This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

WaterCAD age calculation fails

I have run into a recent problem.  I've been trying to do a age calculation on our water system and for some reason the largest duration I can calculate is 128 hrs.  If I change it to 129 hrs the calculation fails and I get an error:

Calculation Engine Failed.  Please see user notifications for possible causes.

To obtain a complete list of problems choose 'Analysis|Validate'.

The user notifications indicate:

41110    "Age-128 HR"    "Scenario"    "12518"    "Age-128 HR"    (N/A)    "Cannot solve network hydraulic equations."    Pressure Engine

Element ID: 12518 is my Age scenario. 

running Validate gives me no errors.

I'm confused as to why I can run my analysis for 128hr and nothing greater than that.

Thanks

  • I assume, you are referring to scenario name "EPS"?

    After running the model in 08.11.08.50 version, I got "Network Unbalanced" message so at this point, I cannot fully trust the results. Despite the Network Unbalanced, I looked at your tanks and I see all of them are fluctuating?

    Looks like I am not on the same page with you.



  • Yes the EPS scenario.  Please double check the following settings for that scenario:

    1. All alternatives set to the "base".

    2. Base-Operational alternative set to "Summer" controls.

    Take a look at the graph "Tank Level vs Pump Flow and make sure its set to display the EPS scenario.  

    When I take a look at that graph, T-2 sits empty for 12hrs before it starts filling again.

  • Before drawing any conclusion on an "Unbalanced Network", please try to work with the model and rectify the issue. Looks like you have fairly complex controls which needs some clean up for example see the highlight below.

    I would also try to break the controls in simpler form rather than trying to make it complex. Simpler controls are easy to understand and troubleshoot.



  • By increasing the number of trials to 200, I believe I was able to get rid of the unbalanced network.  This may not be the correct approach to solving this problem.

      I'm not sure how to simplify the controls much more and have the model operate in the same manner our system is set up.

    Between noon and midnight, if T-1 gets above 52% full(1053.4) or T-3 gets above 54% full (1053.09), Then Pump 7 is to turn off.   Between midnight and noon if T-1 gets above 94% full(1070.29) Pump 7 is to turn off.

    This is the case for nearly every pump in our system that serves a tank.

  • Yes the error message went away after 200 tries however, if you look at the same graph (Tank Level vs Pump Flow), you will see pump(s) are going crazy and I am guessing it's due to some controls.

    "Between noon and midnight, if T-1 gets above 52% full(1053.4) or T-3 gets above 54% full (1053.09), Then Pump 7 is to turn off.   Between midnight and noon if T-1 gets above 94% full(1070.29) Pump 7 is to turn off."

    You can break this up into two controls:
    if T-1 gets above 52% full(1053.4) between noon and midnight, then pump 7 status = off.
    if  T-3 gets above 54% full (1053.09) between noon and midnight, then pump 7 status = off.
    (This way it's little easier to see if you have any conflicting controls) 

    Out of curiosity, I ran the model with WaterGEMS 2.00.12 engine (with default settings) and got the graph like below where pumps are doing somewhat better but still the is some issue.

    With the settings like below in Calc Option, I got graph like below which seems better but your question on tank being empty remains there.
    Convergence Check Frequency = 5,
    Convergence Check Cut off = 10,
    Damping Limit = 0.01 

    Again, out of curiosity, I lowered your demand by 25% and this is what I got:

    Looking at the above graph, it tells me that either you have higher demand in your system that it can currently supply or you are not allowing your pumps to fill the tank.

    Hope this helps.