12 results about "Fault detection rate" patented technology

The application provides a process micro-fault detection method based on stationary subspace integrated probability slow features, and belongs to the technical field of industrial process fault detection. The technical scheme is as follows: the application is composed of an offline stage and an online stage. The application fuses the stationary information of SSA and the time sequence dynamics of SFA to form complementary feature description, finally realizes fine division of non-stationary multivariate data through integrated learning, realizes effective extraction of micro-fault probability information, and greatly improves the micro-fault detection rate.

The application discloses a strip steel hot rolling process multi-performance index abnormality monitoring method and system, and belongs to the technical field of industrial process control and fault diagnosis, and the method comprises the following steps: collecting process variable data and performance index data in a historical strip steel hot rolling process and performing standardization pretreatment; dividing the strip steel hot rolling process into multiple different process subblocks; for each process subblock, performing space-time feature extraction on the corresponding process variable data to obtain a space-time feature representation; constructing a local statistic quantity of each performance index; fusing the local statistic quantities to obtain a global statistic quantity and a global control limit; and based on this, realizing multi-performance index abnormality monitoring and alarm. The application effectively solves the problems that cross-process time lag dependence is difficult to capture, space-time feature extraction is not comprehensive, and single performance index monitoring leads to abnormality missed reports in the strip steel hot rolling process, significantly improves the fault detection rate and reduces the false alarm rate.

The application discloses a kind of lithium battery fault intelligent diagnosis method based on deep learning, comprising the following steps: collecting lithium battery operating data and carrying out quality correction, obtain operating data set;Input to operating condition perception model generates multi-scale feature vector representing different operating states;Initial fault diagnosis model is obtained by initial fault class label, initial fault probability value and diagnostic feature vector;Form uncertainty feature vector set by confidence evaluation;Perform bidirectional backtracking correlation analysis, extract associated time period operating data subset;Diagnosis is carried out again, generates review result and carries out consistency comparison with initial result;Conflict result generates conflict mode information and is stored to fault mode library, without conflict when directly output final diagnosis result;The diagnosis result of new data is corrected using fault mode library Pattern matching, output corrected final diagnosis result.The application can significantly improve fault detection rate, reduce misjudgment rate and shorten diagnosis duration.

The application discloses an industrial process fault detection method and system based on double subspace division, which firstly performs normality evaluation on process variables, divides a data space into Gaussian subspace and non-Gaussian subspace; a fault detection model based on principal component analysis is established in the Gaussian subspace, and a fault detection model based on support vector data description is established in the non-Gaussian subspace; current test data variables are divided into the Gaussian subspace and the non-Gaussian subspace, and are respectively input into the trained models, so that fault detection statistics corresponding to each subspace are calculated; each subspace statistic is converted into a fault probability by using Bayesian inference and is weightedly fused to construct a comprehensive monitoring statistic, and fault detection is realized. The application effectively solves the problem that a traditional method has poor adaptability to Gaussian and non-Gaussian mixed distribution data, improves the fault detection rate while significantly reducing the false alarm rate, and is suitable for real-time monitoring requirements of complex industrial processes.

This invention discloses a health monitoring system and method for construction machinery equipment. The system includes: a perception layer; an edge computing gateway, including a dynamic self-configuration module, a working condition perception engine, an intelligent hierarchical transmission module, and a local data buffer module; a network layer, including a multi-link adaptive routing module and a protocol adaptation and encryption module; a platform layer, including an IoT access and service module, a hybrid storage module, and a computing engine layer; and an application layer, including an equipment profiling module, a transfer learning prediction module, a knowledge graph-enhanced diagnostic module, and a visualization and interaction module. This invention achieves plug-and-play sensor functionality through dynamic self-configuration, improves fault detection rates through working condition adaptive edge computing, reduces communication costs through intelligent hierarchical transmission, shortens model deployment cycles through equipment profiling transfer learning, and reduces false alarm rates through knowledge graph-enhanced diagnostics, thus realizing full lifecycle health monitoring of construction machinery equipment.

This invention provides a method for detecting minor process faults based on sliding window shared dictionary learning, belonging to the field of data-driven technology. It solves the technical problems of domain shift in existing dictionary learning methods and insufficient sensitivity of the KNN algorithm for minor fault detection. The technical solution includes the following steps: acquiring multivariate time series data of an industrial process; extracting the mean and variance statistical features from the sliding window; constructing a shared dictionary learning dataset; learning the shared dictionary using a clustering algorithm; calculating the reconstruction error; setting adaptive detection control limits; and performing minor fault detection and evaluation. The proposed method has been applied to the detection of several different types of minor faults in the Eastman Chemical Company in Tennessee. Simulation results show that, compared with principal component analysis and the KNN method, the proposed shared dictionary learning method has a higher fault detection rate and a lower false alarm rate.

