Design of Neuro Fuzzy Sliding Mode Controller for Active Magnetic Bearing Control System

Active Magnetic Bearing system (AMB) is a nonlinear mechatronic device utilized for levitat-ing the rotating components of a machine without physical contact with the stationary parts. The nonlinearityproperties pose a challenge for ensuring system stability. To achieve stable and optimal performance,controllers must be incorporated. A model of the proposed AMB has been derived and results a 2x2MIMO system. To simplify it, the model is linearized using Taylor series expansion technique. Simulationexperiments have been conducted using MATLAB. In this paper, a neuro-fuzzy sliding mode controller(NFSMC) has been designed to control the position of the AMB system. Simulink models of the AMBwith different control strategies have been developed for conducting simulation studies. The comparisonresults indicate that, with SMC, the position of the rotors follows the reference input but it is affected bychattering. It is further observed that FSMC reduces the chattering, while NFSMC controllers automaticallyeliminate the chattering problem and operate more robustly and smoothly. The system's stability has beenanalyzed using the Lyapunov theorem. According to the results gained, the NFSMC improves the overallsystem performance. The NFSMC controller yields an optimal result having no overshoot, a settling time of0.5 sec, and a rise time of 0.10 sec in comparison with FSMC controller which exhibits overshoot of 0.392%,a settling time of 0.75 sec, and a rise time of 0.133 sec, while the SMC shows overshoot of 9.047%, a settlingtime of 7.5 sec, and a rise time of 0.17 sec.