摘要
列车运行次序调整是列车运行调整的核心问题,一定程度上决定了干扰对行车的影响程度。列车运行次序调整判定模型,选取运行次序是否调整作为模型判定因子,基于列车运行实绩大数据,以朴素贝叶斯分类器和决策树作为机器学习预选模型。经训练集训练模型,测试集测试,得到判定值与真实值。通过混淆矩阵、准确率、ROC曲线验证,基于朴素贝叶斯的列车到发次序调整模型能够达到良好的判定效果,具有应用和研究价值。
The adjustment of train operation sequence is the core issue of train operation adjustment,which determines the degree of influence of the interference on the train operation to a certain extent.The train arrival and departure sequence adjustment determination model,select whether the arrival and departure sequence is adjusted as the model determination factor.Based on the train actual performance big data,naive Bayes classifier and decision tree are used as machine learning pre-selection models.After training the model on the training set and testing on the test set,the determination value and the true value are obtained.The determination model of train arrival and departure sequence adjustment based on Naive Bayes,verified by confusion matrix,accuracy rate,and ROC curve,can achieve good judgment results,and has application and research value.
作者
石晶
贾云光
杨明春
崔俊锋
岳朝鹏
SH Jing;JIA Yunguang;YANG Mingchun;CUI Junfeng;YUE Chaopeng(Beijing National Railway Research&Design Institute of Signal&Communication,Beijing 100073,China)
出处
《综合运输》
2021年第10期68-72,78,共6页
China Transportation Review
基金
基于北斗定位的列车运行智能控制系统研究(KSF202010)
新型列控系统-系统功能应用优化研究(2300-K1210003.01)。
关键词
运行次序调整
列车调整策略
朴素贝叶斯分类器
高速铁路
Arrival and departure sequence adjustment
Train adjustment strategy
Naive Bayes Classifier
High-speed railway
