摘要
电工材料是电气装备的基础,其材料特性直接决定电气装备的极限电磁参数。随着科学技术的进步,生产、生活水平的不断提高,人们对电气装备具有的功能和性能的需求日益增加。理论研究和工程实践表明,以传统电工材料为基础生产的电气装备在功能和性能方面不能完全满足人类社会对先进电气装备快速增长的需求。电工新材料及其应用研究应使未来的电工装备具有挑战更高电磁参数极限的能力。材料基因工程为电工材料设计筛选提供了新的技术手段,通过利用高通量的方式设计与筛选材料,大幅缩短了材料的研发周期,显著降低了研发成本。本文综述了近几年材料基因工程先进理念与前沿技术在电工材料开发中的应用案例,这些案例表明了材料基因工程技术在电工材料研发中的适用性。本文还重点分析了以机器学习为代表的大数据技术在电工材料研发中的应用现状。
Electrical materials are the basis of electrical equipment,and their characteristics directly determine the limit of electromagnetic parameters of electrical equipment.With the improvement of science,production,and living standards,the requirements for the functions and performance of electrical equipment have increased.Theoretical research and engineering practice show that electrical equipment based on traditional electrical materials cannot fully satisfy the rapidly growing needs of human society.The research of new electrician materials and their applications should allow the future electrician equipment to have the ability to challenge the limit of electromagnetic parameters,Material Genome Engineering provides new technical means for the design and screening of electrical materials in a high-throughput way,reducing the development cycle and cost of electrical materials by more than half.This paper presents examples of material genome engineering applied to the development of electrical materials.These cases show the applicability of material genome engineering techniques in the development of electrical materials.This paper also focuses on the current status of the application of big data technology represented by machine learning in the research and development of electrical materials.
作者
王博
盛鹏
徐丽
李圣驿
白会涛
李慧
薛晴
WANG Bo;SHENG Peng;XU Li;LI Shengyi;BAI Huitao;LI Hui;XUE Qing(State Grid Smart Grid Research Institute Co.,Ltd.,Beijing 102209,China)
出处
《材料导报》
EI
CAS
CSCD
北大核心
2024年第13期208-224,共17页
Materials Reports
基金
国网智能电网研究院有限公司科技项目(525500200052)。
关键词
电工材料
材料基因工程
机器学习
高通量计算
高通量实验
electrical material
material genome engineering
machine learning
high-throughput computing
high-throughput experiments