基于IC卡和RBF神经网络的短时公交客流量预测

A Short-Term Public Transit Volume Forecasting Model Based on IC Card and RBF Neural Network

在公交客流量特性分析基础上,通过IC卡获取了实时公交客流量数据;结合GPS数据,利用OD反推法分析了实时客流分布;进而建立了基于IC卡和RBF神经网络的短时公交客流量预测模型并介绍了具体预测过程。对重庆市841公交线路进行了实例分析,得到上下车客流真实值与预测值的平均绝对相对误差均小于1.5%,实例计算结果表明该模型能获取实时客流数据,预测精度高,具有一定的实际应用价值。

On the base of the analysis on characteristics of bus passenger volume, the real-time data of public transit volume was obtained by IC card. Combining with GPS data, the real-time distribution of passenger volume was analyzed by OD back-stepping method. And then the forecasting model of short-term public transit volume based on IC card and RBF neural network was established, meanwhile, the specific forecasting process was also introduced. No. 841 bus route in Chongqing was taken as an example to verify the proposed forecasting model. It is found that the average absolute relative error of the real value and the predicted value of the passenger flow is less than 1.5%. The results of ease study show that the proposed model can obtain real-time traffic data and achieve high prediction accuracy, which has certain practical application value.