Artificial intelligence for accelerating time integrations in multiscale modeling

Journal of Computational Physics(2021)

引用 19|浏览7
暂无评分
摘要
•AI-guided data-driven adaptive time stepping for accelerating multiscale modeling of demanding biophysical systems.•AI-ATS encapsulates physics-informed feature extraction and fully connected neural networks for inferencing optimal time-stepping.•An integrated workflow of multiscale simulator with AI-ATM facilitates seamless data flows and the AI inference on-the-fly.•We reduced 40% calculations while keeping 97% accuracy for modeling a flowing platelet with the standard methods.
更多
查看译文
关键词
Adaptive time stepping,Artificial intelligence,Multiscale modeling,Platelet dynamics
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要