论文信息 - A Method of TE Process Fault Detection Based on Improved WNBC

A Method of TE Process Fault Detection Based on Improved WNBC

This paper propose a weighted Naive Bayesian classification (WNBC) fault detection method for industrial processes with simultaneous time-varying, multi-mode and complex data distribution. Estimated the weight of the algorithm by using the test data and selected the attributes according to the relationship between attributes. Based on the Tennessee Eastman(TE) data, this paper validates that the improved algorithm is effectiveness and practicability, which by designing different test scenarios and comparing them with similar methods in the literature.

Xingang Wang | Xiu Sun

[1] Leo H. Chiang,et al. Fault diagnosis in chemical processes using Fisher discriminant analysis, discriminant partial least squares, and principal component analysis , 2000 .

[2] Steven X. Ding,et al. An Integrated Design Framework of Fault-Tolerant Wireless Networked Control Systems for Industrial Automatic Control Applications , 2013, IEEE Transactions on Industrial Informatics.

[3] Chih-Jen Lin,et al. A Practical Guide to Support Vector Classication , 2008 .

[4] Shen Yin. Data driven design of fault diagnosis systems , 2012 .

[5] Mauricio Maestri,et al. Automatic qualitative trend simulation method for diagnosing faults in industrial processes , 2014, Comput. Chem. Eng..

[6] Ping Zhang,et al. A comparison study of basic data-driven fault diagnosis and process monitoring methods on the benchmark Tennessee Eastman process , 2012 .

[7] Jan Magott,et al. Timing analysis of safety properties using fault trees with time dependencies and timed state-charts , 2012, Reliab. Eng. Syst. Saf..