为更精确地对港口或航道内船舶交通流量进行预测,分别建立BP神经网络预测模型和RBF神经网络预测模型进行仿真,并以宁波港船舶交通流量为例进行验证.结果表明,在宁波港现有发展基础和港口设施状况下,RBF神经网络用于宁波港船舶交通流量预测误差较小,预测值与实际值相近. In order to more accurately predict the traffic flow in the port or navigation channel, BP neural network prediction model and RBF neural network prediction model are respectively established for simulation and the traffic flow in Ningbo Port is taken as an example to verify the results. Under the condition of development foundation and port facilities, RBF neural network is used to predict the traffic flow of ships in Ningbo Port with less error and the predicted value is close to the actual value.