Cap Maximum ODE solve time

I am solving a set of ODEs with several different initial conditions some of which will be badly conditioned and get bogged down. I want to cap the run time of an individual solve to say 100 seconds. If it reaches 100 seconds, I want to exit and just mark the solve as fail. I haven't been able to find a way to do this on the forum. Does anyone have any suggestions?

[Sam Chak]

Hi @mihir,

What is the order of your dynamic system? You can determine this by identifying the number of state variables. If, from the map of the state space in the operating range, you can identify which regions are badly conditioned, would you avoid selecting the initial values from those regions?

Answer 1

[Torsten]

Josh Meyer's answer shows you one possible way to do this:

https://uk.mathworks.com/matlabcentral/answers/511547-possibility-to-add-a-timeror-on-ode15s-which-after-a-certain-time-limit-the-solver-is-stopped-or-gi

I suggest you also define the stop time in your script and pass it to the event function via the call to the ODE integrator instead of setting it in the event function itself.

Answer 2

[John D'Errico]

Simpler is to just set the maximum number of evals or iterations. This way you don't need to do anything special, just set a flag that already exists. It will do effectively the same thing as what you asked for.

[Torsten]

I don't see such an option in "odeset".