[Getdp] ... improving GetDP...

ALEX RIBEIRO bakali_1st at yahoo.com
Sat Mar 17 21:48:43 CET 2007


Hi, my name is Alex, and i am a graduate engineering student at departement of mechanical engineering at Univesidade de Coimbra.

i have recently downloaded GetDP and Gmsh so as to use them in a neural-genetic framework of structural multidisciplinary optimization tool in mechanical design:

The Neural network builds a scalar mapping that holds the "cultural belief" of genetic compatibility between individuals in a population, its currently a 3 layer backprop network.

The Genetic Algorithm explores the decision space of multidisciplinary designs, using Variable population size, fized mutation and crossover rates and a parametric tuning between exploration or exploitation of design space, i.e: generalization through unbiased crossover, or specification through seduction (culturally biased crossover by Neural Network).
   
  The structure of the framework will be built around a parametric geometry builder mesher, GMSH, and a structural design function evaluator, GetDP, using a loop scripted in TCSH wilh file (population individual) parsing and updating with awk and Vi.
   
  I would like to know if GetDP can be updated or upgraded in terms of its solver capabilities to solve shock dynamics problems, formulated by the equations of Euler, particularly Radiation Shock Hydrodynamics in dense plasmas, so as to evaluate or quantify the onset of a thermonuclear ignition process in a Z-Pinch plasma implosion.
  That is charaterized by the Lawson parameter, and the transport of energy by the imploding cylindrical shock wave.
  (it is an infinite electromagentic cumulation process ocurring in a column of plasma under a strong electric field and electric current...)
   
  If it is possible, i would like how to contribute to such endeavour...
   
  Thanks in advance...
   
  Alexandre Ribeiro  


 
---------------------------------
The fish are biting.
 Get more visitors on your site using Yahoo! Search Marketing.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.geuz.org/pipermail/getdp/attachments/20070317/5220bd9f/attachment.html>