Reliability Analysis of Foundation Pit by Improved Response Surface Method

Article Preview

Abstract:

Lacking of explicit limit state function (LSF) will result large quantities of computational efforts for a FEAM based structural reliability analysis. An improved response surface (RS) method is proposed to analyze the failure probability of foundation pit through combining uniform design (UD) and non-parametric regression (NPR). Deferent levels of design parameters are first delicately selected according to UD and then FEAM is used to analysis corresponding pit response parameters including maximum lateral displacement of wall, settlement of ground, safety factor of overall stability, safety factors of against overturning, heave and piping. The RS relationship is then established through NPR based on inputs and responses. At last, a direct Mont Carlo Simulation is carried out to obtain the probability density function of response parameters.

You might also be interested in these eBooks

Info:

Periodical:

Pages:

197-202

Citation:

Online since:

December 2011

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2012 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

[1] China Ministry of Construction. Guideline of risk management for construction of subway and underground works [C]. Beijing: China Architecture & Building Press(2007).

Google Scholar

[2] Søren Degn Eskesen, Per Tengborg, Jørgen Kampmann, Trine Holst Veicherts. Guidelines for Tunnelling Risk Management[C]. International Tunnelling Association, Working Group No. 2(2002).

DOI: 10.1016/j.tust.2004.01.001

Google Scholar

[3] Huang HW, Xie XY et. al. Risk management in construction of transportation rail engineering and its application [M]. Tongji University Press (2009).

Google Scholar

[4] Zhao Guo-fan, Jin Wei-liang, Gong jin-xin. The theory of structure reliability [M]. Bejing: China Architecture and Building press (2000).

Google Scholar

[5] Fang Kai-tai., Ma Chang-xin. Orthogonal and uniform experimental design [M]. Beijing: Science press (2001).

Google Scholar