Aiming at poor control effect of Fractional Order Proportional-Integral-Derivative (FOPID) controller and the characteristics of wide range and high complexity of parameter tuning for FOPID controller, an improved Particle Swarm Optimization (PSO) method was proposed to optimize the parameters of FOPID controller. In the proposed algorithm, the upper and lower limits of inertial weight coefficients in PSO were defined and decreased nonlinearly with the iteration times in form of Gamma function, meanwhile, the inertia weight coefficients and learning factors of particles were dynamically adjusted according to the fitness value of particles, making the particles keep reasonable motion inertia and learning ability, and improving self-adaptive ability of the particles. Simulation experiments show that the improved PSO algorithm has faster convergence rate and higher convergence accuracy than the standard PSO algorithm in optimizing the parameters of FOPID controller, which makes the FOPID controller obtain better comprehensive performance.