COMPARATIVE ANALYSIS OF MAGNETIC LEVITATION SYSTEM USING DIFFERENT CONTROL TECHNIQUES

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 help of an efficient controller. This dissertation work is focused on how the Magnetic Levitation System works in presence of disturbance. The tracking performance and robustness is also checked for this system. For tracking, two type of reference trajectory are modelled. One is sine wave and other is a set of constant point varying at different levels. Lastly for robust performance, parameter variation is done in MLS. For this task we have designed Interval Type-2 Fuzzy Logic Controller (IT2FLC), Interval Type-2 Single Input Fuzzy Logic Controller (IT2SIFLC), Interval Type-2 Fuzzy Sliding Mode Controller (IT2FSMC) based on theory of type-2 fuzzy logic systems. Uncertainty is an inherent part of intelligent systems used in real world applications. Conventional controllers cannot fully handle the uncertainties present in real-time systems. Type-2 fuzzy sets that are used in type-2 fuzzy systems can handle such uncertainties in a better way because they provide us with more parameters. The type-2 fuzzy controller is designed in such a way that it can be implemented with signed distance method. With help of signed distance, interval type-2 fuzzy input variable in our simple fuzzy logic controller called type-2 single input FLC. Which has the advantage of the total number of rules are abruptly reduced compared to IT2FLC. The IT2FSMC is a combination of the interval type-2 fuzzy logic control and the sliding mode control which inherits the benefits of these two methods. The objective of the controller is to allow the system to move to the sliding surface and remain in on it. The IT2FSMC can reduce the complexity of analysis and implementation by using the sliding surface. Fuzzy logic based on interval value sets is capable of modelling the uncertainty and precision in a better way. However, in real time application uncertainty associated with the available information is always occurs. The designed controller's performance is compared with the feedback linearization control and more robust against mass of ball, inductance and resistance variations. Simulation results show that the designed controller is fast with high degree of uncertainty handling capacity. Simulation results analysed for all the controllers and validated in the real time model of the MLS.