建筑楼面活荷载的小样本统计推断

发布时间：2018-04-02 14:37

本文选题：楼面活荷载　切入点：标准值　出处：《西安建筑科技大学》2013年硕士论文

【摘要】：建筑结构楼面活荷载标准值的推断是结构可靠性评定工作中的一项非常重要的工作，是既有结构进行改造加固设计的重要依据，也是确定荷载规范中拟建结构楼面活荷载标准值取值的重要依据。在对建筑楼面活荷载标准值进行推断的时候，若推断值过大，会造成材料的浪费、成本的提高；若推断值过小，则会导致结构可靠性的降低、安全性的不足。因此合理准确的推断建筑楼面活荷载标准值是非常重要的。目前我国对大样本情况下的楼面活荷载标准值的推断理论和方法已经比较成熟，我国的规范已给出了建筑楼面活荷载标准值的推断方法。然而规范给出方法是以大样本为前提条件的，但在工程实际中并不只存在大样本的情况，有时会因为各种原因使得样本容量不足而出现小样本问题，这时我们将不能再按照规范给出的大样本方法进行推断，，否则推断结果将不够准确，不能满足实际情况的要求。而目前小样本条件下的建筑楼面活荷载标准值的推断还没有很成熟的理论方法。针对至今还没有很好地解决的小样本问题，本文提出了两种方法：第一种是利用极大值I型分布分位值的估计方法，建立的小样本条件下建筑楼面活荷载标准值的推断方法，可称之为线性回归推断方法；第二种方法则是以信度的相关理论为基础，提出的小样本条件下建筑楼面活荷载标准值的贝叶斯推断方法，并对方差已知和方差未知两种情况分别进行了推导，给出了推导结果。文章最后以某商场楼面活荷载调查统计的数据为基础，按照三种方法分别对其进行了推断分析，并对推断结果进行了对比分析，得出了结论。通过对比分析可以发现线性回归方法相对比较准确，而贝叶斯方法的误差较大，但相对比较保守。文章最后给出了两种小样本推断方法中一些参数的推荐取值。论文的研究成果为小样本条件下楼面活荷载标准值的推断提供一种较为合理的方法，有利于对既有结构的楼面活荷载标准值做出更合理的估计，为确定拟建结构的楼面活荷载标准值提供了另一种方法，可为我国现行荷载标准的修订提供理论和方法上的参考。
[Abstract]:The estimation of the standard value of the living load on the floor of the building structure is a very important work in the reliability evaluation of the structure, and it is also an important basis for the reconstruction and reinforcement design of the existing structure.It is also an important basis for determining the standard value of the living load of the proposed structure floor in the load code.When inferring the standard value of living load of building floor, if the inferred value is too large, it will cause the waste of materials and the increase of cost, and if the inferred value is too small, it will lead to the reduction of structural reliability and the shortage of safety.So it is very important to infer the standard value of building floor live load reasonably and accurately.At present, the theory and method to infer the standard value of floor live load under the condition of large sample in our country has been quite mature, and the standard value of living load of building floor has been given in the code of our country.However, the method given in the specification is based on large samples, but there are not only large samples in engineering practice, but sometimes small sample problems occur due to the insufficient sample size for a variety of reasons.In this case, we will no longer be able to infer according to the large sample method given by the specification, otherwise, the inference results will not be accurate enough to meet the requirements of the actual situation.At present, there is no mature theoretical method to infer the standard value of living load of building floor under the condition of small sample.Aiming at the problem of small sample which has not been solved well up to now, this paper puts forward two methods: the first method is to estimate the standard value of living load of building floor under the condition of small sample by using the method of estimating the quantile of maximum I distribution.It can be called linear regression inference method, the second method is based on the theory of reliability, and the Bayesian inference method for the standard value of living load of building floor under the condition of small sample is put forward.The known variance and unknown variance are derived, and the results are given.Finally, based on the survey data of living load on the floor of a shopping mall, three methods are used to infer and analyze the living load of a shopping mall, and the inferred results are compared and analyzed, and the conclusion is drawn.Through comparative analysis, we can find that the linear regression method is relatively accurate, while the Bayesian method has a large error, but is relatively conservative.Finally, the recommended values of some parameters in two small sample inference methods are given.The research results of this paper provide a more reasonable method for estimating the standard value of floor live load under the condition of small sample, which is helpful to make more reasonable estimation of the standard value of floor live load of existing structure.Another method is provided for determining the standard value of floor live load of the proposed structure, which can be used as a theoretical and methodological reference for the revision of current load standards in China.
【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】：2013
【分类号】：TU312

【参考文献】