An adaptive neuro-fuzzy approach to control a distillation column

In this paper we use a control strategy that enhances a fuzzy controller wi th self-learning capability for achieving the control of a binary methanol- propanol distillation column. An adaptive-Network-based Fuzzy Interference System (ANFIS) architecture extended to cope with multivariate systems has been used. This allows the tuning of parameters both of the membership func tions and the consequents in a Sugeno-type interference system. To satisfy the control objectives the backpropagation gradient descent through the pla nt method is applied, hence identification of the plant dynamics is also ne eded. The performance of the resulting neuro-fuzzy controller under differe nt reference settings for the concentration of methanol demonstrates the st abilisation of the concentration profiles in the column leading to an effec tive methanol composition control.