating a path coverage test sequence through the first agent; generating a boundary value test data set through the second agent; generating abnormal test scene information through a third agent; and generating a test case set based on the path coverage test sequence, the boundary value test data set and the abnormal test scene information. In conclusion, by constructing the system dependency model and utilizing the multi-agent collaborative analysis process logic path, boundary condition and dependency relationship, the multi-surface coverage of the industrial control system test scene is realized, the automation of the test case generation is realized, the effects of improving the generation efficiency and enhancing the comprehensiveness of the test coverage are achieved, and the method is suitable for industrial control system test. Especially for the coverage of boundary conditions, abnormal scenes and concurrent paths, the manual dependence is reduced, and the reliability and consistency of the test are improved.

The invention discloses a multi-modal data driven commercial vehicle frame performance prediction method and system. The method comprises the following steps: uniformly coding design variables of a frame; constructing a multi-modal performance response data set based on finite element simulation and test results of mass production vehicle models, and realizing fusion of simulation and test data by adopting a maximum mean difference and related alignment algorithm; constructing a graph perception Transform multi-task prediction model fusing the structure topology and the physical position features, and predicting key performance indexes of the frame under a plurality of typical working conditions; through weighted multi-task loss function joint training, an uncertainty mechanism is introduced to dynamically adjust task weights; and after training is completed, deploying to an inference engine to realize second-level prediction and support increment fine adjustment updating. According to the method, repeated modeling and solving processes are avoided, the frame performance prediction efficiency and the adaptive capacity are remarkably improved, and the method is suitable for rapid evaluation of the frame performance of commercial vehicles of various structural configurations and material types.

The invention discloses a multi-dimensional intelligent analysis and fault traceability system for ACU final test data, and relates to the technical field of data analysis. The multi-source heterogeneous data acquisition module is used for acquiring operation parameters, function test data and assembly data through a CAN bus, test equipment and a production system, and is provided with a high-precision timestamp; the data cleaning and preprocessing module is used for standardizing data, filtering and denoising, synchronizing time and complementing missing values; the multi-dimensional feature extraction module is used for extracting features from a time domain, a frequency domain, a time sequence and a space; the intelligent fault diagnosis module is used for classifying faults by using a random forest and a D-S evidence theory and evaluating severity; the fault traceability reasoning module is used for constructing a fault propagation graph and positioning root causes through a Bayesian network; and the visual display and early warning module is used for generating a three-dimensional fault tree and a thermodynamic diagram and triggering graded early warning. According to the invention, the fault detection rate and traceability efficiency are improved, the false alarm rate is reduced, early warning is realized, the ACU test process is optimized, and the product quality and safety are guaranteed.

The invention provides a simulation test system simulation degree evaluation method and device, equipment, a storage medium and a product. The method comprises the steps of determining a test scene corresponding to a to-be-tested vehicle according to a design operation range of the to-be-tested vehicle; according to the test scene, determining a corresponding evaluation item, and respectively testing in the simulation test system and the actual environment to obtain simulation test data corresponding to the simulation test system and actual test data of the to-be-tested vehicle in the actual environment; based on the simulation test data, obtaining a simulation evaluation index corresponding to each evaluation item; based on the actual test data, obtaining an actual evaluation index corresponding to each evaluation item; and obtaining a simulation degree evaluation result of the simulation test system based on the simulation evaluation index and the actual evaluation index. According to the method, the simulation data and the actual data are compared according to the scene characteristics and the evaluation indexes related to the driving scene, and targeted evaluation results can be accurately obtained in different driving scenes.

The invention discloses an aerospace equipment data-real fusion performance test method and device based on reinforcement learning, and the device comprises a digital twin model building module which is used for building an aerospace equipment high-fidelity mechanism model fusing multi-physical field features, a lightweight digital twinborn body adaptive to reinforcement learning training is generated through a model order reduction technology; the digital twinborn model correction module is used for carrying out consistency calibration of multi-dimensional features and key physical parameter correction on the reduced-order model based on object reference data; the performance test module based on reinforcement learning actively explores a coupling scene of an extreme working condition and multiple faults in an interaction environment by constructing an attacker agent, and autonomously explores a performance boundary and a limit envelope of equipment; and the number-real fusion verification module is used for carrying out a hardware-in-the-loop point compensation test by taking the performance limit obtained by the digital test as guidance so as to assist in decision making and optimization of a test scheme. According to the method, the coverage, the safety and the test efficiency of the limit performance test of the aerospace equipment are improved.

