Optimization of controller parameters for a DC attraction type levitation system by evolutionary technique

In this paper, evolutionary computation algorithms such as a novel Gravitational Search algorithm (GSA) and Real Coded Genetic Algorithm (RGA) are used to optimize the parameters of PID controller for a single actuator based DC attraction type levitation system (DCALS). A comparative study between performances of classical, GSA and RGA based nominal PID controller while operating at nominal and off-nominal (low and high) air-gaps has been carried out. It is seen that the GSA based optimized nominal PID controller produces best performances for the proposed DCALS with a wide variation of operating air-gaps.