Sensitive Analysis of Soil Parameters on Stability Numbers

Kumar, Bimlesh and Samui, Pijush (2007) Sensitive Analysis of Soil Parameters on Stability Numbers. In: Civil Engineering in the New Millennium:Opportunities and Challenges, 11-14 January, West Bengal,India.

Preview

PDF GT009.pdf Download (126kB)

Abstract

In general, Stability number can be expressed as a function comprising of pore water pressures coefficient $(r_u)$, horizontal earthquake coefficient $(k_h)$, slope angle $(\beta)$ and friction angle $(\phi)$ in associated flow rule. Then it is important to know how different parameters influence the stability number. These influences may be obtained through the sensitivity analysis of the parameters. Due its facility in solving non linear problems, Artificial Neural Networks (ANN) has been proposed, as a powerful tool, to represent the stability number. Neural Networks are typically thought of as black boxes trained to a specific task on a large number of data samples. In many applications it becomes necessary to ook inside of these black boxes before they can be used in practice. Sensitivity analysis is one of best ways to look inside the neural model and ascertain that which parameters are having more effect on the output. Sensitivity analysis is a method for extracting the cause and effect relationship between the inputs and outputs of the network. The basic idea is that each input channel to the network is offset slightly and the corresponding change in the output(s) is reported. So, through differentiation of a previously trained net, the sensitivity factors of the main parameters of solar collectors is calculated and discussed. The sensitivity factors show how much the input variables influence the output variables. In this paper, the sensitivity analysis for stability number’s main parameters is discussed. The result indicates that the ru is having more influence on stability number than any other parameter followed by \beta $k_h$ and \phi. The stability numbers have been found to decrease continuously with (i) increase in $r_u$, (ii) increase in kh and (iii) increase in slope angle $(\beta)$ (iv) decrease in \phi.

Item Type:	Conference Paper
Keywords:	Slope stability;Artificial Neural Network;Stability number
Department/Centre:	Division of Mechanical Sciences > Civil Engineering
Date Deposited:	25 Aug 2008
Last Modified:	19 Sep 2010 04:34
URI:	http://eprints.iisc.ac.in/id/eprint/9623

Actions (login required)

View Item