How to handle convergence
I am using a simple 2D grid to simulate the growth of hydrates after some gas injection occurs in the system. Ideally I would like to shut down the source (injector) and let the system evolve with the injected gas. Nevertheless, when I set the injection rate to 0 in the `GENER` section, the simulation crashes.
I get the following error. What would be the best way to tackle this. Would I have to use the UPDATE section, or is there something to set in the PARAM section that can help me.
In <SS_Rate_Interpolation>, the time value 1.7356700E+11 is outside the range ( 0.0000000E+00, 1.7356700E+11) At <Tot_NumTimeSteps> = 145 the time exceeds the maximum value in the generation time table at elementA2Q09 (source GIn00) ==> will reduce timestep >>>>>>>> CONVERGENCE FAILURE on time step # 145 with Dt = 3.281375E+00 seconds, following two steps that converged on <NumNRIterations> = 1 ==> STOP EXECUTION AFTER NEXT TIME STEP
-
That would typically mean the simulation time has exceeded the maximum time defined in the table of injection rates. Or are you doing constant injection? Try commenting out the GENER block to disable injection if you're changing the rate via a restart.
-
Matthew Reagan Hi Matthew, thanks for your reply. I am using a table with varying rates over time. A simplified version I have added is below with a plateau of constant injection over 4000 years and decline from 4000 to 5500 years. This is just some preliminary testing for some work in relation to basin modelling.
GENER A2Q09GIn00 3 COM23 1.000e-04 0.00000e+00 1.26230e+11 1.73567e+11 1.00000e-04 1.00000e-04 0.00000e+00 0.00000e+00 0.00000e+00 0.00000e+00
Do you mean that if I want the simulation to continue after the source is shut I should include that time in the table?
-
