Aerodynamic topology optimisation using an implicit representation and a multiobjective genetic algorithm

摘要: Given the focus on incremental change in existing empiricalaerodynamic design methods, radical, unintuitive, new optimal solutionsin previously unexplored regions of space are very unlikely to befound using them. We present a framework based an implicit shaperepresentation and multiobjective evolutionary algorithm that aims toproduce variety flow topologies for given requirement,providing designers with insights into possibly radical solutions. A revolutionaryintegrated simulation system developed specifically fordesign work is used evaluate candidate designs.