Intelligent Train Operation Models via KNN and Adaboost.M1

Abstract

Traditional control algorithms in Automatic Train Operation (ATO) system have some drawbacks, such as high energy consumption and low comfort. In this paper, two Intelligent Train Operation (ITO) models based on data mining for the subway train control are proposed combining with the driving experience. Firstly the training data set is sorted out and sieved out from the real train operation data set by drivers in Yizhuang-line of Beijing subway to establish the standard database. By using KNN and Adaboost.M1 respectively, two ITO models are dug out to represent the output of controller with speed limit, running time and gradient. In the train control simulation platform, ITO model is compared with the traditional PID control algorithm of ATO. The results show that the proposed ITO models perform better than PID control algorithm on saving energy, increasing comfortable degree, and reducing number of controller's output change. Moreover, Adaboost.M1 is better than KNN especially on energy consumption and comfort.