沉降计算经验系数的Bootstrap法置信区间估计

来源期刊:中南大学学报(自然科学版)2011年第9期

论文作者:谢桂华 张家生 刘荣桂

文章页码:2843 - 2847

关键词:沉降;经验系数;小样本;Bootstrap法;置信区间

Key words:settlement; empirical coefficients; small sample; Bootstrap method; confidence interval estimation

摘    要:采用Bootstrap法推断小样本条件下的沉降计算经验系数概率特征及置信区间,以解决工程中沉降计算经验系数样本量很少,难以应用常规统计方法分析其概率分布的问题。直接对样本重抽样,根据Bootstrap样本逼近参数真实分布,并纠正Bootstrap仿真过程中出现的偏差,以获得参数的无偏置信区间。通过某客运专线路基工程实例分析,得出:采用Bootstrap法可较好地推断小样本条件下沉降计算经验系数未知参数的概率分布,获得参数的置信区间;当面荷载在40~240 kPa范围内取值时,红黏土地基的沉降计算经验系数均值随载荷的增大而减少,经验系数均值服从正态分布,置信度为95%的经验系数均值的置信区间按荷载范围依次为[0.637, 0.884],[0.637, 0.884]及[0.901, 1.165]。

Abstract:

Because of the scarcity of samples in engineering and the difficulty in inferring the probabilistic distribution of empirical coefficient by regular methods during the calculation of foundation settlement, the probabilistic distribution and confidence interval of empirical coefficient were determined by Bootstrap method. By re-sampling from the existing samples, the true distributions of parameters were approximated by Bootstrap method and the possible deviation was corrected. As a result, unbiased confidence interval estimation was obtained. A subgrade of a passenger dedicated railway was given to illustrate the proposed method and the results show that, under the condition of small sample of empirical coefficient in calculating foundation settlement, Bootstrap method is efficient to estimate the probability distributions and confidence intervals of unknown parameters; when the range of surface load is 40-240 kPa, the empirical coefficients of foundation settlement decrease with increasing loads for red clay; the mean value of empirical coefficients obeys normal distribution, and its confidence interval is [0.637, 0.884], [0.637 0.884] and [0.901, 1.165] respectively at the confidence of 95%.

相关论文

  • 暂无!

相关知识点

  • 暂无!

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号