Software Version 1.1.2

for a complete description of the following software.

*Written by*

Professors: | Dr. Gordon A. Fenton | Dr. D.V. Griffiths | ||

EMAIL: | Gordon.Fenton@dal.ca | d.v.griffiths@mines.edu | ||

Phone: | 902-494-6002 | 303-273-3669 | ||

Location: | Dalhousie University | Colorado School of Mines | ||

Halifax, Canada | Golden, Colorado |

**
See also the complete list of RFEM Publications.
**

In 1992, Drs. G.A. Fenton and D.V. Griffiths combined random field simulation with the finite element method to produce the Random Finite Element Method, or RFEM. The method has been used to investigate the random behaviour of soils in the context of a variety of classical geotechnical problems, ranging from settlement of shallow foundations to the collapse of rock pillars. All but the most recent publications arising out of this research can be found here. (For the most recent publications, you will have to obtain the journal versions.)

The authors are making all of their RFEM software freely available, along with the finite element libraries written by I.M. Smith and D.V. Griffiths (which have been public for many years) and the simulation and numerical library written by G.A. Fenton.

When results derived from the RFEM software available at this site are used in any publication, the authors must cite at least one of the journal papers as well as the textbook written by the developers. The appropriate references can be found in RFEM Publications.

The RFEM front end executables (under Windows) include the following;

- mrbear2d: 2-D shallow foundation stochastic bearing capacity analysis,
- mrdam2d: 2-D stochastic earth dam analysis,
- mrearth2d: 2-D stochastic earth pressure analysis,
- mrflow2d: 2-D stochastic seepage analysis,
- mrflow3d: 3-D stochastic seepage analysis,
- mrpill2d: 2-D stochastic pillar analysis,
- mrpill3d: 3-D stochastic pillar analysis,
- mrsetl2d: 2-D shallow foundation stochastic settlement analysis,
- mrsetl3d: 3-D shallow foundation stochastic settlement analysis,
- mrslope2d: 2-D stochastic slope stability analysis,

In order to look at some of the output of these programs, you will need to
install a Postscript viewer. See
here for the required
software (or search for ghostscript and ghostview for the most recent
versions). You will need to install ghostscript first and then ghostview. The
32bit executables should work on any computer, the 64bit versions are tailored
for 64bit CPUs. Install only one set or the other. *Remember what folder you
install the Postscript viewing program into* -- you will need to know the
folder and the executable name in order to get the RFEM programs to view
graphical results -- this information must be provided in the "rfem.inf" file
discussed shortly. Use Windows Explorer to find the executable, which is
(probably) called gsview32.exe (or gsview64.exe).

The complete current version of the RFEM distribution for the Windows environment is RFEM_Windows_1.1.2.exe, which is a self-extracting executable. The easiest way to install the RFEM distribution is to extract it into the "c:" drive -- do this just by running the downloaded executable and accepting the default options. This will create a subdirectory called "c:\rfem" within which all of the executables, libraries, and source codes are placed. If you wish to install rfem on a different drive, just change "c:" to whatever drive you wish. In the following, the c: drive is assumed.

REVISION HISTORY:

Version 1.1.2 (Aug 29, 2012):
mrslope2d.exe was broken (opened off the edge of the screen) -- fixed in this
revision.

Version 1.1.1 (Jan 8, 2012):
Added the ability for rslope2d to consider any positive gradient
(rather than just integer gradients).

Version 1.1.0 (Oct 13, 2011):
I discovered an element numbering error in one
of the LAS3G subroutines (plan3d.f) which led to an incorrectly generated 3-D
LAS random field in the event that k1 = 1. This would have occurred , for
example, when trying to generate a field of size 4 x 64 x 64, i.e. a very
narrow field in the x-direction. Similarly, in the 2-D LAS generator, when
trying to generate a field of size 2 x 512, the final set of 4 cells were not
being averaged to the correct final parent cell. Both problems have been
corrected in this version.

After extracting the distribution, you will need to perform the following steps;

- edit c:\rfem\rfem.inf using an ordinary editor (e.g. Notepad,
*not*MS Word). - set BINDIR to the path to the rfem\bin folder (c:\rfem\bin by default),
- set HELPDIR to the path to the rfem\lib\help folder (c:\rfem\lib\help by default),
- set GHOSTVIEW to the path and executable name of a Postscript viewing program (see discussion about GSView above), the default is c:\ghostscript\gsview\gsview32.exe.
- move or copy the rfem.inf file into the directory c:\windows\inf.

To uninstall the RFEM distribution simply remove the rfem directory and the c:\windows\inf\rfem.inf file (if you copied the latter over).

Now, if you look in the rfem\bin directory, you should be able to run any of the programs listed above, e.g. mrsetl2d.exe. Some hints at running these programs;

- to get the output to appear in a specific folder, simply set the basename to include the folder path, e.g. d:\gordon\temp\rsetl2d
- virtually all of the entries have a little button beside it labeled "Help". If in doubt, click on it.
- the non-linear programs (mrbear2d, mrslope2d, mrpill?d, mrearth2d) may take a LONG time to run. You may want to test it using just a few realizations to get a sense for timing prior to cranking the final run(s).
- the leading "m" in these programs stands for "graphical front end" (in whatever language "graphical front end" starts with the letter "m"! Actually, it was sort of a mistake -- I usually name just the main routine of a program with a leading "m", and my early attempts at Borland C++ Builder named the executable the same name as the main...). Anyhow, what this means is that when you "run" mrbear2d, for example, by clicking on the little black triangle, it actually executes another program named "rbear2d" which reads the data file you have just created and runs the RFEM simulations. "rbear2d.exe" is in the rfem\bin directory. The source code for rbear2d.exe is in rfem\sim\rbear2d. The source code for mrbear2d.exe is in rfem\sim\rbear2d\gui.

