RBF Neural Network Model Based on Improved PSO for Predicting River Runoff

Based on the observed river runoff data obtained from Yichang hydrological station in the middle of the Yangtze River, Radial Basis Function neural network (RBF) based on improved particle swarm optimization (PSO) was applied to predict river runoff in the Yangtze River. The capacity of solving nonlinear problems is enhanced effectively through adjusting inertia factor dynamically in the algorithm of particle swarm optimization. Improved PSO is applied to optimize the parameters of the neural network and overcome the over-fitting problem and a faster convergence rate is reached. MATLAB was applied to simulate the model. The theoretical analysis and simulations show that the prediction model is more practical and has better generalization performance and prediction accuracy than the traditional one.