Design of PID Controller for Maglev System Based on an Improved PSO with Mixed Inertia Weight

A Maglev system was modeled by the exact feedback linearization to achieve two same linear subsystems. The proportional-integral-differential controllers (PID) based on particle swarm optimization (PSO) algorithm with four different inertia weights were then used to regulate both linear subsystems. These different inertia weights were Fixed Inertia Weight (FIW), Linear Descend Inertia Weight (LIW), Linear Differential Descend Inertia Weight (LDW), and mixed inertia weight (FIW–LIW-LDW). On the other hand, the parameters C of the PSO-PID controllers via mixed inertia weight (FIW– LIW-LDW) were optimized, the parameter values C in the electromagnet 1 and electromagnet 2 were both 0.4. Simulation results demonstrate that the control performance and robustness of PSO-PID based on mixed inertia weight (FIW–LIW-LDW) was superior to that of three PSO-PID controllers based on single inertia weights. For electromagnet 1, the overshoot of PSO-PID controller with mixed inertia weight reduced 3.36% than that of PSOPID controller with FIW, 5.81% than that of PSO-PID controller with LIW, and 6.34% than that of PSO-PID controller with LDW; for electromagnet 2, the overshoot of PSO-PID controller with mixed inertia weight reduced 1.07% than that of PSO-PID controller with FIW, 12.56% than that of PSO-PID controller with LIW, 7.97% than that of PSO-PID controller with LDW; the adjusting time of PSO-PID controller with mixed inertia weight reduced 0.395s than that of PSO-PID controller with FIW, 34.1s than that of PSO-PID controller with LIW, and 33.494s than that of PSO-PID controller with LDW.