Applied Research of PSO in Parameter Estimation of Richards Model

Abstract

It's significant to establish a mathematical model for the spread of epidemic, to help control the epidemic situation and minimize their impacts. In this paper, Richards model is proposed to fit the spread and PSO is employed to estimate the parameters of Richards model. Meanwhile concave function decreasing strategy and linear decreasing strategy are adopted to update the particle's velocity inertia weights respectively, and a new object function in the sense of normalized cross-correlation is built. The experiment result indicates that, PSO is a valid method for the parameter estimation of Richards model.