基于遗传算法与近似模型的全局气动优化方法

Global aerodynamic optimization method using genetic algorithms and surrogate model

关键词:  叶型  三维动叶片+  遗传算法  响应面模型+  流场求解程序+

Key words:  Airfoil  3-D rotor blade+  Genetic algorithm  Response surface model+  Flow field solver+

摘要:

开发了基于精英保留策略与小生境技术的求解有约束优化问题的改进遗传算法,并用一多峰值有约束函数对其寻优能力进行了测试。为了降低气动优化设计时间,提出了集成实验设计、三维粘性流场求解程序、二次多项式近似模型、改进遗传算法的具有全局寻优能力的三维气动优化设计体系。采用该方法对NASA Rotor37动叶片叶型进行了以绝热效率最大为目标的优化设计,优化后叶片绝热效率提高1.1%。该结果表明了本优化设计体系省时、高效的特点。

Abstract:

Firstly an improved genetic algorithm for constrained optimization problems was developed based on elitism and niche technique.A multimodal constrained function was used to test the performance of the new GA.To reduce the time-consuming for aerodynamic optimization design,an effective three-dimensional aerodynamic optimization design system,combined with design of experiment(DOE),the 3-Dimensional Navier-Stocks solver code,quadratic response surface model and improved genetic algorithm,was developed.The system was used to the optimization designs of airfoils of NASA Rotor37 blade with the objection of maximum isentropic efficiency.It was found that the efficiency of rotor blade increased by 1.1%.The results show that the presented optimization system is time saving and has fine performance.