@article{scholars12606,
            year = {2020},
       publisher = {American Chemical Society},
         journal = {Industrial and Engineering Chemistry Research},
           pages = {18595--18606},
          number = {41},
            note = {cited By 15},
          volume = {59},
             doi = {10.1021/acs.iecr.0c02288},
           title = {Multiscale framework for real-time process monitoring of nonlinear chemical process systems},
        keywords = {Accident prevention; Chemical industry; Error statistics; Fault detection; Monitoring; Optimization; Process control; Process monitoring; Wavelet transforms, Chemical process systems; Continuous stirred tank reactor; Kernel principal component analyses (KPCA); Monitoring techniques; Multi-scale frameworks; Process disturbances; Real-time process monitoring; Squared prediction errors, Real time systems},
             url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85096033121&doi=10.1021\%2facs.iecr.0c02288&partnerID=40&md5=826402f230faa0d32700b43f35dbb54e},
        abstract = {Process monitoring techniques are used in the chemical industry to improve both product quality and plant safety. In chemical process systems, real-time process monitoring is one of the most crucial and challenging tasks for efficient quality control of the final products and process optimization. The existing multiscale process monitoring techniques use offline decomposition tools that restrict their applications to real-time process monitoring. In this study, to improve the performance of monitoring real-time process data, we have combined moving window-based wavelet transform and kernel principal component analysis (KPCA). A case study is performed on a typical continuous stirred tank reactor system. Performance analysis (based on T2 and squared prediction error statistics and contribution plots) shows that the technique successfully detects and identifies process disturbances, sensor bias, and process faults. Moreover, a comparison with PCA and KPCA methods shows that the proposed approach provides a 100 fault detection rate for the step-change fault patterns and has considerably improved detection rates for the random and ramp-change fault patterns. {\^A}{\copyright} 2020 American Chemical Society},
            issn = {08885885},
          author = {Maulud, A. S. and Nawaz, M. and Zabiri, H. and Suleman, H. and Tufa, L. D.}
}