We use cookies to make your experience of our website better. You can delete or block cookies, but some parts of this website won't work without them. By using this website you accept our use of cookies.

prabhanjan

16/06/2015 07:36:46

I've run the model with a constant flow (unsteady). I've generated the CSV file from the ZZN for min, max and mean values. While comparing the results for stage, I noticed that min and max values are same (expected, as it is a constant flow simulation).

However, the mean value is differing. When I see full time series, stage at all times is the same value.
Just wanted to know how does the model calculate mean values while generating CSV? I've used .mmm file for generating CSV.

Has anyone experienced this problem or is it a bug?

Thanks

Konrad

Flood Modeller Support

16/06/2015 14:36:26

It calculates the mean over all *calculated* timesteps, whereas the zzn file will give you results at each output interval. It is conceivable that the model could have been settling down (especially if the initial conditions were not at steady-state) during the first few timesteps, and therefore skewing the mean (although I'd expect the min and max to be different if this were the case).

Were you using adaptive timestepping? I believe it includes all calculations (but only when converged), even if those are subsequently discarded in favour of a lower timestep ... but if the model converged this should not change the calculated value (and therefore the mean).

There was an issue (since fixed) with the mmm file in ISIS v3.7.0 using Double precision simulations. Could this be the case here?

Otherwise, please feel free to send in your model and we'll have a look under the bonnet/hood to see what's happening.

Konrad

Sign in to create or reply to forum posts.

Privacy & Terms

Copyright