An efficient method of aerodynamic shape optimization assisted by aerodynamic database

The management and reuse of aerodynamic data is of great significance for computational fluid dynamics (CFD)-based analysis and design optimization. Based on eXtensible Markup Language (XML) document, a general aerodynamic database is established, which realizes management and reuse of aerodynamic data. On the basis of surrogate based optimization (SBO), an efficient method of aerodynamic shape optimization assisted by aerodynamic database is proposed, which contains two new database-assisted optimization strategies. The first one is the database-assisted quick design, which can generate optimal design just in seconds without any additional CFD evaluation. The result of quick design is almost same with a 50 CFD evaluations’ SBO result. The second strategy is the database-assisted optimization based on hierarchical kriging (HK) model, which can accelerate the original SBO. This strategy converged to the best solution quickly with only 40% computational cost compared to SBO. It is proved to be robust and efficient by a rudder optimization test case. These cases show that the proposed method has great potential to become a useful optimization tool for detailed design of aircraft in the future.