Adaptive interval type-2 fuzzy neural network control of nonlinear systems with time-delay: direct, indirect and hybrid
Tsung-Chih Lin (Taiwan)

In engineering, information about the system is uncertain and imprecise. Moreover, many control systems such as aircraft, chemical or process control systems involve tine-delay either in the state, the input or the measurements. Time-delay deteriorates the system performance and is a source of instability, since delay systems are infinite dimensional in natural. Furthermore, due to type-1 fuzzy logic controller (FLC) cannot fully handle the linguistic uncertainties and the uncertainties associated with the use of noisy training data, type-2 FLC is proposed to accommodate such uncertainties. In this talk, I will present nonlinear systems with time-delay control using adaptive  interval type-2 fuzzy neural network (FNN) controller included direct, indirect and hybrid three categories: (1) The direct adaptive FNN controller uses fuzzy logic systems as controller in which linguistic fuzzy control rules can be directly incorporated into the controller. (2) The indirect adaptive FNN controller uses fuzzy descriptions to model the system plant. (3) The hybrid adaptive FNN controller is a combination of direct and indirect adaptive FNN controllers. A weighting factor, which can be adjusted by the tradeoff between plant knowledge and control knowledge, is adopted to sum together the control efforts from direct adaptive FNN controller and indirect adaptive FNN controller. A comparative analysis between adaptive  type-1 FNN control and adaptive  interval type-2 FNN control is given. Finally, the resulting adaptive interval type-2 FNN control systems show better performances, i.e., tracking error and control effort can be made smaller.