A Novel Fuzzy Neural Network Controller for Maglev System with Controlled-PM Electromagnets

This chapter proposed an intelligent control method for the positioning of a hybrid magnetic levitation (Maglev) system, using the emerging approaches of fuzzy logic and artificial neural network (ANN). A Maglev system depends on controlling the air gap of the electromagnetic actuator. In practice, no precise mathematical model can be established because this hybrid Maglev system is inherently unstable in the direction of levitation, and the relationships between current and electromagnetic force are highly nonlinear. Fuzzy logic has emerged as a mathematical tool to deal with the uncertainties in human perception and reasoning. It also provides a framework for an inference mechanism that allows for approximate human reasoning capabilities to be applied to knowledge-based systems. Moreover, ANNs have emerged as fast computation tools with learning and adaptive capabilities. Recently, these two fields have been integrated into a new emerging technology called fuzzy neural networks (FNN) which combine the benefits of each field. In the method that is proposed herein, the control model uses Takagi-Sugeno fuzzy logic, in which the back-propagation algorithm processes information from neural networks to make suitable adjustments to the parameter of the fuzzy logic controller (FLC) and the control signal for object output tracking of the input. This method can then be applied to improve the control performance of nonlinear systems. System responses transient performance and steady-state performance various processes that are by using a FNN that must be trained through a learning process, to yield suitable membership functions and weightings. The result on the Maglev system of a simulation indicates that the system response satisfies the control performance without overshoot, zero-error steady state, and obtaining the rise time within 0.1 s. The proposed controller can be feasibly applied to Maglev systems with various external disturbances, and the effectiveness of the FNN with self-learning and self-improving capacities is proven.