|February 19, 2015, 06:07||
Optimization using Steepest Descent Algorithm
Join Date: Aug 2014
Posts: 3Rep Power: 2
Hi! I am trying to Optimize the shape of a diffuser in order to reduce the Pressure Drop across it. To do so, I am using Adjoint Flow Solver in STAR CCM+ which gives me the Sensitivities(Gradients, i.e. X,Y and Z Pressure Drop Derivative). I use that data to generate new points using the Steepest Descent Algorithm with the help of a formula Xnew = Xold - (Step Size * X Gradient) and respectively for Y and Z.
The problem which I'm facing is finding the Global Minimum of the Solution. Sometimes the Initial Geometry lies somewhere around the local minimum which doesn't leaves much scope for improvement.
Can anyone please put forward some suggestions?
|adjoint, algorithm, optimization, star ccm+|
|Thread||Thread Starter||Forum||Replies||Last Post|
|could not open file .vtk in paraview||ali_atrian||OpenFOAM Paraview & paraFoam||8||August 27, 2014 10:31|
|Paraview 3.98 - errors when saving geometry file||pajot||OpenFOAM Paraview & paraFoam||1||September 28, 2013 10:45|
|whats the general way to add an optimization algorithm?||immortality||OpenFOAM Running, Solving & CFD||5||March 29, 2013 02:22|
|COMSOL: optimization applied to a laminar flow||tdh89||Main CFD Forum||0||April 30, 2012 08:57|
|steepest descent and conjugate gradient||devalpandya||Main CFD Forum||0||April 23, 2009 04:25|