Mixed Qualitative/Quantitative Dynamic Simulation of Processing Systems

In this article the methodology proposed by li and wang for mixed qualitative and quantitative modeling and simulation of temporal behavior of processing unit is reexamined and extended to more complex case. the main issue of their approach considers the multivariate statistics of principal component analysis (pca), along with clustered fuzzy digraphs and reasoning. the pca and fuzzy clustering provide tools to categorize the quantitative dynamic trends, describing the temporal behavior of joint human-process interactions qualitatively, and through the proposed neuro-fuzzy reasoning the system responses can be obtained when the system is exposed to uncertain disturbances. first, the method is applied to a continuous stirred tank reactor – cstr and then to a distillation column to demonstrate the accuracy level and capability of the approach to handle more complex processes.