文献类型：会议论文

英文题名：Development of Driving Cycle for Light Vehicle Based on the AP Clustering Method

作者：Chen, Jiarui[1]; Chen, Baoqin[1]

机构：[1] Faculty of Mathematics and Computer Science, Guangdong Ocean University, Zhanjiang, 524088, China

会议论文集：Computational and Experimental Simulations in Engineering - Proceedings of ICCES 2020

会议日期：January 6, 2021 - January 10, 2021

会议地点：Phuket, Thailand

语种：英文

外文关键词：Parameter estimation - K-means clustering

外文摘要：In order to establish the light vehicle driving cycle of Fuzhou city, a driving cycle construction method was proposed based on t-SNE combined with AP clustering algorithm, which can well reflects non-linear relationship between parameters and obtain more stable clustering effect. Based on the micro-trip theory, 2011 micro-fragments are identified from 496464 effective speed information and an evaluation criterion consisting of 15 characteristic parameters was founded. Then, t-SNE was utilized to reduce the dimensionality of characteristic parameters and all micro-fragments were classified into different categories by AP clustering. Finally, according to the relative error of six characteristic parameters and the SAFD, the target driving cycle is compared with the other driving cycle obtained by the method of principal component analysis combined with k-means clustering, which verified the accuracy of the proposed method. ? 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.