期刊
RELIABILITY ENGINEERING & SYSTEM SAFETY
卷 204, 期 -, 页码 -出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.ress.2020.107194
关键词
Complex systems; Prognostics and health management (PHM); Probabilistic risk assessment; Logic modeling; Deep learning
资金
- U.S. Nuclear Regulatory Commission [31310018M0043]
Recent advances in sensing and computing technologies have resulted in an abundance of data in various formats and more processing power for using this data. Consequently, there has been an interest in using these advances to enhance modeling and assessment techniques for safety and reliability of a variety of systems. To date, this has occurred under two distinct aspects of reliability engineering. Prognostics and Health Management (PHM) has developed powerful new algorithms for understanding and predicting the mechanical and electrical devices' health. For complex systems, the techniques of Probabilistic Risk Assessment (PRA), which provide a system-level perspective, have become increasingly dynamic. Both PHM and PRA bring unique advantages and limitations. PHM excels at data handling and supports prediction, but the methods applicable at the component level are not suitable for modeling complex engineering systems (CES). PRA provides a comprehensive approach suitable for drawing together many types of data and assessing complex systems, but is limited in the ability to exploit advanced ML methods or enable prediction. In this paper, we explore how to systematically draw together the advances in PHM and PRA to provide a more forward-looking, model- and data-driven approach for assessing and predicting the risk and health of CES.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据