EVOLUTION OF PROBABILITY DISTRIBUTIONS IN DIFFERENT CONSTRCTION STAGES -THE SIMULATION OF A FOUNDATION PIT
The risk of foundation pit often changes along with the construction process,and estimating risk accurately in various stages is important to ensure construction safety and to implement monitoring construction.For this,based on a new response surface (RS) method,Monte Carlo Simulation (MCS) is used for risk analysis in various stages.Firstly,The RS construction is through uniform experiment design (UD) and non-parametric regression (NPR),thus more accurate RS for different stages is obtained with fewer calculations.Further based on different RS,MCS is applied for obtaining the evolution of probability distribution for risk analysis along with various stages.With a practical project,the above methods are used to simulate evolution of probability distribution of pile deformation in various stages.The result suggests that the proposed method can provide an efficient approach to simulate the dynamic evolution of probabilities in different construction stages.Because of the introducing of UD and NPR,the RS functions can contain more changing parameters and be constructed precisely with fewer calculations.These features help solve the complex reliability problems of those practical engineering.
Deformation foundation pit evolution umiform experiment ACE Monte Carlo
Haiying Wang Hongwei Zheng Jing Cao Zhigang Song
Faculty of Civil Engineering, Kunming University of Science and Technology, Kunming 650500, P.R.China
国际会议
The Twelfth International Symposium on Structural Engineering (第十二届结构工程国际研讨会)
武汉
英文
411-416
2012-11-17(万方平台首次上网日期,不代表论文的发表时间)