[Time] 15:00 – 16:00, December 20 (Friday)
[Venue] Room 412, Shunde building
[Speaker] Dr. Xiaowei Yue
[Host] Dr. Chen Zhang
[Title] Engineering-Driven Data Analytics for System Informatics

[Abstract]
Engineering-driven data analytics focuses on developing new methodologies for system-level fusion of data-driven models and physical models in order to realize real-time monitoring of system operations, accurate detection of system faults, quick diagnosis of root causes, predictive control and process optimization. In this presentation, three research examples will be covered: (1) High-dimension nonlinear data are commonly encountered in many complex engineering systems. A penalized mixed-effects decomposition (PMD) method was proposed to decompose high-dimensional functional data into different components. The extracted features can represent corresponding quality characteristics in the scalable nanomanufacturing process. (2) Dimensional shape control of composite parts is vital for large-scale production and integration of composite structures. An automated shape control system was proposed to adjust composite parts to an optimal configuration in an effective and efficient manner. (3) A physics-driven deep learning architecture, “StressNet” was developed to predict the maximum stress in fracture propagation of brittle materials. The proposed methods are formulated into a systematic data decomposition framework and it will be briefly discussed.
[Bio]
Xiaowei Yue is an assistant professor at the Grado Department of Industrial and Systems Engineering, Virginia Tech. He got his Ph.D. degree in industrial engineering (Minor: Machine Learning), M.S. in Statistics from Georgia Tech, M.S. in Engineering Thermophysics from Tsinghua, B.S. in Mechanical Engineering from Beijing Institute of Technology. His research interests focus on engineering-driven data mining and analytics for advanced manufacturing. He is a recipient of Mary G. and Joseph Natrella Scholarship from American Statistical Association, and IISE Pritsker Doctoral Dissertation Award, FTC Early Career Awards from ASQ, and several best paper awards, e.g. IEEE Transactions on Automation Science and Engineering Best Paper Award, INFORMS Data Mining Best Paper Finalist Award, etc.