Internal model control with a fuzzy model: application to an air-conditioning system

Fuzzy models can represent highly nonlinear processes and can smoothly integrate a priori knowledge with information obtained from process data. A nonlinear controller can be designed by incorporating an inverted fuzzy model of the process in an internal model control (IMC) scheme. This paper presents an identification procedure for a Takagi-Sugeno fuzzy model, which is based on product-space fuzzy clustering. The obtained model can be inverted analytically and hence can be easily included in a nonlinear IMC scheme. The described method is applied to temperature control in an air-conditioning system. The performance is compared with the performance of a well-tuned PID controller.