Also included in this distribution are the following;

- DVGlib: double precision finite element library, written by I.M. Smith and D.V. Griffiths.
- VGlib: single precision finite element library, written by I.M. Smith and D.V. Griffiths. The compiled version of this library is in rfem\lib\vfem.lib. This library is needed to recompile any of the r-progs, e.g. rbear2d.
- doc: documentation for; the r-progs, a series of r-prog utilities (e.g. hist_* are programs designed to create histograms of the output samples from the r-progs), plotps, display, and a bunch of other things which you should be able to find somewhere in this distribution.
- gaf77: simulation routines (e.g. LAS) written by Gordon A. Fenton. This library includes LOTS of other things -- a complete list of the routines can be found in rfem\gaf77\Summary.txt, which is an ordinary text file. Each routine has a lengthy commentary at its start explaining the routine in detail. The compiled library is in rfem\lib\gaf77.lib. It is needed if you want to recompile virtually all of the programs in this distribution (NOTE: to recompile a particular program, you may need to edit its "Makefile" to tell it where the library is prior to running "make" -- however, all of the Makefiles are now "drive independent" so as long as \rfem\lib is on the same drive as the program being compiled there should be no problem).
- graphics: this folder actually includes three
subfolders, all written by Gordon A. Fenton:
- pslib: a postscript plotting library. See the files PSLIB.3f (text), PSLIB.ps (postscript), or Summary.txt (text) in rfem\graphics\pslib for more details.
- plotps: a program which produces x-y type plots in postscript. See rfem\doc\plotps.txt for more details.
- display: a program which produces plots of 3-D data in postscript. See rfem\doc\display.txt for more details.

- lib: where the compiled libraries and help files are stored.
- sim: includes a series of sub-folders which
contain the source code for various programs. The following are the
r-progs written by Gordon A. Fenton and D.V. Griffiths;
- rbear2d: 2-D shallow foundation stochastic bearing capacity analysis,
- rdam2d: 2-D stochastic earth dam analysis,
- rearth2d: 2-D stochastic earth pressure analysis,
- rflow2d: 2-D stochastic seepage analysis,
- rflow3d: 3-D stochastic seepage analysis,
- rpill2d: 2-D stochastic pillar analysis,
- rpill3d: 3-D stochastic pillar analysis,
- rsetl2d: 2-D shallow foundation stochastic settlement analysis,
- rsetl3d: 3-D shallow foundation stochastic settlement analysis,
- rslope2d: 2-D stochastic slope stability analysis,
- rpile1d: 1-D deep foundation stochastic analysis. This is a recent addition, with no corresponding graphical front-end at this time.

- rfest1d: a suite of programs developed by Gordon A. Fenton while at the Norwegian Geotechnical Institute to statistically analyze CPT soundings.
- simqke2: a program developed by Gordon A. Fenton, Erik H. Vanmarcke, and Ernesto Heredia to simulate spatially correlated and optionally conditioned earthquake ground motions
- test routines: a set of folders (test* and tst*) containing programs designed to run the various random field generators. They are primarily included here for those of you interested in writing your own simulation programs -- these can be used as templates.
- utils: a series of programs to compute things such as local average variances, generate random variables, and compute normal distribution probabilities. See the Makefile for brief descriptions of each program.

- utils: this folder includes several
subfolders:
- cor2d: a program which estimates a 2-D correlation structure, written by Gordon A. Fenton. See cor2d.1 (text) for details.
- krige: a program which computes best linear unbiased estimates (kriging), written by Gordon A. Fenton. See krige.1 (text) for details.
- mkhst: a program which plots a histogram of data with optional fitted distribution, written by Gordon A. Fenton. See mkhst.1 (text) for more details.
- mmat: a program which performs various matrix operations (e.g. multiplication, Cholesky decomposition, LU decomposition, solution to linear system, matrix inverse, condition number, singular-value-decomposition). Written by Gordon A. Fenton. See mmat.1 (text) for more details.
- psd: a program which estimates a power spectral density function, written by Gordon A. Fenton. See estpsd.1 and psdana.1 (text) for details.
- reg1d: a program which performs a 1-D regression, written by Gordon A. Fenton. See regr1d.1 (text) for details.
- reg2d: a program which performs a 2-D regression, written by Gordon A. Fenton. See regr2d.1 (text) for details.
- stats: two programs which estimate mean, variance, and covariance, written by Gordon A. Fenton. See mncov.1 and mnvar.1 (text) for details.

NOTE: THIS SOFTWARE COMES WITH NO WARRANTY. YOU ARE WELCOME TO COPY AND/OR MODIFY THE CODE TO SUIT YOUR NEEDS, BUT THE NAMES OF THE ORIGINAL AUTHORS (WHO ARE COPYRIGHT HOLDERS) MUST REMAIN WITH THE CODE (IN THOSE CODES WHERE THESE NAMES ARE CURRENTLY PROVIDED).

FOR EACH PROGRAM INCLUDED IN THIS DISTRIBUTION AND/OR DERIVED FROM THIS DISTRIBUTION, THE FOLLOWING PROVISIONS HOLD:

BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION. YOU ARE WELCOME, HOWEVER, TO CONTACT THE PROGRAM AUTHORS TO HELP PROVIDE CORRECTIONS.

IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING WILL ANY COPYRIGHT HOLDER OR AUTHOR, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.