3-22. Data-driven approach to understand environmental corrosion

Dawei Zhang1,3*, Tao Yang2,3, Xuequn Cheng1,3, Dongmei Fu2,3, Xiaogang Li1,3

1. Beijing Advanced Innovation Center for Materials Genome Engineering, University of Science and Technology Beijing;

2. School of Automation and Electrical Engineering, University of Science and Technology Beijing;
3. National Materials Corrosion and Protection Data Center

Abstract: According to a 2017 nationwide survey, corrosion is estimated to cost over 2.1 trillion RMB annually in China, representing ~3.34% of the gross domestic product. The recent Materials Genome Engineering (MGE) initiative provides new opportunities to demystify the complex corrosion problems and to more effectively design corrosion-resistant materials. A data-driven approach is proposed within the MGE framework by integrating i) data accumulation, ii) data standardization and classification, iii) data mining and modelling and iv) data sharing and applications. The talk will summarize some of the recent progresses made by the National Materials Corrosion and Protection Data Center in the development of on-line corrosion monitoring sensors and data mining techniques for better understanding and predication of atmospheric corrosion.

Keywords: Atmospheric corrosion, Data mining, Corrosion monitoring 

数据驱动材料环境腐蚀失效研究

张达威 1,3*，杨焘 2,3，程学群 1,3，付冬梅 2,3，李晓刚 1,3 

1.北京科技大学，北京材料基因工程高精尖创新中心;

2.北京科技大学自动化学院;3.国家材料腐蚀与防护科学数据中心 

摘要:我国每年的腐蚀损失超过 2.1 万亿元，约占国内生产总值的 3.34%。材料基因工程的研究 方法为更为深入的理解复杂腐蚀问题和更有效的设计先进耐腐蚀材料提供了新的机遇。通过集 成 i)材料腐蚀大数据积累、ii)数据标准化和分级分类、iii)数据挖掘和建模以及 iv)数据共 享和应用，本文在材料基因工程的研究体系内提出了数据驱动下材料环境腐蚀研究方法，总结 了国家材料腐蚀与防护科学数据中心在发展大气腐蚀在线监测传感器和数据挖掘技术方面取得 的一些最新进展。
关键词:大气腐蚀;数据挖掘;腐蚀监测