Visualization of Global Trade-Offs in Aerodynamic Problems by ARMOGAs

Abstract

Trade-offs is one of important elements for engineering design problems characterized by multiple conflicting design objectives to be simultaneously improved. 
In many design problems such as aerodynamic design, due to computational reasons, only a limited number of evaluations can be allowed for industrial use. 
Efficient MOEAs, Adaptive Range Multi-Objective Genetic Algorithms (ARMOGAs), to identify trade-offs using a small number of function evaluations have been developed. 
In this study, ARMOGAs are applied to aerodynamic designs problems to identify trade-offs efficiently. 
In addition to identify trade-offs, trade-off analysis is also important to obtain useful knowledge about the design problem. 
To analyze the high-dimensional data of aerodynamic optimization problem, Self-Organizing Maps are applied to understand the trade-offs.