Experimental Combined Grouping Analysis Approach for Robust Battery pack design for Electric Vehicles with Higher Performance

International Conference on Sustainable Energy and Green Technology 2018

能源学;生态环境科学

Yun, Liu^1 ; Shui, Li^1 ; Gao, Liang^2 ; Fan, Zhun^3 ; Ruhatiya, C.^1^4 ; Wang, Chin-Tsan^5 ; Garg, Akhil^1

Intelligent Manufacturing Key Laboratory, Ministry of Education, Shantou University, Guangdong, China^1

State Key Lab of Digital Manufacturing Equipment and Technology, School of Mechanical Science and Engineering, Huazhong University of Science and Technology, Wuhan, China^2

Guangdong Provincial Key Laboratory of Digital Signal and Image Processing, College of Engineering, Shantou University, Shantou

515063, China^3

School of Engineering Sciences, Mahindra Ecole Centrale, Telangana, Hyderabad

500043, India^4

Department of Mechanical and Electro-Mechanical Engineering, National i Lan University, ILan, Taiwan^5

关键词: Analysis approach;    Battery pack design;    Clustering methods;    Energy efficient;    Intelligent clustering;    Multiple parameters;    Self-organizing map neural network;    Single cell performance;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/268/1/012020/pdf
DOI  :  10.1088/1755-1315/268/1/012020

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

Battery consistency is an important factor for battery pack performance. Excellent battery consistency can make battery packs more energy efficient and electric vehicles can have longer mileage and higher safety. Thus, in this study a comprehensive intelligent clustering methodology for the design of Li-ion battery pack on the basis of uniformity and equalization criteria of the cell was proposed. Firstly, multiple parameters (capacity, voltage, temperature and resistance) test of single cell performance was performed. Secondly, a clustering method combine with self-organizing map neural network (SOM) was proposed. Furthermore, a validation experiment (pack level) was carried out to verify the accuracy of proposed clustering algorithm. It can be concluded that the battery pack formed from SOM sorting results perform better than the battery pack having random cells combination as well as the pack originally purchased from the manufacturer.