Zhaoyi is a Ph.D. candidate in the Guggenheim School of Aerospace Engineering at Georgia Institute of Technology. His research focuses on machine learning (ML) in engineering applications, with an emphasis on ML for reliability and system safety, as well as design optimization. He is also proficient in computational fluid dynamics (CFD).

Zhaoyi is experienced in ML-aided reliability analysis, numerical simulation, and optimization. He developed a high accuracy deep neural network (DNN) surrogate model for helicopter accident probability estimation. The model significantly improves prediction accuracy and provides novel safety insights for improving helicopter accident analysis and prevention. In his prior CFD work, Zhaoyi worked on turbulent flow simulation. He designed a conditional generative adversarial network (cGAN) of time-averaged flow-field prediction, and long-short term memory (LSTM) auto-encoder surrogate model to alleviate the significant computational burden associated with turbulence propagation. In his present work, Zhaoyi is leveraging Bayesian optimization and Monte-Carlo simulation to build a robust optimization framework with uncertainty quantification (UQ) to improve decision-making and system design.

Zhaoyi received his M.S. degree in aerospace engineering from the Georgia Institute of Technology. His work on supercritical airfoil optimization, using the genetic algorithm (GA) with ML surrogate was recently published in the AIAA Journal (2019, 57-10). He received his B.S. degree in aerospace engineering at Northwestern Polytechnical University (China) in 2015 where he received an outstanding undergraduate thesis award based on his work on transonic unsteady flow.