@inproceedings{scholars11384,
            year = {2019},
             doi = {10.1063/1.5121128},
          volume = {2138},
            note = {cited By 2; Conference of 4th Innovation and Analytics Conference and Exhibition, IACE 2019 ; Conference Date: 25 March 2019 Through 28 March 2019; Conference Code:150892},
       publisher = {American Institute of Physics Inc.},
         journal = {AIP Conference Proceedings},
           title = {Online process monitoring using multiscale principal component analysis},
            issn = {0094243X},
          author = {Nawaz, M. and Maulud, A. S. and Zabiri, H.},
            isbn = {9780735418813},
             url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85071638860&doi=10.1063\%2f1.5121128&partnerID=40&md5=0c794244a14529cf8a5b7f6548a58470},
        abstract = {Fault detection and identification are challenging tasks in chemical processes, the aim of which is to decide out of control samples and find faulty sensors timely and effectively. Process data from chemical processes are highly correlated and generally have multiscale features. Multiscale process monitoring techniques based on wavelets have been regarded as powerful tools because these can efficiently separate deterministic and stochastic features. An online multiscale fault detection approach using principal component analysis (PCA) is proposed in this paper by introducing a moving window into traditional wavelet transform. Various windows in wavelet decomposition are used to determine the appropriate window size for model development. The results demonstrate that the approximation and the highest detail functions are adequate to detect the fault. The proposed approach performance is validated using simulated data from a continuously stirred tank reactor (CSTR) system. The proposed method shows a substantial improvement over conventional PCA and multiscale PCA. {\^A}{\copyright} 2019 Author(s).}
}