Selection of the PSO Parameters for Inverting of Ellipsometry

To guarantee the convergence of PSO applied to the inverting of ellipsometry to the global optima in good time, the optimization performance of PSO under different parameters is compared and analyzed. The result shows that the range of inertia weight ω from 0.5 to 0.8, the sum of learning parameters c₁, c₂ preferably not more than 3, and at the same time, a smaller c₁ and a bigger c₂, ensure the better optimization performance of PSO.