%A KUI Hai-lin, BAO Cui-zhu, LI Hong-xue, LI Ming-da %T Idling time prediction method based on least square support vector machine %0 Journal Article %D 2018 %J Journal of Jilin University(Engineering and Technology Edition) %R 10.13229/j.cnki.jdxbgxb20170904 %P 1360-1365 %V 48 %N 5 %U {http://xuebao.jlu.edu.cn/gxb/CN/abstract/article_13675.shtml} %8 2018-09-20 %X Engine start-stop system is widely used as it can reduce fuel consumption by turning off the engine while idling. But in traffic congestion conditions, the engine start-stop system increases fuel consumption because of turning off engine during short time idling. To solve this problem, Least Square Support Vector Machine (LS-SVM) is applied to predict the idling time, that the engine will not be turned off during short time idling. Experiment results show that the accuracy of idling time prediction using LS-SVM can reach 80%. The engine start-stop system using this method can reduce fuel consumption by about 0.2% during smooth hour. Meanwhile the driving comfort can be improved by reducing more than 50% start-stop numbers during rush hour and more than 15% during smooth hour.