A FUZZY INFERENCE SYSTEM BASED ON OPTIMAL FUZZY CLUSTER AND ITS APPLICATION IN PRODUCT QUALITY ESTIMATION

A fuzzy inference model based on optimal fuzzy clustering in the input space has been established. By combining model structure analysis based on singular value decomposition and fuzzy statistical information criteria, an optimal fuzzy cluster in the input space can be realized, and an optimal fuzzy inference system is attained. This model has been successfully applied to modeling jet fuel endpoint estimator for a fractionator of the hydrocracking unit in the oil refining industry.