Moth-Flame Optimization-Based Fuzzy-PID Controller for Optimal Control of Active Magnetic Bearing System

摘要

In this paper, the application of moth-flame optimization (MFO) algorithm for the optimal control of an active magnetic bearing (AMB) system is studied. Active magnetic bearings are known to be highly nonlinear multivariable systems. An AMB system is used in motors, generators, turbines and various other machineries in different industries to provide an active suspension to the rotor shafts. The comparison of controlled responses of various closed-loop systems resulting from the use of conventional proportional-integral-derivative (PID) and fuzzy logic-based intelligent control strategies is discussed. The heuristic MFO algorithm is applied to optimize the scaling factors of the fuzzy-PID controller. The proposed controller performance is superior as compared to the very famous industrial PID and fuzzy-PID controllers with respect to various time response parameters. A comparative study with other heuristic algorithms such as PSO and SA is also performed. Three performance indices, namely integral square error, integral time absolute of error (ITAE) and integral time absolute of error plus Integral time absolute of control action (ITAE + ITAU), are chosen for designing the optimization problem.