摘要
针对因行驶工况与驾驶风格对剩余续驶里程有显著影响而导致剩余里程估算困难的问题,采用工况识别与驾驶风格识别相结合的方式对电动汽车续驶里程进行估算。通过主成分分析和聚类分析,选出基于大数据、符合郑州市当地交通特征的典型行驶工况片段,并利用选取的典型行驶工况片段,在MATLAB/Simulink下建立主成分分析和学习矢量量化神经网络相结合的工况识别模型、驾驶风格模糊识别模型;通过联合仿真进行剩余续驶里程实时估算,并通过卡尔曼滤波对输出结果进一步优化。仿真分析及半实物测试结果表明,采用融合车、路、人信息的电动汽车续驶里程估算方法,不仅降低了剩余里程估算误差,同时也证实了方法的可行性。
Aiming at driving cycle and driving style influenced significantly on remaining range esti-mation,which resulted in the remaining range estimation difficulty,a estimate method was adopted,which combined driving cycle with driver’s driving style,to estimation the driving range of electric vehi-cles.The typical driving condition fragments were selected based on large sample and traffic characteris-tics of Zhengzhou,which through principal component analysis and fuzzy C-means clustering technique.Under MATLAB/Simulink,a model to identify driving patterns was established based on principal com-ponent analysis and learning vector quantization by the typical driving condition fragments and driving style fuzzy recognition model,the remaining range estimation were carried out,which adopted joint sim-ulation method,and the results of remaining range estimation were optimized by Kalman filtering.The simulation analysis and hardware in the loop test results show that driving range estimation for electric vehicles through vehicles,roads,and human information fusion method,reduces the remaining range es-timation errors,and confirmes the feasibility of the method at the same time.
作者
高建平
高小杰
郗建国
GAO Jianping;GAO Xiaojie;XI Jianguo(Vehicle and Transportation College,Henan University of Science and Technology,Luoyang,Henan,471003;Collaborative Innovation Center of Machinery Equipment Advanced Manufacturing of Henan Province,Henan University of Science and Technology,Luoyang,Henan,471003)
出处
《中国机械工程》
EI
CAS
CSCD
北大核心
2018年第15期1854-1862,共9页
China Mechanical Engineering
基金
国家自然科学基金资助项目(U1604147)
河南省科技攻关计划资助项目(152102210073)
河南省高等学校青年骨干教师培养计划资助项目(2015GGJS-046)
关键词
电动汽车
续驶里程
行驶工况
驾驶风格
electric vehicle
driving range
driving cycle
driving style
