首页|基于大数据驱动的山地环境纯电动汽车运行工况

基于大数据驱动的山地环境纯电动汽车运行工况

扫码查看
针对纯电动汽车日益增多趋势,采用传统燃油车工况的方式已经不再适用于纯电动汽车工况研究.为获取专用于纯电动汽车的行驶工况并从数据的角度将其与传统工况进行比较,首先,根据重庆地区的纯电动汽车行驶数据,利用专有的大数据平台的整车和电池数据,采用了改进的短行程法,筛选出运动片段;然后,采用特征工程的相关方法构造运动片段特征,对高维特征数据采用主成分分析方法进行数据降维并计算特征权重,以消除特征的相关性影响;之后,采用K-means++聚类方法对行车速度和电池电流曲线进行结构划分,从而构建低速、中速、中高速、高速 4 种类型的工况类型.接下来,通过对每一类工况以聚类中心点为原点进行距离排序,筛选出最合适的短时工况,以数据集中各类工况的总时长与数据集总时长的比值作为各类工况的权重,采用加权组合的方式对工况进行拼接;然后通过误差分析,选取特征偏离误差最小的一条工况曲线作为代表工况,从而确定整车和电池的代表工况.最后,通过对比国际国内现有典型工况,证明了在山地环境下纯电动汽车工况曲线的可靠性.结果表明:与非山地环境相比,山地环境下纯电动汽车匀速行驶时间更短,怠速行驶时间更长,减速度更大,减速时间占比更小;电池放电效能更高;车速保持在适中水平.
Battery Electric Vehicle's Driving Cycle in Mountain Environment Based on Big Data
In view of the trend of battery electric vehicles increasing,it is no longer applicable to use traditional fuel vehicle's driving cycle for battery electric vehicle's driving cycle research.In order to obtain the driving cycle of the battery electric vehicle and compare with the traditional driving cycle from the data point of view,firstly,according to the driving data of the battery electric vehicle in Chongqing area,by using the whole vehicle and battery data of the proprietary big data platform,the short-stroke method is adopted to screen out the kinematic fragments.Then,the feature of kinematic fragments is constructed by the correlation method of feature engineering,and the dimension of high-dimensional feature data is reduced by principal component analysis and the feature weight is calculated to eliminate the influence of feature correlation.Next,K-means+ + clustering method is used to partition the structure of the driving speed and battery current curves,and thus to construct 4 types of working conditions of low speed,medium speed,medium-high speed and high speed.Moreover,the most suitable short-time working conditions are selected by sorting the distance of each working condition with the cluster center as the origin,the ratio of the total duration of the data set to the total duration of the data set is taken as the weight of the data set.Then,through the error analysis,the characteristic deviation error of the smallest curve is selected as a representative of the working condition,so as to determine the representative of the vehicle and battery working conditions.Finally,the reliability of battery electric vehicle operating curve in mountain environment is proved by comparing the international and domestic typical operating conditions.The result shows that(1)compared to non-mountainous environments,in mountain environment,pure electric vehicles have shorter constant speed driving times,longer idling times,greater deceleration and a smaller proportion of deceleration time;(2)battery discharge efficiency is higher,and the vehicle speed is maintained at a moderate level.

automotive engineeringdriving cycleprincipal component analysiskinematic fragmentK-means++ clustering analysisbattery electric vehiclemountain environment

徐婷婷、龙方家、胡晓锐、朱蜀江

展开 >

国网重庆市电力公司,重庆 400015

中国汽车工程研究院股份有限公司, 重庆 401122

汽车工程 行驶工况 主成分分析 运动学片段 K-means++聚类分析 纯电动汽车 山地环境

国网重庆市电力公司科技项目

5220002000C0

2024

公路交通科技
交通运输部公路科学研究院

公路交通科技

CSTPCD北大核心
影响因子:1.007
ISSN:1002-0268
年,卷(期):2024.41(1)
  • 14