The invention discloses a method and device for automatically generating a Java method unit test case based on AI, and the method comprises the steps: building a structured annotation specification, carrying out the automatic traversal and intelligent analysis of a Java project, extracting a method signature, a parameter type and a parameter constraint condition, constructing a test data generation model, and carrying out the automatic generation of the test case. The model is transmitted to an AI service to analyze business logic and boundary conditions, test case codes are generated, test cases are optimized and integrated to a project test directory, finally, code areas with insufficient coverage are identified through coverage rate analysis, a test completion strategy is generated, and iterative perfection of the test cases is completed. The technical problems that in the prior art, project-level automatic test case generation cannot be achieved, the generated test case cannot fully cover business logic and boundary conditions, and context relation understanding is lacked are solved, full-automatic project-level test generation is achieved, and the test efficiency and quality are improved.

The invention provides a light-weight aerial ship image target detection method and device based on yov8n and a storage medium. According to the scheme, the method comprises the following steps that an image data set of an aerial ship is obtained and preprocessed; the method comprises the following steps of: constructing a network model of the improved yov8n, replacing part of an original CBS module with a GAC module, replacing part of an original C2f module with a C2fADWC module, modifying a detection head and a loss function, and selecting a rotating frame OBB; the training set is input into the improved yolov8n-obb model to be trained; a trained yolov8n-obb improved model is obtained; and inputting the test data set into the trained improved lightweight yolov8n-obb model to obtain a ship detection and identification result. According to the method, the ship in the aerial ship image can be more accurately identified, the wharf and the ship can be better distinguished, and the detection precision is remarkably improved while extremely little parameter quantity is increased.

The invention provides a chip testing method and device, equipment, a storage medium and a program product. The method comprises the steps that operator codes, comprising multiple chip interface instructions, of a target operator are obtained, the target operator and a reference operator have the same calculation function, the chip interface instructions are used for achieving at least one hardware operation of a chip, and the hardware operations combining the multiple chip interface instructions are used for achieving the calculation function; analyzing the operator code to obtain a machine code file of the chip; test data are obtained, the test data and the machine code file are transmitted to chip simulation equipment, the chip simulation equipment is used for executing the machine code file on the test data, and a simulation calculation result of the target operator for the test data is obtained; obtaining a simulation calculation result from chip simulation equipment, and obtaining a reference calculation result of the reference operator for the test data; and generating a test result of the chip based on the simulation calculation result and the reference calculation result. According to the invention, the chip test progress can be accelerated.

The invention discloses a test data intelligent generation and privacy compliance desensitization method, which belongs to the technical field of electric digital data processing, and comprises the following steps: monitoring the structure change of a preset regulation update source and a target application system, and obtaining compliance dynamic information; analyzing the compliance dynamic information to generate a dynamic compliance rule base; based on the dynamic compliance rule base, generating an initial test data set through a preset intelligent data generator; desensitizing the initial test data set, and outputting a desensitized test data set; scanning the desensitized test data set in real time for compliance verification, and generating a compliance verification report; and dynamically optimizing the generation strategies of the dynamic compliance rule base and the intelligent data generator based on the compliance verification report. Dynamic rule base construction, intelligent separated data generation, closed-loop verification optimization and desensitization audit tracing technologies are adopted, laws and regulations and system changes can be automatically adapted, and the service validity, privacy security and law compliance of test data are synchronously guaranteed.

The invention discloses a method, a system and equipment for predicting the stability of integrated circuit test equipment and a medium, and belongs to the technical field of integrated circuit test. The method comprises the following steps: firstly, carrying out preprocessing and feature selection on historical data in an FT test stage, and adopting a Gaussian mixture model (GMM) to cluster and identify different operation condition clusters of a test machine; then, establishing a health state GMM reference for each working condition cluster, calculating a KL divergence value of a normal sample and the reference, and setting a dynamic anomaly detection threshold by 99.73% quantile of the KL divergence value; and finally, in real-time monitoring, calculating a KL divergence value of real-time data and a corresponding working condition cluster benchmark, and comparing the KL divergence value with a dynamic threshold value to realize accurate anomaly marking. The method effectively solves the problem of abnormal detection of the test data of the integrated circuit under complex and changeable working conditions, and improves the monitoring accuracy and working condition adaptability.