船舶交通流量预测的灰色神经网络模型

摘要

In order to reduce the error and improve the accuracy of ship traffic flow prediction,a Grey neural network model is constructed based on the analysis of the advantages and disadvantages of the tra-ditional Grey model and BackPropagation (BP)neural network model. The real measured data are used as initial data to construct different Grey models. Various prediction results of these models are used as the input of the neural network,and then the optimized prediction model is obtained. A case study shows that the Grey neural network model can improve prediction accuracy,is of good prediction results and better than the single prediction model. The model requires less initial data,is of strong nonlinear fitting ability,and is feasible and effective for the ship traffic flow prediction.%为降低船舶交通流量的预测误差，提高预测精度，在分析传统的灰色模型和反向传播（Back-Propagation，BP）神经网络模型优缺点的基础上，构建灰色神经网络模型预测船舶交通流量。以实际测量值作为初始数据构建不同的灰色模型，各种灰色模型的预测值作为神经网络的输入值，得到最佳预测模型。实例分析表明：灰色神经网络模型可提高预测精度，预测结果比较理想，优于单一预测模型；该模型具有所需初始数据少和非线性拟合能力强的特点，用于船舶交通流量预测是可行和有效的。