Research vision

My research is to improve the practicality of aerodynamic shape optimization in the industrial design of aircraft and wind energy systems. I aim to provide effective solutions to large-scale aerodynamic shape optimization problems, especially those with massive design points, uncertain design variables, discontinuous merit functions, and multiple design objectives.

The difficulties to address the demands come from two fundamental issues: the high dimensionality of shape design variables and the high computational cost of CFD simulations. My research shows that using scientific machine learning is a way to solve these issues and realize practical and large-scale aerodynamic shape optimization for the industry.

My research highlights are:

• Proposed a compact shape parameterization approach that addresses the dimensionality issue in design optimization of aircraft and wind turbines
• Developed fast aerodynamic models for on-design and off-design performance analysis to reduce the computational cost
• Presented a flexible and efficient optimization framework that enables effective aerodynamic shape optimization using any CFD solvers