The accurate mathematical model is an important tool for simulation and design analysis of fuel cell power systems. Semi-empirical models are easier to be obtained and can also be used to accurately predict the performance of fuel cell system for engineering applications. Particle swarm optimization (PSO) is a recently invented high-performance algorithm. In this paper, a parameter optimization technique of PEMFC semi-empirical models based on DKPSO was proposed in terms of the voltage-current characteristics. The simulated and experimental data confirmed the validity of the optimization technique, and indicated that PSO is an effective tool for optimizing the parameters of PEMFC models.