[1]姚继涛,程凯凯,宋 璨.结构性能概率模型的小样本建模方法[J].西安建筑科技大学学报：自然版,2016,48(02):155-159.[doi:10.15986/j.1006-7930.2016.02. 001]
　YAO Jitao,CHENG Kaikai,SONG Can.Small samples method to establishment for the probability characteristic of structural performance[J].J.Xi’an Univ. of Arch. & Tech.：Natural Science Edition,2016,48(02):155-159.[doi:10.15986/j.1006-7930.2016.02. 001]
点击复制

结构性能概率模型的小样本建模方法()
分享到：

西安建筑科技大学学报：自然版[ISSN:1006-7930/CN:61-1295/TU]

卷:
48
期数:
2016年02期
页码:
155-159
栏目:
出版日期:
2016-04-28

文章信息/Info

Title:
Small samples method to establishment for the probability characteristic of structural performance
文章编号:
1006-7930(2016)02-0155-05
作者:
姚继涛; 程凯凯; 宋 璨
西安建筑科技大学土木工程学院，陕西 西安 710055
Author(s):
 YAO Jitao;  CHENG Kaikai ;  SONG Can
 School of Civi. Eng. , Xi′an Univ. of Arch. & Tech. , Xi′an 710055, China
关键词:
结构性能; 结构试验; 概率模型; 统计推断; 统计不定性
Keywords:
structural performance;  structural test;  probability model;  statistical inference;  statistical uncertainty
分类号:
TU375
DOI:
10.15986/j.1006-7930.2016.02. 001
文献标志码:
A
摘要:
目前通过试验建立结构性能概率模型的过程中，一般采用经典统计学中的矩法推断其中关键的计算模式不定性系数的概率特性，并以此为基础推断结构性能的概率特性；但样本容量不足时，矩法的推断结果会受到显著的统计不定性的影响，存在较大的因过高估计结构性能而导致额外失效风险的可能．针对目前试件数量（样本容量）普遍不足的现象，利用贝叶斯法提出建立结构性能概率模型的小样本方法，与矩法、区间估计法、一般贝叶斯法的对比分析结果说明：在统计不定性的反映程度、推断结果的等效置信水平和推断结果本身等方面，文中贝叶斯法都具有较好的表现，推断结果稳妥、适中，且便于应用，可作为小样本条件下建立结构性能概率模型的一个基本方法．
Abstract:
At present, in the process of establishing the structural performance probability model through the structural test, the probability characteristic of key model uncertainty coefficient is inferred from the moment method in classical statistics. On the basis of this, one can infer the probability characteristic of structural performance. However, a limited number of samples will cause uncertainties in the estimations of statistical parameters using the moment method and overestimate the structural performance, which will lead to additional failure risk of structures. Aimed at the insufficiency of sample numbers, the paper proposes the small samples method to establish the probability characteristic of structural performance using the Bayesian method. According to comparative analysis among the moment, the interval estimation methods and the Bayesian method, it shows that the proposed Bayesian method not only does well in reflecting statistical uncertainty, ensuring the confidence level of inference result and the inference result itself, but also its inference result is more accurate and convenient to put it into use, which can be used as a basic method to establish the structural performance
probability model under the condition of small samples.

参考文献/References:

[1] 中华人民共和国建设部．GB/T 50152-2012 混凝土结构试验方法标准[S]．北京：中国建筑工业出版社，2012.

Ministry of construction of the People’s Republic of China．GB/T 50152-2012 Standard methods for testing of concrete structures [S]．Beijing：China Architecture and Building Press，2012.

[2] 中华人民共和国建设部．GB 50152-92 混凝土结构试验方法标准[S]．北京：中国建筑工业出版社，1992.

Ministry of construction of the People’s Republic of China．GB 50152-92 Standard methods for testing of concrete structures [S]．Beijing：China Architecture and Building Press，1992.

[3] 茆诗松，王静龙，濮晓龙．高等数理统计[M]．北京：高等教育出版社，施普林格出版社，1998.

MAO Shisong，WANG Jinglong，PU Xiaolong．Advanced Mathematical Statistics [M]．Beijing：Higher Education Press，Springer-Verlag Press，1998.

[4] Alfredo H-S. Ang，Wilson H. Tang. Probability concepts in engineering planning and design (Ⅱ) [M]．New York: John Wiley & Son, Inc，1984.

[5] JCSS. JCSS probabilistic model code: part 1: basis of design [S]. Copenhagen, Denmark: Joint Committee on Structural Safety, 2001.

[6] 中国建筑科学研究院．建筑结构设计统一标：GBJ68-84[S]．北京：中国建筑工业出版社，1984.

China Academy of Building Research ． GBJ 68-84 Unified standard for design of building structures [S]．Beijing：China Architecture and Building Press，

1984.

[7] International Organization for Standardization. ISO2394: 1998 International Standard: General Principles on Reliability for Structures [S]. Switzerland : International Organization for Standardization ,1998.

[8 ] 茆诗松，王静龙，史定华．统计手册[M]．北京：科学出版社，2003.

MAO Shisong，WANG Jinglong，SHI Dinghua．Statistics handbook [M]．Beijing：Science Press， 2003.

[9 ] 茆诗松．贝叶斯统计[M]．北京：中国统计出版社，2005.

MAO Shisong．Bayesian Statistics [M]．Beijing：China Statistics Press，2005.

[10] JEFFREYS H．Theory of Probability [M]．London：Oxford University Press，1961.

[11] European committee for standardization. EN1990-2002：Eurocode-basis of structural design [S]. London, UK: BSI,2002.

相似文献/References:

[1]赵宪忠,李秋云.土木工程结构试验量测技术研究进展与现状[J].西安建筑科技大学学报：自然版,2017,49(01):48.[doi:10.15986/j.1006-7930.2017.01.008]
　ZHAO Xianzhong,LI Qiuyun.A review on measurement technology for structural testing in civil engineering[J].J.Xi’an Univ. of Arch. & Tech.：Natural Science Edition,2017,49(02):48.[doi:10.15986/j.1006-7930.2017.01.008]

备注/Memo

备注/Memo:
收稿日期：2015-03-06 修改稿日期：2016-04-08
基金项目：国家自然科学基金资助项目(50678143，51278401
作者简介：姚继涛（1965－），男，博士，教授，主要研究方向为工程结构可靠性．E-mail：yaojitao@163.com
更新日期/Last Update: 2016-05-18