Forecasting the vessel traffic flow in the bridge areas is focused on this study. Based on Hybrid Radial Basis Function Neural Network, another novel predictive statistic modeling technique called Enhanced Hybrid Radial Basis Function Neural Network (EHRBF-NN) is proposed in the paper. EHRBF-NN is a flexible forecasting technique that integrates regression trees, particle swarm optimization, with radial basis function neural networks. In this technique, the regression tree is used to determine the centers and radius of the radial basis functions. The Particle Swarm Optimization (PSO) is used to avoid the over fitting and determine the weights of the neural network. Computer simulations have been implemented to validate the EHRBF-NN. Compared forecasting results with actual data, the algorithm of HRBF-NN is more effective than ordinary RBF-NN, RBF-NN with least square method and HRBF-NN, while it uses less computing resources and shorter computing time.
Ship traffic volume forecast in bridge area based on enhanced hybrid radial basis function neural networks
2015-06-01
402326 byte
Conference paper
Electronic Resource
English
Short-Term Freeway Traffic Volume Forecasting Using Radial Basis Function Neural Network
Transportation Research Record | 1998
|Short-Term Freeway Traffic Volume Forecasting Using Radial Basis Function Neural Network
Online Contents | 1998
|Vehicle tracking using radial basis function neural networks
Automotive engineering | 1996
|Vehicle Tracking using Radial Basis Function Neural Networks
British Library Conference Proceedings | 1996
|