PID controller tuning using particle filtering optimization

The PID controller is one of the most popular controllers, due to its remarkable effectiveness, simplicity of implementation and broad applicability. However, the conventional approach for parameter optimization in PID controller is easy to produce surge and big overshoot, and therefore heuristics optimization methods such as genetic algorithm (GA), particle swarm optimization (PSO) are employed to enhance the capability of traditional techniques. One major problem of these algorithms is that they may be trapped in the local optima of the objective and lead to poor performance. In this paper, a novel stochastic optimization technique named particle filter optimization (PFO) is proposed to achieve better performance in dealing with local optima while reduce the computation complexity of PID parameter tuning process. Simulation results indicate that the proposed algorithm is effective and efficient, and demonstrate that the proposed algorithm exhibits a significant performance improvement over several other benchmark methods.