划分时段的公交车辆到站时间复合预测模型

Combination Forecasting Model of Bus Arrival Time by Time Division

在总结国内外现有的公交车辆到站时间预测模型的基础上,选取常用的三种经典预测模型,即Kalman滤波模型、BP神经网络模型、时间序列模型。分析并针对各个模型的特点及适用条件,对影响公交车辆到站时间的要素进行筛选,明确各模型的输入及输出条件,划分公交运行的高峰、平峰、低峰时段,并验证三种预测模型在各个时段的预测精度,进一步确定复合预测模型中各阶段预测方法 ,以实现公交车辆的到站时间的精确预测。

On the basis of the existing domestic and international bus arrival time prediction model,three classical prediction models are selected,namely Kalman filter model,BP neural network model and time series model.After analysising the characteristics and applicable conditions of each model,the factors affecting the bus arrival time are selected,the input and output conditions of each model are defined.Dividing the bus operation time into peak time,flat hump time and low peak time,and verifing the prediction accuracy of the three prediction models in each time period,to further determine the prediction methods of each stage in the combined prediction model.Thus,the accurate prediction of the arrival time of the bus is realized.