您目前所在的位置:首页 - 期刊简介 - 详细页面

铁道科学与工程学报

JOURNAL OF RAILWAY SCIENCE AND ENGINEERING

第11卷    第5期    总第61期    2014年10月

[PDF全文下载]    [Flash在线阅读]

    

文章编号:1672-7029(2014)05-0112-06
基于稳健估计的沉降预测方法
彭仪普1,伍绍浩2,张莹超2

(1.中南大学 土木工程学院,湖南 长沙 410075;
2.中南大学 地球科学与信息物理学院,湖南 长沙 410083
)

摘 要: 双曲线法和Asaoka法在对沉降数据进行预测时,需要对原始数据进行拟合处理,且多采用最小二乘法进行拟合处理。但当沉降数据中混入粗差后,采用最小二乘法拟合存在失准的情况。针对这一问题,引入稳健线性估计理论,在数据预处理过程中采用稳健线性拟合法。并探讨混入粗差后,最小二乘线性拟合与稳健线性拟合分别在双曲线法、Asaoka法的预测效果和适用性。研究结果表明:采用稳健线性拟合法进行数据处理受粗差的影响极小,能起到很好的抵抗粗差的作用;当拟合数据过少时,最小二乘法线性拟合与稳健线性拟合效果无明显差异。

 

关键字: 沉降预测;双曲线法;Asaoka法;最小二乘估计;稳健估计

Settlement prediction based on robust estimation
PENG Yipu1,WU Shaohao2,ZHANG Yingchao2

1. School of Civil Engineering, Central South University, Changsha 410075, China;
2. School of Geosciences and Info-Physics, Central South University, Changsha 410083, China

Abstract:Least square method is adopted to improve the fitting process of the original data when using hyperbolic method and Asaoka method to predict the settlement. But after settlement data with gross error, the least square method fitting results are inaccurate in certain cases. Aiming at this problem, this paper introduced robust linear estimation theory to deal with gross error in the data pre-processing. After discussion of gross error, the applicability of least square linear fitting and robust linear fitting in the prediction were also discussed respectively. Research results show that the influence of gross error plays a minimal role when adopting the robust linear fitting method for data processing, which indicates that this approach can play a very good resistance ability of gross error; when the amount of data is too small to fit, the least squares linear fitting and robust linear effect have no significant difference.

 

Key words: settlement prediction; hyperbola method; Asaoka method; least squares estimation; robust estimation

ISSN 1672-7029
CN 43-1423/U

主管:中华人民共和国教育部 主办:中南大学 中国铁道学会 承办:中南大学
湘ICP备09001153号 版权所有:《铁道科学与工程学报》编辑部
------------------------------------------------------------------------------------------
地 址:湖南省长沙市韶山南路22号 邮编:410075
电 话:0731-82655133,82656174   传真:0731-82655133   电子邮箱:jrse@mail.csu.edu.cn