基于响应面的BLISS改进方法
Improved method of BLISS based on response surface
-
摘要: 在分析研究BLISS(bi-level integrated system synthesis)方法的基础上,提出了一种BLISS改进方法,该方法在子系统级和系统级均引入基于试验设计的响应面进行近似,并设置了可调移动步长参数,不需再进行复杂的灵敏度分析,消除了BLISS方法对每步迭代优化移动步长的限制,减少了优化迭代次数和陷入局部最优的可能性.最后给出两个具体算例对上述方法进行了验证并与传统BLISS方法进行了对比,结果表明:其优化迭代次数分别减少了57.9%和70.3%,证明了该方法的可行性和有效性.Abstract: Based on the analysis of the BLISS (bi-level integrated system synthesis),an improved method of BLISS was proposed,and the response surface based on design of experiments was introduced to make an approximation in both the subsystem and the system-level meanwhile the parameters of moving step were set,making unnecessary for complex sensitivity analysis;the movement limits of each iterative optimization in the BLISS method,and the optimization iterations and the possibility of falling into local optimum were reduced.Finally,two specific examples were given to verify above method compared with traditional BLISS.The results show that the optimization iterations of above methods were respectively reduced by 57.9% and 70.3%,respectively.
-
[1] Sobieszczanski-Sobieski J,Agte J S,Sandusky R R,Jr.Bi-level integrated system synthesis (BLISS) [R].AIAA-98-4916,1998. [2] Kodiyalam S,Sobieszczanski-Sobieski J.Bi-level integrated system synthesis with response surfaces [R].AIAA-99-1306-wip,1999. [3] Sobieszczanski-Sobieski J,Altus T D,Phillips M,et al.Bi-level integrated system synthesis (BLISS) for concurrent and distributed processing [R].AIAA 2002-5409,2002. [4] XU Lin,YAN Li,CHEN Xiaoqian,et al.Enhanced bi-level integrated system synthesis 2000(EBLISS 2000) [R].AIAA 2009-2202,2009. [5] 赵勇,杨维维,黄奕勇,等.一种改进的混合BLISS多学科设计优化方法[J].国防科技大学学报,2011,33(5):17-21. ZHAO Yong,YANG Weiwei,HUANG Yiyong,et al.An improved MDO procedure:hybrid bi-level integrated system synthesis[J].Journal of National University of Defense Technology,2011,33(5):17-21.(in Chinese) [6] Prince A R,Keane J,Holden C M,et al.On the coordination of multidisciplinary design optimization using expert systems[J].AIAA Journal,2011,49(8):1778-1794. [7] Amer F R,He L S,Zeeshan Q,et al.Integrated system design of air launched small space launch vehicle using genetic algorithm [R].AIAA 2009-5506,2009. [8] 陈伟,杨树兴,赵良玉.BLISS方法的基本理论及应用[J].弹箭与制导学报,2006,27(5):229-236. CHEN Wei,YANG Shuxing,ZHAO Liangyu.The basic theory and application of BLISS[J].Journal of Projectiles,Rockets,Missiles and Guidance,2006,27(5):229-236.(in Chinese) [9] Montgomery D C.Design and analysis of experiments[M].7th ed.New York:John Wiley & Sons,2008. [10] Giunta A A.Aircraft multidisciplinary design optimization using design of experiments theory and response surface modeling methods [D].Blacksburg,Virginia:Virginia Polytechnic Institute and State University,1997. [11] Liu W, Batill S.Gradient-enhanced response surface approximations using kriging models [R].AIAA 2000-5498, 2000. [12] Koch P N,Wujek B.Facilitating probabilistic multidisciplinary design Optimization using kringing approximation models [R].AIAA 2002-5415,2002. [13] Stelmack M A,Batill S M.Neural network approximation of mixed continuous/discrete system in multidisciplinary design [R].AIAA 98-0916,1998. [14] 王洪元,史国栋.人工神经网络技术及其应用[M].北京:中国石化出版社,2002. [15] Suykens J,Gestel T V,Brabanter J D,et al.Least squares support vector machines[M].London:World Scientific Publishing,2002. [16] Guo Z W,Bai G C.Application of least squares support vector machine for regression to reliability analysis[J].Chinese Journal of Aeronautics,2009,22(2):160-166. [17] Azarm S,Li W C.Optimality and constrained derivatives in two-level design optimization[J].ASME Journal of Mechanical Design,1990,112(12):563-568.
点击查看大图
计量
- 文章访问数: 1298
- HTML浏览量: 2
- PDF量: 927
- 被引次数: 0