摘要
如今金属材料的发展进入了瓶颈期,急需一个新的发展方式来突破当前的瓶颈。上世纪50年代,人工智能逐渐兴起,经过60多年的发展,人工智能技术趋于成熟,大幅度的应用在各种领域中,材料领域也有所涉及。数据与人工智能结合的数据驱动方式成为改变金属材料发展瓶颈的新方式,有望大幅度提升金属材料的研发速度。介绍了金属材料领域的合金设计现状,在传统的“试错法”已经不能满足现有金属材料研发的基础上,综述了机器学习在金属材料力学性能预测等方面的一些应用。总结了机器学习存在的不足和需要优化的地方,展望了机器学习在金属材料领域中的发展方向。
Nowadays,the development of metal materials has entered a bottleneck,and a new way of development is urgently needed to break through the current bottleneck.Artificial intelligence gradually emerged in the 1950s,and after more than 60 years of development,AI technology has matured and has been substantially applied in various fields,including the field of materials.The data-driven approach combining data and AI has become a new way to change the development bottleneck of metal materials,which is expected to significantly increase the speed of metal materials research and development.This paper introduces the current situation of alloy design in the field of metal materials.Given that traditional"trial-and-error"methods can no longer meet the requirements of current metal material research and development,it summarizes some applications of machine learning in predicting mechanical properties of metal materials.It also summarizes the shortcomings of machine learning and areas that need optimization,and finally provides prospects for the development of machine learning in the field of metal materials.
作者
程洪
葛美伶
司天宇
张欢
何忠平
CHENG Hong;GE Meiling;SI Tianyu;ZHANG Huan;HE Zongping(College of Mechanical Engineering,Chengdu University,Chengdu 610106,China)
出处
《材料研究与应用》
CAS
2023年第6期1070-1077,共8页
Materials Research and Application
关键词
金属材料
机器学习
性能预测
合金设计
数据挖掘
metallic materials
machine learning
property prediction
alloy design
data mining
