DESIGN OF TSPID AND TS-PDC TYPE FUZZY CONTROLLER AND STABILIZATION OF MAGNETIC LEVITATION SYSTEM

Abstract

Magnetic Levitation System (MLS) is an unstable, nonlinear complex system. The basic aim of our research work is for suspending the steel ball without any mechanical support in desired position with the help of an efficient controller. This dissertation work is focused on how the Magnetic Levitation System works in presence of disturbance. For this task first of all we have designed the conventional Takagi —Sugeno type fuzzy controller. Uncertainty is an inherent part of intelligent systems used in real world applications. Conventional controllers cannot fully handle the uncertainties present in realtime systems. Conventional Takagi-Sugeno type fuzzy controller cannot give the satisfactory performance and cannot handle the uncertainties present in the Magnetic Levitation System. To overcome this problem we combine the two controllers, Proportional Integral Derivative controller (PID) and Takagi-Sugeno type fuzzy controller and designed Takagi-Sugeno PID type fuzzy controller. The TS-PID type fuzzy controller is designed in such a way that it's rule consequences is in the form of PID expressions. A custom membership functions are designed which are based on the parameters of the magnetic levitation system. The unique features of the TS-PID type fuzzy controller are as follows. First, the proportional, integral and derivative gains constantly vary with the output of the system under control. Here we use - negligible Integral gain. The gain variation leads to a shorter rise-time, a less overshoot and a smaller settling-time as compared to a comparable linear PID controller. Second, the characteristics of the gain variation are determined by the T-S fuzzy rules, and can intuitively be designed. By the use of singleton fuzzifier, the product inference and a center-average defuzzifier, we determine the global model of TS-PID type fuzzy controller. For the purpose of design and stability analysis of the TS-PID type fuzzy controller we use Lyapunov Stability theory. Stability for magnetic levitation system has been explored using Lyapunov Stability theory. The derived stability conditions contain smaller number of Lyapunov expressions. TS-PID type fuzzy controller is capable of modeling the uncertainty and precision in a better way. However, in real time application uncertainty associated with the available iv information is always occurs. The designed TS-PID type fuzzy controller is compared with the conventional Takagi-Sugeno type fuzzy controller and more robust against mass of ball, inductance and resistance variations. Simulation results shows that designed controller is fast with high degree of uncertainty handle capacity. Simulation results of TS-PID type fuzzy controller gives the better results as compared to conventional Takagi- Sugeno type fuzzy controller. Simulation results analyzed for both the controllers and validated in the real time model of the MLS.