报告讲座 > 正文

Yi Wang: Computational and Data-enabled Science and Engineering for Engineering Applications
发布日期:2019-07-11  字号:   【打印

报告时间:2019年7月17日(星期三)9:00

报告地点:机械楼二楼214

  :Yi Wang (王懿) 副教授

工作单位:美国南卡罗莱纳大学机械工程系

举办单位:机械工程学院

报告人简介

王懿,美国南卡罗莱纳大学机械工程系副教授。1998年、2001年于上海交通大学动力与机械工程系获得学士及硕士学位,2005年获得美国卡内基梅隆大学机械工程博士学位。主要从事数值计算与数据科学工程在多物理系统中的研究与应用,具体包括降阶模型,大规模及实时数据分析学,在线机器学习及实时系统仿真与预测。应用对象包括生物微流体芯片分析,空气弹性及空气弹性伺幅分析,半导体热加工过程预测,海量工程计算数据管理,异常系统在线机器学习与控制补偿等。共发表SCI检索期刊论文及会议论文80余篇(包括一篇“热门文章”及两篇封面文章),及4篇书籍章节。获得5项美国专利以及美国航空航天局颁发最佳科技通讯奖。作为项目负责人,开发的太空水环境检测系统及再生医学微流体芯片技术两次被美国广播公司独家专访。作为项目负责人,共获得40余项美国联邦级别项目资助 (共计1500万美元)。在加入南卡罗莱纳大学之前,2005至2017年间,王懿博士在美国CFD科技公司先进技术创新中心先后担任工程师,高级工程师,项目经理及主任,从事多物理系统数值建模及微流体芯片技术研究。

报告简介

Engineering systems feature unprecedented complexities arising from multiphysics and multiscale, posing critical challenges to the engineering community. Computational and data-enabled science and engineering (CDS&E), recently emerging as a focal point across multidisciplinary fields, including aerospace, energy management, additive manufacturing, and biomedicine, enables major engineering breakthroughs by discovering underlying physics, identifying decisive patterns, and accelerating system developments. A recent trend is to apply CDS&E in essentially each phase of multiphysics system and technology development, from conceptualization, virtual prototyping and design, and automation and control, to final verification and validation. 

In this seminar, I will first highlight the significance and impacts of CDS&E to solving practical engineering problems. Then main thrust areas and projects of CDS&E research undertaken in my group at the University of South Carolina will be presented, including equation-based and data-driven reduced order modeling, large-scale data analytics and management, and online machine learning for anomaly detection and compensation in mechanical systems. Practically relevant case studies, such as aeroservoelasticity, microfluidic thermal analysis, semi-conductor rapid thermal processing, and scalable massive CFD data analytics, and others, will be discussed to demonstrate salient computing performance enabled by CDS&E (i.e., orders of magnitude of speedup, remarkable accuracy, and dramatic improvement in resource utilization) and its great promise for system performance assessment, real-time simulation/prediction, design and optimization, control synthesis, and feature detection and signature analysis.

(陈华/文)  
编辑:徐小红
0