Improved PSO Algorithm Trained BP Neural Network: Application to Groundwater Table Prediction

Abstract

Groundwater table often shows complex nonlinear characteristic. Back Propagation (BP) neural network is increasingly used to predict groundwater table. But man-made selecting the structure of BP neural network has blindness and expends much time. In order to overcome shortcomings of traditional BP neural network, Particle Swarm Optimization (PSO) algorithm was adopted to automatically search BP neural network weight matrix and threshold matrix. At the same time, linear inertia weight and chaos variation operator are presented to improve traditional PSO algorithm searching capacity. Study case shows that, compared with groundwater level prediction model based on BP neural network, the new prediction model based on PSO and BP neural network can greatly improve the convergence speed and precision.