A Short-term Traffic Forecasting Model Based on Wavelet Neural Network with Novel Teaching Learning Based Optimization

Short-term traffic flow prediction is of vital significance to traffic monitoring and induction. It is a nonlinear problem with considerable uncertainty. In this paper, a novel teaching-learning based optimization (NTLBO) algorithm is proposed for wavelet neural network, whose parameters are optimized by NTLBO. Parameters of the network are divided into four parts to initialize the population of the novel algorithm. The four parts of parameters are optimized separately in teaching process. A self-learning process for teacher is attached as well to combine local search strategy and global search method in network training, which enhance the search ability of the algorithm. Finally, the model will be verified by time-series data, compared with other neural network methods. The final experimental results indicates that the new algorithm performs better in forecasting and nonlinear curve fitting.