RB F 神经网络在公交行程时间预测中的应用

摘要

为提高公交行程时间预测结果准确度、减少预测过程花费时间，提出基于RB F神经网络公交行程时间预测模型。综合分析公交车辆行程时间动静态影响因素后，将网络变量输入模型对行程时间进行预测，以重庆市462公交线路为例，对模型进行验证，对比BP网络预测结果，表明RBF模型在速度和精确度上优于BP网络，具有一定实际应用价值。%In order to improve the accuracy of bus travel time prediction and reduce the prediction time ,a prediction model of bus travel time was puts forward based on RBF neural network ,analyzing the influence factors of bus travel time comprehensively ,then use the RBF neural network model to predict the travel time .Taking 462 bus line as an example in Chongqing City to verify the model ,contracting with prediction results of BP network and indicating that the RBF model is better than BP network in the speed and accuracy .And the model has practical application value .