Application of Particle Swarm Optimization to Minimize Active Magnetic Bearing Forces

Abstract

Active Magnetic bearings (AMB) are widely used in the rotating machine. Its dynamic behavior is impacted by the AMB parameters. That’s why, a good selection of the bearing parameters is important to achieve an accurate performance of the rotating machine. This work attempts application of Particle Swarm Optimization (PSO) is applied to select the optimal parameters of bearing leading to minimum bearing forces. Bearing forces are depends on several parameters like rotational speed, control-current, derivative gain and number of windings around the core half. One objective function, three limited parameters and one material constraints are considered in the optimization processes. A rotating composite shaft mounted by two AMB is studied. The rotor motion equation has been established and numerically solved by combining Newmark and Newton-Raphson methods. Results show that bearing parameters affect the bearing force value and the dynamic behavior of the rotor.