地铁列车冲突不确定性的预测模型及验证方法

Prediction Model with Probabilistic Train Conflicts and the Verification Method for Metro Trains

地铁运营中,精准高效的列车间冲突预测是调度员行车调整的依据和基础。因此,精准高效地预测列车之间对行车资源请求的冲突一直是行业内持续研究的问题。列车实际运行状态的动态性导致列车间冲突的不确定性。针对目前普遍对于列车间冲突确定性的预测方式,引入地铁列车间冲突不确定性的预测方式,利用样本平均近似法(SAA)对列车运行状态的不确定性进行描述,继而表述列车之间对行车资源请求冲突的不确定性,然后提出一种列车间冲突不确定性的预测模型,并针对模型的预测结果设计验证其预测精度的方案。最后,利用南京地铁3号线列车运行实绩数据进行算例验证,验证结果表明模型预测结果具有较高的准确性。

An accurate and efficient prediction of conflicts between trains is the basis for dispatching in metro network.Therefore,predicting conflicts between trains over the request for operation resources accurately and efficiently has always been a constant research focus in the industry.The dynamic train running states lead to the probabilistic conflict between trains.In view of the commonly used method of predicting the certainty of conflicts between trains,this paper innovatively introduced a prediction method for the probabilistic conflicts between trains,and used the Sample Average Approximation(SAA)method to describe the dynamic train running state,thus describing the probabilistic train conflicts over the request for operation resources.Then,a prediction model for probabilistic train conflicts was proposed,with the scheme designed for verifying the accuracy of the prediction results.Finally,the study used the train record data of Nanjing Metro Line 3 for numerical example verification.The verification results show that the results of model prediction are of high accuracy.