Applying neural network with particle swarm optimization for energy requirement prediction

Prediction of energy requirement is an important research topic. For fulfilling such prediction, neural network (NN) has testified to be a cost-effective technique superior to traditional statistical methods. But their training usually with back-propagation (BP) algorithm or other gradient algorithms, and some problems are frequently encountered in the use of these algorithms. In this paper, particle swarm optimization (PSO) is proposed to train artificial neural networks (ANN), and as a result, a PSO-based neural network approach is presented. The approach is demonstrated by predicting energy requirement in Xipsilaan city in China. The results show that the proposed approach can effectively improve convergence speed and generalization ability of NN.