Hi Mohsen
"Probability Analysis of Slope Stability"
try to contact prof Helmut Schweiger [log in to unmask]
or dr. Robert Thurner [log in to unmask]
sincerely
Radu
SIMON CROXFORD schrieb:
> Mohsen
>
> Another way of getting around the lack of statistically valid data indicated by Jonathan is to use a large data set that has been published in the public domain. Such published data could be on the effective stress parameters of burnt and unburnt colliery spoil in the UK. (Composition & engineering properties of British colliery discards, Taylor, 1984). Where the large number of effective stress tests carried out by the former National Coal Board have all been published. This would provide statisitically valid data as well having the additional advantage of providing a large number of man made slopes of which there are published case studies of both successes and failures.
>
> Simon Croxford
> Engineering Geologist
>
> IMC Consulting Engineers Ltd
> PO Box 18
> Mill Lane
> Huthwaite
> Sutton in Ashfield
> Notts NG17 2NS
>
> Tel : 01623 444611
> Fax : 01623 440021
> E Mail : [log in to unmask]
>
> >>> Jonathan Welch <[log in to unmask]> 12/03/01 20:36:17 >>>
> Hi Mohsen
>
> "Probability Analysis of Slope Stability"
>
> Finding a suitable distribution can be difficult due to a shortage of good
> data. Effective stress testing is both expensive and time consuming and so
> you are unlikely to get enough data for sophisticated statistical analysis.
> However, here are some suggestions to get you started. For geometrical data
> consider using a uniform distribution; if you need answers for a range of
> angles, heights and lengths, maybe no one value is any more likely than
> another. For soil parameters consider using a triangular distribution in
> which you specify the lowest, most likely and maximum likely values. I have
> found that the resulting solutions are not significantly different from
> those obtained using normal distributions. Try it out for yourself! For
> parameters that depend upon several multiplier factors such as the
> coefficient of permeability (these properties are commonly known "order of
> magnitude" parameters) you can use logarithmic triangular distributions.
> Note that some people do not use Monte Carlo simulation for parameters which
> are simply just not well known, prefering to specify distributions only for
> parameters that show a natural variability.
>
> Good luck with your project, I am looking forward to learning from other
> contributors!
> Jonathan Welch
>
> ----- Original Message -----
> From: "mohsen sabermahani" <[log in to unmask]>
> To: <[log in to unmask]>
> Sent: Sunday, March 11, 2001 11:14 AM
> Subject: probability analysis of slope stability
>
> > Hi Dear All,
> >...
> > Now, I would like to have specifications of slided
> > slopes(such as geometry, soil parameters:
> > c,phi,gama,...) as case studies.
> > ...
> > Best wishes.
> > M.Sabermahani
--
Dr. Radu Schwab
Bundesanstalt für Wasserbau / Federal Waterways
Engineering and Research Institute
Kussmaulstr. 17
76187 Karlsruhe, Germany
Phone: +49 (0)721 9726-3810
URL: www.baw.de
|