In May, Grade 11 students Grace Y. and Akira Y. competed in the Canada-Wide Science Fair. The weeklong Canada-Wide Science Fair (CWSF) brings together the country’s top young scientists in Grades 7 to 12. In addition to scientific, social, and cultural activities, the 500 finalists compete for medals, cash prizes, scholarships and exclusive opportunities. Grace and Akira were awarded the Gold Excellence Award (awarded to the top 10 projects in every division) and the YouthCan Innovate Award (awarded to the top 8 projects in the senior division).
Improving the efficiency of airfoil design is key to the creation of more fuel-efficient airplanes, reducing their carbon footprints. Unfortunately, current component optimization is inaccessible due to its reliance on iterative Computational Fluid Dynamics (CFD), which is expensive, computationally costly, and difficult to use, resulting in inefficient designs. Rapid advancements in machine learning techniques open new avenues to streamline this process.
We investigated the simulation runtime and accuracy of generating airfoil designs through the construction of a series of machine learning models. Models utilizing ensemble methods and K-Nearest Neighbors Bagging, as well as the multilayer perceptron, produce highly accurate and efficient predictions for familiar data, though error increases substantially when assessed with completely unseen data. Other models are either unsuitable, inefficient, or inaccurate due to our limitations when tuning hyperparameters. Overall, machine learning models present significant improvements in runtime, ease of use and cost over traditional CFD software. Here is a link to their project board from the fair: https://projectboard.world/ysc/project/on-the-generative-design-of-airfoils-bvenjj?rc=yxxpvxby