The present application belongs to the technical field of industrial equipment operation state monitoring, and discloses a dust removal fan fault prediction system based on a time series prediction model, which comprises a multi-source sensing module, a data preprocessing module, an edge computing module, an early warning output module and a feedback optimization module. The multi-source sensing module collects mechanical, performance and environmental multi-dimensional parameters in a distributed manner. The data preprocessing module realizes data synchronization optimization through a DTW algorithm and a filter circuit. The edge computing module takes an LSTM-ARIMA hybrid architecture as the core and combines a gate weight adjustment circuit to adapt to working condition fluctuations. Precise positioning is realized through Grad-CAM visualization and a fault mode library. The early warning output module realizes hierarchical alarm and equipment linkage. The feedback optimization module ensures the long-term adaptability of the device. The system solves the problems of single monitoring dimension, high false alarm rate and response lag in the prior art, has small wind pressure difference prediction error, high filter bag damage fault detection rate and low false alarm rate, can greatly reduce filter material loss, and is suitable for high dust industrial scenes.

The application discloses an Android application testing method based on Monte Carlo tree search and deep reinforcement learning, and belongs to the technical field of software testing.The method adopts deep reinforcement learning to perform adaptive learning on a testing strategy, a fine-grained state representation mode is designed for an application program interface state, a new reward function is used during exploration, state changes caused by interface jumps can be rewarded, and fine-grained control position and text changes in the interface can also be rewarded, so that the learned testing strategy is more comprehensive and detailed, the Monte Carlo tree search method is used to optimize the testing strategy, potential states are provided with opportunities for long-term exploration, and local optimization is avoided.The method can continuously reach some application program states that are difficult to traverse previously, high-code-coverage Android application program testing is realized, and the code coverage and fault detection rate performance of Android application testing are improved.

This invention discloses a sampling and evaluation method for equipment testability testing based on three-dimensional fault data fusion, belonging to the technical field of equipment testability testing and evaluation. It includes: acquiring three-dimensional evaluation data for all fault modes of the equipment; normalizing the data to obtain a normalized three-dimensional feature vector for each fault mode; performing weighted fusion using weight coefficients to obtain initial fusion weights; revising the lower limit of the initial fusion weights to obtain final fusion weights; obtaining the three-dimensional fusion weighted sampling probability for each fault mode after normalization; calculating the total sample size of the test; performing PPS sampling allocation using forced sample position allocation and low-difference sequences respectively; conducting physical fault injection tests based on sample allocation; calculating the unbiased estimate of the fault detection rate and the one-sided confidence lower limit of the fault detection rate; and completing the equipment testability evaluation. This invention can improve the scientific rigor of equipment testability evaluation.

The application discloses a kind of incremental feature decoupling self-encoders' core temperature anomaly detection method.The application is aimed at the problem that traditional decoupling feature learning is difficult to specify feature dimension in advance, designs a kind of feature increment strategy, and its hidden space feature is generated step by step when self-encoder model is trained and feature dimension is adaptively determined.Meanwhile, an iterative training strategy based on double performance indicators is proposed for model training, so that the features extracted by the self-encoder model have strong reconstruction ability for data and meet the decoupling requirements of hidden space features.Finally, the feature space and residual space of the core temperature data are described using statistical quantities, and comprehensive anomaly detection of the reactor core temperature is realized.The method can effectively reduce the false alarm rate of faults and improve the fault detection rate in the anomaly detection task of multiple measurement point temperature data of the nuclear reactor core, providing practical help for the safe and stable operation and intelligent operation and maintenance of the nuclear reactor.

The application discloses a kind of multi-die interconnection integrity test optimization method and system, it is related to die test technical field, including the weighted connection structure model of CTE mismatch, aspect ratio and hollow density correction is constructed, four kinds of failure of sensitivity model are established to thermal fatigue, TSV leakage, bonding hollow and electromigration, risk index is calculated and three-level coverage baseline is divided by fusing FMEA and safety level, compressed test vector is generated according to clock domain clustering, dynamic rerouting is realized through IEEE1838 hierarchical TAP, online response freezes and extracts adjacent line to form fault cluster, execute hardware level isolation and directional diagnosis, update weight based on fault result, monitor process drift and dynamically adjust threshold, offline update topology graph forms closed loop iteration.The application realizes cross-die differentiation test strategy, improves test efficiency and fault detection rate.