Nonlinear identification of a batch reflux ratio controlled distillation column

• Published in: ARPN Journal of Engineering and Applied Sciences
• Main Author: Rustam I.; Wan W.A.K.; Ismail M.S.N.; Ali M.A.; Sidek M.N.
• Format: Article
• Language: English
• Published: Asian Research Publishing Network 2017
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85020075747&partnerID=40&md5=da97b574c7ff465620d8ee0182c8a59e

An attempt towards representation of a nonlinear process dynamic through the use of empirical NARX-OLS approach is discussed in this paper. Structured data mining was incipiently conducted to provide exposure over the process properties elucidated by the plant taken from a MISO control perspectives that is allocated in accordance to various sets of reflux ratio operation. Establishment over the process dynamic identification is made possible via OLS model structure and QR factorization parameter selection technique. An optimization of the resultant finding was then implemented through ERR optimization for comparison study. A good comparable yielded result from the model estimators insinuated the possibility of constructing a well-defined empirical dynamic model through a structured data mining processes without a priori knowledge of the system.