论文部分内容阅读
为了对导弹进行多学科设计优化(MDO),建立了包含气动、推进、质量、控制和弹道的多学科分析模型,并采用协作优化对战术导弹多目标多学科设计优化问题进行了表述。针对多目标多学科优化设计问题的计算复杂性,提出了一种新的处理约束多目标优化问题的基于Kriging多目标遗传算法(MOKGA)。MOKGA采用物理规划法将多目标优化转化为单目标优化,然后构建目标函数的考虑约束的EI(Expected Improvement)模型,并采用遗传算法进行求解。将MOKGA与多目标优化算法NSGA-II进行了比较。结果表明,NSGA-II和MOKGA两种算法的优化结果均较初始方案得到明显改进,但MOKGA的精确分析次数较NSGA-II减少了40%,降低了多学科设计优化问题求解过程中的计算复杂性。
In order to carry out multidisciplinary design optimization (MDO) for missiles, a multidisciplinary analysis model including aerodynamics, propulsion, mass, control and trajectory was established, and the multi-objective multi-disciplinary design optimization problem of tactical missile was described by using collaborative optimization. Aiming at the computational complexity of multi-objective multi-disciplinary optimization design problems, a new Kriging-based multi-objective genetic algorithm (MOKGA) is proposed to deal with constrained multi-objective optimization problems. MOKGA uses the physical programming method to transform multi-objective optimization into single-objective optimization, and then constructs the EI (Expected Improvement) model with the objective function and uses genetic algorithm to solve it. The MOKGA is compared with the multi-objective optimization algorithm NSGA-II. The results show that the optimization results of both NSGA-II and MOKGA algorithms are significantly improved compared with the original scheme, but the number of accurate analysis of MOKGA is reduced by 40% compared with that of NSGA-II, which reduces the computational complexity of solving multidisciplinary design optimization problems Sex.