Soft sensing modeling using neurofuzzy system based on rough set theory

Contributes to the development of soft sensor models combining the theory and methodology of data mining technology. Rough set theory, which can extract reduction rules out of data, is an efficient tool of data mining. In the paper, rough set theory is used to obtain the reduction rules, which are used as the fuzzy rules of the fuzzy system. Then the fuzzy system is represented via an equivalent artificial neural network (ANN). Because the initial parameter of the ANN is reasonable, the convergence of the ANN training is fast, and since the rules are reducts, the structure size of the ANN becomes small. The neurofuzzy approach based on rough set theory is used to build a soft sensor model for estimating the freezing point of the light diesel fuel in a fluid catalytic cracking unit.