基于DNN-NKDE的公交车驻站时间概率区间预测

Bus Station Time Probability Interval Prediction Based on DNN and NKDE

针对公交车在站台驻站时间的影响因素分析,考虑在复杂的乘车模式下引入驻站时间概率区间预测的概念。通过DNN算法构建驻站时间预测模型得到预测结果,统计预测值与真实值的误差集并划分为多个子区段,对每个子区段采用非参数核密度估计算法构建概率区间预测模型,估计重组各子区段在某置信水平下的公交车驻站时间概率区间预测结果。利用济宁市公交车到离站等信息数据,验证方法的有效性和可行性。

The influence factors of time for the bus at the station platform analysis,considering complex drive mode is introduced into the concept of probability interval prediction station time.The station time prediction model is established through algorithm within DNN forecasting results are obtained,the statistical error of the predicted values and the real value and multiple sub divided into sections,for each segment nonparametric kernel density estimation algorithm is used to construct the probability interval prediction model,estimate the reorganization of the predicted results of each section under a certain confidence level of the bus station time probability interval prediction results.Using the data from the jining city bus to leave the information such as the feasibility and effectiveness of the validation method.