Ensemble prediction of monthly mean rainfall with a Particle Swarm Optimization-neural network model

A nonlinear statistical ensemble prediction modeling method has been developed for predicting monthly mean rainfall using Particle Swarm Optimization (PSO) algorithm and neural network (NN) technique. Comparison results of prediction experiments show that the PSO-NN ensemble prediction (PNNEP) model is superior to the traditional linear statistical forecast method in prediction capability. Computation and analysis of the PNNEP also demonstrate that the prediction of the ensemble model integrates predictions of dozens of ensemble members and the network structure of each member is objectively determined by means of PSO algorithm, so the generalization capacity of the ensemble prediction model is also enhanced, suggesting that the PNNEP model opens up a vast range of possibilities for operational weather prediction.