基于概率模型的抗力设计值推断方法
本文选题:结构设计 切入点:结构试验 出处:《西安建筑科技大学》2015年硕士论文 论文类型:学位论文
【摘要】:基于试验的设计方法是国际上提出的新的结构设计方法,它将结构试验引入具体的设计过程中,通过载荷试验和统计推断直接确定设计对象的抗力设计值,具有显著的实证性,适用于大批量相似构件、复杂结构和新型结构的设计,是目前结构设计方法发展的重要方向之一。基于分析模型的推断方法是基于试验的设计方法的重要内容,是推断抗力设计值的关键方法,它在推断中利用了抗力分析模型中已知影响因素的概率特性,可在较大程度上减小推断中的不确定性,得到更优的推断结果。但是国际标准ISO2394:1998中基于分析模型的推断方法是一种间接的推断方法,它采用了设计中的分析模型,未能直接反应各影响因素概率特性的影响,故在推断过程中不能保证抗力设计值具有一致的保证率,不利于对结构可靠度的控制;欧洲规范EN1990:2002中的方法虽然可在统一的保证率下推断抗力的设计值,但存在理论上的缺陷,是一种经验性方法,其推断结果偏于保守,甚至出现抗力设计值为零的情况。目前,国际上尚未很好的解决如何在统一的保证率下,以一种理论基础严密的方法推断抗力设计值的问题。本文针对目前国际上抗力推断方法的不足,直接依据抗力的概率模型,完全按照统计学的方法,提出了基于概率模型的抗力设计值推断方法。该方法在统一的保证率下,分别针对“计算模式不定性系数的变异系数未知”和“计算模式不定性系数的变异系数已知”两种情况提出抗力设计值的推断公式,不但具有严密的理论基础,而且克服了之前抗力推断方法的缺点。为了检验该方法的适用性,本文首先通过引入一个工程实例,分别利用区间估计法和文中方法对受弯构件承载力的设计值进行推断,并对其推断结果进行分析。然后在相同保证率下,对比了欧洲规范中方法与本文方法的推断结果,对比结果表明,基于概率模型的抗力设计值推断方法不仅符合统计学的一般规律,且可给出适宜的推断结果。基于概率模型的抗力推断方法发展了基于分析模型的抗力推断方法,是对目前结构构件抗力推断方法的一个补充。
[Abstract]:The design method based on test is a new structural design method put forward in the world. It introduces the structural test into the concrete design process and directly determines the resistance design value of the design object through load test and statistical inference. It is one of the most important directions in the development of structural design methods to apply to the design of large quantities of similar components, complex structures and new structures. The inference method based on analytical model is an important part of the experimental design method. It is a key method to infer the design value of resistance. It makes use of the probability characteristics of known factors in the model of resistance analysis, which can reduce the uncertainty in inference to a large extent. But the inference method based on analytical model in the international standard ISO2394:1998 is an indirect inference method, which adopts the analysis model in the design, and fails to directly reflect the influence of the probability characteristics of each influence factor. Therefore, in the process of inference, the design value of resistance can not be guaranteed consistently, which is not conducive to the control of structural reliability. Although the method in European Specification EN1990:2002 can infer the design value of resistance under the unified guarantee rate, it has some theoretical defects. It is an empirical method, whose inference results are conservative, and even the resistance design value is zero. At present, the international community has not solved the problem well under the unified guarantee rate. The problem of inferring the design value of resistance by a method with a strict theoretical basis is discussed in this paper. In view of the deficiency of the present method of resistance inference in the world, the method is directly based on the probability model of resistance and is completely in accordance with the statistical method. A method of inferring resistance design value based on probabilistic model is proposed. In the case of "unknown coefficient of variation of uncertain coefficient of calculation mode" and "coefficient of variation of uncertain coefficient of calculation mode known", the inferring formulas of resistance design value are put forward respectively, which not only have a strict theoretical foundation, In order to test the applicability of the method, an engineering example is introduced to infer the design value of the bearing capacity of flexural members by using interval estimation method and the method in this paper, respectively. Then, under the same guarantee rate, the inferred results of the methods in European norms and the methods in this paper are compared, and the comparison results show that, The inference method of resistance design value based on probabilistic model not only accords with the general rule of statistics, but also gives the appropriate inferential result. The resistance inference method based on probabilistic model has developed the resistance inference method based on analytical model. It is a supplement to the current method of estimating the resistance of structural members.
【学位授予单位】:西安建筑科技大学
【学位级别】:硕士
【学位授予年份】:2015
【分类号】:TU317
【相似文献】
相关期刊论文 前10条
1 李启信,W.White,楚剑,赵敏敏;土层的概率模型及其在桩基分析中的应用[J];岩土工程学报;1989年06期
2 徐人平;叶司一;李淑兰;;断裂韧性的概率模型[J];云南工学院学报;1993年01期
3 王阜;地震发生的空间概率模型[J];地震工程与工程振动;1986年03期
4 任飞正,孔力;客运量的合成预测与概率模型[J];铁道运输与经济;2003年02期
5 刘任先;路面不平的概率模型[J];农业机械学报;1990年01期
6 贺红艳,周杰娜;基于概率模型和博弈模型的发电厂报价策略研究[J];贵州工业大学学报(自然科学版);2004年01期
7 徐楠;王威强;王胜春;李艳;;疲劳概率模型的综合评价方法[J];中国机械工程;2008年23期
8 胡泳芬;出游法中概率模型的改进[J];江西电力;1997年04期
9 梁磊;;船撞桥概率模型分析[J];物流工程与管理;2009年02期
10 井浩涌,,刘军,潘文新,张玉德;群体决策的概率模型与分组德尔菲方法[J];大连铁道学院学报;1995年03期
相关会议论文 前8条
1 徐道远;张林兵;符晓陵;寿朝晖;;混凝土裂缝失稳扩展的概率模型研究[A];第二届全国结构工程学术会议论文集(上)[C];1993年
2 王冰洁; 沈玉清;;中国上市公司被收购风险概率模型的实证分析[A];中国灾害防御协会风险分析专业委员会第二届年会论文集(一)[C];2006年
3 季Y
本文编号:1574994
本文链接:https://www.wllwen.com/jingjilunwen/jianzhujingjilunwen/1574994.html