PDA

View Full Version : Balancing Guidelines



B_L
April 18, 2017, 08:04 AM
Is there a step by step procedure that can be used to correct the balance when running a model? I am using H20Map for my particular model.

Patrick Moore
April 20, 2017, 01:16 PM
Brian,

We are a little unsure exactly what you are asking for. Could you clarify what you mean by "Balancing Guidelines"?

Were you asking about how to get a model run to converge or more on how to get your model and field results to better match (like calibration)?

Thanks for clarifying...

Patrick Moore

Patrick Moore
April 21, 2017, 02:20 PM
Brian,

for the sake of others on the forum I am going to post what we discussed in email:

It is difficult to give hard and fast guidelines but here are a few:



If your model is not converging try these options: (similar settings in H20MAP water or InfoWater)

Generally if a model is having difficulty converging under different conditions the best way to find out what is occurring is to look at the Output text report in the run manager. To get the most information in this report you will want to use the following settings:
(click if need larger image)
395


Once you update these settings rerun the model and then look at the report.

You want to look for


Warning messages – Investigate each occurrence as it will often explain what the issue is
Look for things repeating in the run report (Look for repeated ID’s things changing status like Open to closed, closed to active, active to open.) If you repeatedly see an ID come up in the run report the instability is often caused by that one or more elements.


These issues need to generally get resolved to best resolve instability issues.

Accuracy – The general acceptable range is 0.01 to 0.001 in most circumstances. Accuracy values higher than 0.01 can lead to inaccuracy in the run results.

Advanced run parameters – If a model is dealing with instability, the advanced run parameters can help.

Try opening the run manager -> Standard tab -> Simulation options
These settings often solve many instability issues:
(click if need larger image)
396

(click if need larger image)
397


Notes:


200 trials – If a model will not converge in 200 trials there is an issue that needs to be resolved
Unbalanced – Continue – By continuing you can get results to help troubleshoot
Extended Run – set to 50 this allows additional runs with certain conditions to find a solution
Advanced parameters

Relaxation Factor : 0.6 This is the most helpful if models are having difficulty converging. (1 solves faster but can have more difficulties with instability at times)
Damping Limit – Set to 10X the accuracy. If accuracy is 0.001 then damping Limit is 0.01.



If you are still having issues please upload the model to support@innovyze.com and we can look at it in detail.


Other things to look at


Tank group graph – look for tanks going 100% full or 0% full right before the crash

Pipes connected to tanks that reach 100% or 0% full will get closed and could create an impossible way to satisfy demands or in rare cases, no place to send flow added to model and can cause a crash. This could be due to zone interface being open or missing or bad controls. Never set off controls equal to the max level in a tank as it will never exceed this value and most controls are at “greater than level X” which if “X” is max level it would never reach.


Are valves fighting for control? Review settings to see if they are set to close.


Hope this helps.


second email:

One last thing

Convergence problems mean a certain timestep cannot reach an answer before max trials is reached. Any solution given may be “unstable” or unrealistic.

Calibration issues generally mean the model runs but that results are not accurate when compared to the real world and can be caused by a host of issues. The only way to know how “good” a model is would be to compare it against real data.

Steady state flow testing typical calibration criteria


Pressures within 5 psi of static measured pressure
Pressures within 5 psi dynamic (when fireflow occurs) measured pressure
Pressure drop static- dynamic is within 5 psi of field pressure drop

EPS typical calibration criteria


Pressures within 5 psi
Flows within 8-10%
Tank levels within 1 ft


Many calibration issues are due to connectivity errors in the model. This means something looks visually right but is not connected as the model understands things connected. Using node in close proximity and pipe split candidates and diameter discrepancy checks in the software are helpful to identify the most hydraulically significant issues.


Hope this also helps.


Patrick Moore