Simulation of neuro — Fuzzy controller for a flow process using MATLAB

Proportional integral derivative (PID) controllers are widely used in process control applications. For non linear systems such as flow process and multi-process control loops, PID controllers do not exhibit good performance. In this paper, a flow process loop is taken into consideration. Transfer function is derived by forcing the system to step signal and recording the reaction. Hence the second order system is approximated into a first order system with dead time. Design and simulation of a fuzzy controller is done based on the characteristics of the second order system. Response of the fuzzy controller is improved than the PID controller. Simulation of adaptive neuro-fuzzy controller is implemented for the flow process. The dynamic responses are better than fuzzy controller.