data-driven fault detection technique has been widely applied for process monitoring, which can effectively detect faults happened in industrial processes. It is extremely significant fir guaranteeing the normal opera...
详细信息
ISBN:
(纸本)9781538635247
data-driven fault detection technique has been widely applied for process monitoring, which can effectively detect faults happened in industrial processes. It is extremely significant fir guaranteeing the normal operation of processes. Independent Component Analysis (ICA), a type of data-driven fault detection technique, has been successfully applied to Blind Source Separation and process faultdetection. There exist two classical methods, Gradient and Newton Iteration Methods, to determine independent components. ICA based on Newton Iteration Method has been well investigated and successfully applied into practice. However, ICA based on Gradient. Method gains less attention. In order to give a comparison between Gradient and Newton Iteration Method based ICA in process monitoring, principles of both methods are demonstrated in detail. Moreover, numerical simulations are finished in the platform of Tennessee Eastman(TE) process. Results show that efficiency of process monitoring of ICA based on Gradient Method is better for TE process.
In this paper the telemetry data of Satellite TX-I are analyzed in order to have a better understanding of the satellite operating status,and to lay the foundation for faultdetection *** the high dimensional data,the...
详细信息
ISBN:
(纸本)9781479946983
In this paper the telemetry data of Satellite TX-I are analyzed in order to have a better understanding of the satellite operating status,and to lay the foundation for faultdetection *** the high dimensional data,the locally linear embedding (LLE),a kind of manifold learning schemes,is applied to perform dimensionality reduction and feature *** the data-driven fault detection can be effectively implemented by means of the statistic indexes T 2 and *** results presented in the paper demonstrate that not only the data processing,like feature extraction,but the faultdetection scheme is effective.
Tools of sensor-data-driven anomaly detection facilitate condition monitoring of dynamical systems especially if the physics-based models are either inadequate or unavailable. Along this line, symbolic dynamic filteri...
详细信息
Tools of sensor-data-driven anomaly detection facilitate condition monitoring of dynamical systems especially if the physics-based models are either inadequate or unavailable. Along this line, symbolic dynamic filtering (SDF) has been reported in literature as a real-time data-driven tool of feature extraction for pattern identification from sensor time series. However, an inherent difficulty for a data-driven tool is that the quality of detection may drastically suffer in the event of sensor degradation. This paper proposes an anomaly detection algorithm for condition monitoring of nuclear power plants, where symbolic feature extraction and the associated pattern classification are optimized by appropriate partitioning of (possibly noise-contaminated) sensor time series. In this process, the system anomaly signatures are identified by masking the sensor degradation signatures. The proposed anomaly detection methodology is validated on the International Reactor Innovative & Secure (IRIS) simulator of nuclear power plants, and its performance is evaluated by comparison with that of principal component analysis (PCA).
暂无评论