Multi-Objective Optimization of Low Reynolds Number Airfoil Using Convolutional Neural Network and Non-Dominated Sorting Genetic Algorithm

The airfoil is the prime component of flying vehicles. For low-speed flights, low Reynolds number airfoils are used. characteristic a laminar separation bubble and an associated drag rise. This paper presents framework for design airfoil. contributions proposed research twofold. First, convolutional neural network (CNN) designed aerodynamic coefficient prediction airfoils. Data generation discussed in detail XFOIL selected to obtain coefficients. performance CNN evaluated using different learning rate schedulers adaptive optimizers. trained model can predict coefficients with high accuracy. Second, used non-dominated sorting genetic algorithm (NSGA-II) multi-objective optimization at specific angle attack. A similar performed NSGA-II directly calling XFOIL, Pareto fronts both optimizations compared, it concluded that replicate actual considerably less time.