Dear all, I see that Sage has some ODE solvers from GNU Scientific Library.
Is there any DAE solver readily available to Sage? I have here a second order nonlinear system which boils down to some system as: M(y) ydot = f(t,y) My mass matrix is state dependent, not time dependent (yet). It is also nondiagonal. The 'y' vector represents some generalized coordinates resulting from a Galerkin discretization of a electromechanical problem. I've being doing the symbolic calculation on Maxima down to the above system. Up to 3 modal coupled equations I can invert the M(y) and rewrite things as a coupled ODE an run the native RK4 on Maxima. I don't like the invert step since I don't know if M(y) will be invertible all the time. It indeed breaks down if I try to run it with higher number of modes. I did a dry test hand copying the DAE system from Maxima output and plugged into Matlab ode15s, ode23t. It works as expected, but I also don't like the Matlab step. My plan is to bring all the symbolic and numerical time integration of this DAE system into Sage. Later I will be investigating the stability and nonlinear dynamics of this system. Therefore I need a DAE solver which I didn't find on Sage manual or mailing lists. It seems that Mathematica does the DAE job with the IDA solver from Sundials. I saw two interfaces around: Python-Sundials and PySundials. I've seen that the PyDSTool also has a Radau5 Python wrapped already. Are there other options? Comments are welcomed. Kind regards, Guilherme Brondani Torri -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
