Premium
Scale‐sifting multiscale nonlinear process quality monitoring and fault detection
Author(s) -
Liu Yang,
Zhang Guoshan
Publication year - 2015
Publication title -
the canadian journal of chemical engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.404
H-Index - 67
eISSN - 1939-019X
pISSN - 0008-4034
DOI - 10.1002/cjce.22221
Subject(s) - benchmark (surveying) , sma* , computer science , fault detection and isolation , nonlinear system , process (computing) , scale (ratio) , kernel (algebra) , artificial intelligence , scale invariant feature transform , pattern recognition (psychology) , fault (geology) , key (lock) , data mining , feature extraction , algorithm , mathematics , actuator , physics , geodesy , quantum mechanics , combinatorics , seismology , geology , geography , operating system , computer security
We demonstrate a novel multiscale nonlinear process monitoring and fault detection method, called the scale‐sifting multiscale algorithm (SMA). The key innovative feature of SMA is essential scale data reconstruction without prior knowledge of signals monitored compared with state‐of‐the‐art multiscale monitoring methods. The SMA includes a scale‐sifting benchmark, data decomposition and data reconstruction, and dynamic kernel partial least squares. The scale‐sifting benchmark is developed to sift out special scales with the essential features of abnormal situations. Then, the data are reconstructed corresponding to selected scales. Finally, dynamic KPLS is applied to analyze data reconstructed for online quality process monitoring and fault detection. The application results illustrate the effectiveness of the proposed method