Please wait a minute...
工程设计学报
Chinese Journal of Engineering Design  2011, Vol. 18 Issue (5): 327-331    DOI:
english     
Nonprobabilistic reliability analysis based on support vector machines
ZHENG Yan1, CHENG Wenming1, CHENG Yue2, WU Xiao1
1.Research Institute of Mechanical Engineering, Southwest Jiaotong University, Chengdu 610031, China;2.School of Industrial Manufacturing, Chengdu University, Chengdu 610106, China
Download: HTML     PDF(750KB)
Export: BibTeX | EndNote (RIS)      

Abstract  For nonprobabilistic reliability analysis of uncertainty structure with implicit limit state function, a method for nonprobabilistic reliability analysis of uncertainty structure was presented on the basis of support vector regression, and analysis process of the method was given. Taking data samples and optimizing support vector regression parameters, the support vector machine after training was used to replace implicit limit state function. Using interval analysis method of nonprobabilistic set theory and introducing size scale factor, suitable nonprobabilistic reliability index iterative algorithm of uncertainty structure was presented. Finally, a numerical example and an engineering example were used to test the presented method. The numerical example shows that support vector regression model forecast precision is high, and using the model instead of implicit limit state function to analyze nonprobabilistic reliability is feasible. The engineering example shows that the presented method has three virtues of feasibility, correctness and high accuracy. Therefore, the presented method is of good practical engineering value, and it also has a certain useful value for other uncertainty structure nonprobabilistic reliability analysis with implicit limit state function．

Key words： support vector regression      nonprobabilistic reliability analysis      implicit limit state function      interval analysis     
Published: 28 October 2011
CLC:  TB 1143  
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
CHENG Yue
ZHENG Yan
CHENG Wen-Ming
WU Xiao
Cite this article:

ZHENG Yan, CHENG Wen-Ming, CHENG Yue, WU Xiao. Nonprobabilistic reliability analysis based on support vector machines. Chinese Journal of Engineering Design, 2011, 18(5): 327-331.

URL:

https://www.zjujournals.com/gcsjxb/     OR     https://www.zjujournals.com/gcsjxb/Y2011/V18/I5/327


基于支持向量机的非概率可靠性分析

针对具有隐式极限状态函数的不确定性结构的非概率可靠性分析,提出一种基于支持向量机回归的不确定性结构非概率可靠性分析方法,并给出了该方法的分析流程.抽取数据样本,优化支持向量机回归参数,使用训练后的支持向量机替代隐式极限状态函数.利用非概率集合理论中的区间分析法,引入尺寸比例因子,构造合适的不确定性结构非概率可靠性指标迭代算法.最后用一个数值算例和一个工程算例来验证该方法.数值算例表明支持向量机回归模型预测精度高,用支持向量机回归模型替代结构隐式极限状态函数进行非概率可靠性分析是可行的.工程算例证明该方法具有可行性、正确性和高精度的优点.因此,该方法具有较好的工程实用价值,并对其它具有隐式极限状态函数的不确定性结构的非概率可靠性分析具有一定的参考价值.

关键词： 支持向量机回归,  非概率可靠性分析,  隐式极限状态函数,  区间分析 
[1] Hao ZHOU,Shang-lin LIU,Kai-hong YANG,Si-yang ZHOU,Qian ZHANG. Research on prediction method of driving load of shield machine based on mutual information and support vector regression[J]. Chinese Journal of Engineering Design, 2022, 29(3): 286-292.
[2] YANG Chao， DI Peng， CHEN Tong. Fuzzy reliability allocation method for warship armaments based on interval analysis[J]. Chinese Journal of Engineering Design, 2015, 22(5): 317-323.
[3] YANG Chao， DI Peng， CHEN Tong. Fuzzy reliability allocation method for warship armaments based on interval analysis[J]. Chinese Journal of Engineering Design, 2015, 22(4): 317-323.