Project Grant 2226347

The National Science Foundation awarded North Carolina State University $282,315 under the Engineering (47.041) federal grant program to develop new digital twin calibration methods using stochastic optimization techniques. The two-year project will contribute to national prosperity by providing robust estimation approaches for parameter calibration of digital twins with large, complex datasets. Key activities include developing stochastic optimization reconciled with statistical theories to guide simulation experiments through efficient subset sampling, extending the integrative optimization framework to multi-dimensional, functional, and time-variant calibration problems, and incorporating input uncertainty into optimization to enhance solution robustness while maintaining computational tractability. Validation case studies in building energy and wind power systems are also part of the project. The award reflects NSF's mission to advance engineering research and was deemed worthy following intellectual merit and broader impacts review.