论文部分内容阅读
气动外形的全局优化设计会产生大量的过程数据,其中隐含的设计知识具有较高的挖掘价值.数据挖掘有助于获取直观、可定性描述的设计知识.本文采用基于本征正交分解的数据挖掘方法从气动优化设计的过程数据中获取设计知识,数据挖掘对象为跨音速压气机转子叶片NASA Rotor 37的优化过程数据,该数据由基于粒子群方法的绝热效率最大化优化设计产生.结果表明:基于本文数据挖掘方法获取的设计知识能够直接反映气动外形的变化规律,为叶片的气动外形设计提供参考;数据挖掘的设计知识成功地验证了优化设计结果的有效性.
The global optimization design of aerodynamic shape will produce a large amount of process data, and the implicit design knowledge has a high value of mining.Data mining can help to obtain intuitive and qualitative description of the design knowledge.In this paper, based on the eigen-orthogonal decomposition Data Mining Method The design knowledge was obtained from the process data of aerodynamic optimization design and the data mining object was optimized process data of the transonic compressor rotor blade NASA Rotor 37. The data was generated by the optimization of adiabatic efficiency based on particle swarm optimization.Results The results show that the design knowledge obtained based on the data mining method can directly reflect the changing rule of the aerodynamic shape and provide a reference for the aerodynamic profile design of the blade. The design knowledge of data mining has successfully verified the validity of the optimized design results.