Convergence Errors be resolved automatically while the model is simulated
Sometime a TSR running for few weeks throws up convergence hence these sims need to be rerun with shorter timestep which is both inefficient and a lot of time is wasted. It would be good if somehow the simulations dialog allowed to run the model with Max and Min timestep so that when it encounters convergence there after the model runs with shorter timestep. (I understand the model does timestep halving but even then it fails to converge)
Salam Jamadar
shared this idea
Hello,
Thank you for this, there is already a mechanism available for achieving something very similar to this. If you have a network and then go to the simulation parameters for this network via Model > Model parameters > Simulation parameters then within these there is an entry for “Max number of timestep halvings”. Depending on the timestep in the run dialog and the entry in the simulation parameters here you will be able to determine the minimum computational timestep that the engine goes to before failing a simulation.
Hope this helps.