Online Multivariable Identification of a MIMO Distillation Column Using Evolving Takagi-Sugeno Fuzzy Model

In this paper, an evolving Takagi-Sugeno (eTS) fuzzy model has been utilized for online identification of a multi-input, multi-output (MIMO) distillation column. In this approach, the rule-base structure and the model parameters of the consequent parts of fuzzy IF-THEN rules gradually evolve during the online identification process. In addition, an exponential time-varying weight is included in the original rule generation condition in order to control the rate of rule generation at the start of the training process and consequently reduce the total number of generated rules in comparison with the original MIMO eTS algorithm. Recursive-Least Squares (RLS) algorithm is employed to estimate the consequent part of each rule. The results show that the modified condition reduces the total number of generated rules for a certain data set with lower RMSE error in comparison with the original eTS method. ,