Adaptive nonlinear PID controllers based on neurofuzzy networks

PID controllers are popular in industrial applications, as they are easy to install and reasonably robust. However, for highly nonlinear systems, the performance of PID controllers can deteriorate quite fast. It is necessary to develop nonlinear PID controllers for controlling nonlinear processes. An approach to design these controllers is to switch between several linear PID controllers using fuzzy logic based on the Takagi-Sugeno model. The nonlinear PID controllers derived here follows this approach. However, they are implemented based on B-spline neurofuzzy networks. Design guidelines and online training of the proposed controller are devised. The implementation and performance of the proposed controllers are illustrated by a simulated three-tank water level control system.