Learning-Guided Exploration in Airfoil Optimization
نویسندگان
چکیده
A learning-based exploration approach is proposed to escape from the basins of attraction of converged-to optima, by selecting on what is termed the interestingness of a solution. This interestingness is based on the modeling error made by a surrogate model that is trained on all solutions encountered earlier during the search. Compared to multiple standard optimization runs, a learning-guided restart scheme that alternates between a quality optimization phase and an exploration phase directed by interestingness finds solutions that are more diverse and of higher quality.
منابع مشابه
Airfoil Design Optimization for Airplane for Mars Exploration
Aerodynamic design optimization of an airfoil for the Mars exploration airplane has been demonstrated by using an evolutionary algorithm. The adaptive range genetic algorithm is used for efficient and robust design optimization. Two-dimensional Navier-Stokes solver is used for accurate aerodynamic performance evaluation. The present computation is parallelized on the SX-6 vector computers in In...
متن کاملA Study on Airfoil Design for Future Mars Airplane
An optimum airfoil design for future Mars airplane for Mars exploration is obtained by evolutionary computation coupled with a two-dimensional Reynolds-averaged Navier-Stokes solver. The optimized airfoil design is also compared with other airfoil designs optimized at different Reynolds number or at different Mach number to discuss Reynolds number and Mach number effects on airfoil design. Thes...
متن کاملReinforcement Learning of a Morphing Airfoil-Policy and Discrete Learning Analysis
An episodic unsupervised learning algorithm using the Q-Learning method is developed to learn the optimal shape and shape change policy of a morphing airfoil. Optimality is addressed by reward functions based on airfoil properties such as lift coefficient, drag coefficient, and moment coefficient about the leading edge representing optimal shapes for specified flight conditions. The reinforceme...
متن کاملEfficient Aerodynamic Optimization Using a Multiobjective Optimization Based Framework to Balance the Exploration and Exploitation
In many aerospace engineering design problems, objective function evaluations can be extremely computationally expensive, such as the optimal design of the aerodynamic shape of an airfoil using high-fidelity computational fluid dynamics (CFD) simulation. A widely used approach for dealing with expensive optimization is to use cheap global surrogate (approximation) models to substitute expensive...
متن کاملAirfoil Shape Optimization with Adaptive Mutation Genetic Algorithm
An efficient method for scattering Genetic Algorithm (GA) individuals in the design space is proposed to accelerate airfoil shape optimization. The method used here is based on the variation of the mutation rate for each gene of the chromosomes by taking feedback from the current population. An adaptive method for airfoil shape parameterization is also applied and its impact on the optimum desi...
متن کامل