Department of Energy
University Information
Collaborators
Predictive Engineering and Computational Sciences
201 E. 24th Street
Austin, TX, 78712
Phone: 512-471-3168
Fax: 512-232-8067
Site Administrator
201 E. 24th Street
Austin, TX, 78712
Phone: 512-471-3168
Fax: 512-232-8067
Site Administrator
Solution Verification
Activities
Modeling Domains
In computational science, one of the most important considerations is whether a numerical solution of a problem is a sufficiently accurate representation of the exact solution. Moreover, one’s assessment of the accuracy of the solution depends on the quantities one is interested in determining from the solution. The critical step in solution verification is thus to determine an estimate of the errors in the quantities of interest. Such an a posteriori error estimate can be found from the solution of an adjoint problem, and adaptive grid refinement can by driven based the estimates. Goal-oriented error estimators and refinement such as this are well-established for relatively simple systems. The challenge in the PECOS project is to apply these techniques to a complex multi-physics problem such as a reentry vehicle.
A related topic in verification, “Code Verification“, is supported at the PECOS center through the use of the MASA library.