HDMR在多维随机变量结构全局灵敏度分析的应用
|
摘要
针对多随机变量作用的框架结构体系,采用构造近似模型替代传统的大规模计算任务,并获取结构体系的全局灵敏度信息进行分析。结果表明,HDMR可以在一定样本的条件下准确构造近似模型并获取灵敏度信息,对结构优化与安全评估具有重要意义。
High dimensional model representation can reduce the complexity of the high-dimensional model problem from polynomial level by exponential growth to the polynomial level by using the hierarchical property of the function,and reduce the computational cost,and it has the function of revealing the output of the system with respect to the multi-input variable. Therefore,the HDMR construction approximation model is used to replace the traditional large-scale computation task for the frame structure system with multi-random variables,and the global sensitivity information of the structural system is analyzed. The results show that HDMR can accurately construct approximate model and obtain sensitivity information under certain sample conditions,which is important for structural optimization and safety assessment.
High dimensional model representation can reduce the complexity of the high-dimensional model problem from polynomial level by exponential growth to the polynomial level by using the hierarchical property of the function,and reduce the computational cost,and it has the function of revealing the output of the system with respect to the multi-input variable. Therefore,the HDMR construction approximation model is used to replace the traditional large-scale computation task for the frame structure system with multi-random variables,and the global sensitivity information of the structural system is analyzed. The results show that HDMR can accurately construct approximate model and obtain sensitivity information under certain sample conditions,which is important for structural optimization and safety assessment.
引文
[1]Rabitz H,Alis.O.F.General Foundations of High Dimensional Model Representations[J].Journal of Mathematical Chemistry,1999,25(2-3):197-233.
[2]宋梦,于继来,李碧君,等.HDMR在电网潮流概率评估与调控中的应用[J].电网技术,2014,38(6):1585-1592.
[3]汤龙,李光耀,王琥.Kriging-HDMR非线性近似模型方法[J].力学学报,2011,43(4):780-784.
[4]李伟平,窦现东,王振兴,等.BPNN-HDMR非线性近似模型方法及应用[J].湖南大学学报(自科版),2014,41(5):32-38.
[5]李亮,孙秦.SVM-HDMR高维非线性近似模型构造法[J].计算机工程与应用,2013,49(15):6-9.
[6]黄志远,邱浩波,蔡习文.IDIRECT-HDMR高维近似模型方法及工程应用[J].机械制造与自动化,2015(3):100-103.
[7]Ziehn T,Tomlin A S.GUI–HDMR–A software Tool for Global Sensitivity Analysis of Complex Models[J].Environmental Modelling&Software,2009,24(7):775-785.
[8]Li G,Rabitz H,Wang S W,et al.Correlation Method for Variance Reduction of Monte Carlo Integration in RS-HDMR[J].Journal of Computational Chemistry,2003,24(3):277-83.
[9]王晓霞,贾玉玺,董抒华.基于Morris方法的纤维复合材料结构件固化均匀性的全局灵敏度分析[J].复合材料学报,2015,32(4):1211-1217.
[10]周长聪[1].随机激励下的随机结构全局灵敏度分析[J].高技术通讯,2015,25(10-11):956-963.
[11]Li G,Hu J,Wang S W,et al.Random Sampling-high Dimensional Model Representation(RS-HDMR)and Orthogonality of Its Different Order Component Runctions.[J].Journal of Physical Chemistry A,2006,110(7):2474-85.
[2]宋梦,于继来,李碧君,等.HDMR在电网潮流概率评估与调控中的应用[J].电网技术,2014,38(6):1585-1592.
[3]汤龙,李光耀,王琥.Kriging-HDMR非线性近似模型方法[J].力学学报,2011,43(4):780-784.
[4]李伟平,窦现东,王振兴,等.BPNN-HDMR非线性近似模型方法及应用[J].湖南大学学报(自科版),2014,41(5):32-38.
[5]李亮,孙秦.SVM-HDMR高维非线性近似模型构造法[J].计算机工程与应用,2013,49(15):6-9.
[6]黄志远,邱浩波,蔡习文.IDIRECT-HDMR高维近似模型方法及工程应用[J].机械制造与自动化,2015(3):100-103.
[7]Ziehn T,Tomlin A S.GUI–HDMR–A software Tool for Global Sensitivity Analysis of Complex Models[J].Environmental Modelling&Software,2009,24(7):775-785.
[8]Li G,Rabitz H,Wang S W,et al.Correlation Method for Variance Reduction of Monte Carlo Integration in RS-HDMR[J].Journal of Computational Chemistry,2003,24(3):277-83.
[9]王晓霞,贾玉玺,董抒华.基于Morris方法的纤维复合材料结构件固化均匀性的全局灵敏度分析[J].复合材料学报,2015,32(4):1211-1217.
[10]周长聪[1].随机激励下的随机结构全局灵敏度分析[J].高技术通讯,2015,25(10-11):956-963.
[11]Li G,Hu J,Wang S W,et al.Random Sampling-high Dimensional Model Representation(RS-HDMR)and Orthogonality of Its Different Order Component Runctions.[J].Journal of Physical Chemistry A,2006,110(7):2474-85.