KDE based output PDF control for nonlinear non-Gaussian systems using PSO algorithm

This paper considers a simplified PDF control strategy for non-linear non-Gaussian stochastic distribution systems. The objective of this approach is still to make the output probability density function track a desired PDF. The output PDF, which is assumed to be immeasurable, is estimated according to kernel density estimation by analyzing the output sequence of the system. The performance index is constructed with respect to the control objective and is minimized based on particle swarm optimization. A PDF tracking controller is then designed and a simulation is finally given to demonstrate the effectiveness and efficiency of the proposed algorithm.