Adaptive Type-2 Fuzzy Logic Control of Non-Linear Processes Bartolomeo Cosenza, Mosè Galluzzo* Dipartimento di Ingegneria Chimica dei Processi e dei Materiali, Università degli Studi di Palermo Viale delle Scienze, ed. 6, 90128 Palermo, Italy. Email:galluzzo@unipa.it The main objective of this study is to provide a valid and effective approach for the design and development of an adaptive type-2 fuzzy controller (AT2FLC), based on the analysis of the nonlinear process dynamics and the use of an ANFIS technique for the optimization of the controller. The performance of the obtained AT2FLC, characterized by a few number of rules, is higher than the performance of a traditional type-2 fuzzy controller with a larger rule base. The proposed controller is particurarly suitable for the control of processes characterized by uncertainty and time varying parameters. 1. Introduction In the last decades nonlinear control techniques have received considerable attention in the industrial process field, although all traditional approaches present many difficulties connected with the restrictive applicability conditions and the computational complexity. The heuristic approach used in the design of fuzzy logic controllers, built up making use of type-1 fuzzy sets, can be seen as an answer to the great complexity of traditional nonlinear control strategies in terms of robustness and effectiveness. Although over the past years many successfull fuzzy logic control applications for a number of complex and nonlinear processes have been reported, some difficulties of type-1 fuzzy logic controllers (FLCs) in minimizing the negative effects of uncertainties in the plant model parameters have come out. More recently a new generation of fuzzy controllers, type-2 FLCs (Mendel, 2001; Hagras, 2007; Castillo et al., 2008; Galluzzo et al. 2008), built up making use of type-2 fuzzy sets, characterized by a larger number of parameters and design freedom degrees, has shown to be able to handle uncertainties better than traditional type-1 fuzzy systems. However when processes are characterized by time varying parameters, as chemical industrial processes, simple type-2 FLCs may not be able to assure a lasting effective control. The variations of system parameters with time may deteriorate the control action and only the introduction of an adaptive mechanism able to modify the controller action, according to the actual system parameters, can make the control system more robust to parameter changes and to disturbances acting in the system. A procedure for designing adaptive type-2 FLCs has been developed. An ANFIS (Adaptive Neuro Fuzzy Inference System) technique is used to reduce the computational load of adaptive type-2 FLCs without losing the control efficiency. In