Machine learning and microstructure design of polymer nanocomposites for energy storage application

作     者：Yu Feng Wenxin Tang Yue Zhang Tiandong Zhang Yanan Shang Qingguo Chi Qingguo Chen Qingquan Lei Yu Feng;Wenxin Tang;Yue Zhang;Tiandong Zhang;Yanan Shang;Qingguo Chi;Qingguo Chen;Qingquan Lei

作者机构：Department of High VoltageKey Laboratory of Engineering Dielectrics and Its ApplicationMinistry of EducationHarbin University of Science and TechnologyHarbinChina School of Electrical and Electronic EngineeringHarbin University of Science and TechnologyHarbinChina 

出 版 物：《High Voltage》 (高电压（英文）)

年 卷 期：2022年第7卷第2期

页      面：242-250页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 0805[工学-材料科学与工程（可授工学、理学学位）] 080502[工学-材料学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：Foundation of China,Grant/Award Number:51807041,U20A20308 and 51977050 the Natural Science Foundation of Heilongjiang Province of China,Grant/Award Number:ZD2020E009 the China Postdoctoral Science Foundation,Grant/Award Number:2020T130156 Heilongjiang Postdoctoral Financial Assistance,Grant/Award Number:LBH‐Z18098 the Fundamental Research Foundation for Universities of Heilongjiang Prov-ince,Grant/Award Number:2019‐KYYWF‐0207 and 2018‐KYYWF‐1624 the State Key Laboratory of Power System and Generation Equipment,Grant/Award Number:SKLD20M13 

主　　题：composites composite prediction 

摘      要：Film dielectric capacitors have been widely used in high‐power electronic *** design of microstructure and the choice of fillers play an important role in nano-composites energy storage *** learning methods can classify and summarise the limited data and then explore the promising composite *** this work,a dataset has been established,which contained a large amount of data on the maximum energy storage density of *** using processed visual image infor-mation to express the internal information of composite,the prediction accuracy of the prediction models built by three machine learning algorithms increase from 84.1%to 91.9%,80.9%to 68.9%,70.6%to 81.6%,*** calculating the branch weight in the random forest prediction model,the influence degree of different descriptors on the energy storage performance of nanocomposites is analysed.A total of 10 groups of composites with different structure and filler amount were prepared in the laboratory,which were used to verify the reliability of prediction ***,the effective filler s structure is explored by three prediction models and some suggestions for the interface design of filler are given.