Simulation Analysis of Non-Linear Fuzzy PID Temperature Controller

Abstract

The objectives of the project are to simulate linear Mamdami type fuzzy temperature controller and non-linear Takegi-Sugeno type fuzzy temperature controllers using MATLAB and Simulink, and to compare the performance between the two controllers. A case study has been created to test the controllers involved a water boiler, where the system is modeled using Joules Law and Law of Thermodynamics. A Proportional-Integral-Derivative (PID) controller was tuned and the PID parameters were then used to obtain the gain of the fuzzy controllers. Simulation results confirmed that non-linear fuzzy controller has smaller overshoot and faster settling time compared to the linear fuzzy controller and PID controller, although an extra derivative gain may be needed for the non-linear fuzzy controller if the integral term is huge enough to affect the stability of the system.