On Fri, Jun 3, 2016 at 1:34 PM, Harriet Li <kame...@gmail.com> wrote:
> Hello all, > > I'm looking to solve a nonlinear problem where the nonlinearity is > controlled by some parameter. The Newton solver does not converge at the > desired parameter value without a better initial guess than I have. > > Is this a problem I can apply the ContinuationSystem class to? Can I solve > the problem at an initial parameter value and use the class's methods to > work towards a solution at a designated final parameter value? (I don't see > a way to tell the continuation where to stop...) > The ContinuationSystem can be used for this. Note that it does arclength continuation (see the Keller reference in continuation_system.h) so it might be overkill for what you are doing if there are no turning points/bifurcations in the nonlinear system (i.e. you just want to slowly ramp up the nonlinear parameter and re-solve the system multiple times). As far as when it should "stop", this is controlled by the Real min_continuation_parameter; Real max_continuation_parameter; parameters in the class. -- John ------------------------------------------------------------------------------ What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic patterns at an interface-level. Reveals which users, apps, and protocols are consuming the most bandwidth. Provides multi-vendor support for NetFlow, J-Flow, sFlow and other flows. Make informed decisions using capacity planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e _______________________________________________ Libmesh-users mailing list Libmesh-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/libmesh-users
