论文部分内容阅读
为了提高涡轮级多学科设计优化的优化效率,基于本征正交分解(Proper Orthogonal Decomposition,POD)技术,并结合快照样本自适应更新方法,提出了一种综合的涡轮级多学科优化系统。首先,通过进行POD分析,仅保留占优势的基函数,并以POD系数作为新的设计变量,设计变量个数由60个缩减为5个,提高了优化效率。然后,基于自适应进化规则,优化过程中对快照样本进行不断的进化和修正,从而提高POD精度。最后将该方法与涡轮多学科优化流程相结合,建立了一种高效率、高精度的优化策略。某涡轮优化的结果表明:该优化策略适于设计变量较多的复杂优化问题,且具有良好的收敛性,优化后设计点等熵效率提高了3.5%。
In order to improve the optimization efficiency of turbine-level multidisciplinary design optimization, an integrated turbo-level multidisciplinary optimization system is proposed based on the Proper Orthogonal Decomposition (POD) technique and the adaptive snapshot sample update method. First of all, by keeping the POD analysis, only the dominant basis functions are reserved. Taking the POD coefficient as the new design variable, the number of design variables is reduced from 60 to 5, which improves the optimization efficiency. Then, based on the adaptive evolution rules, the snapshot samples are continuously evolved and corrected in the optimization process to improve POD accuracy. Finally, this method is combined with the turbine multi-disciplinary optimization process to establish a high-efficiency, high-precision optimization strategy. The results of a turbine optimization show that the optimization strategy is suitable for the design of complex optimization problems with many variables and has good convergence. The isentropic efficiency of the optimized design point is improved by 3.5%.