Results 1 to 4 of 4

Thread: Different Results of H20Map Water vs of Infoworks WS Pro

  1. #1

    Join Date
    Mar 2018
    Posts
    2

    Different Results of H20Map Water vs of Infoworks WS Pro

    Hi,

    I try to import the model from H2OMAP to Infowork WS Pro, but i have a huge different pressure. The different until 4m
    Pressure in Infowork is higher than H2OMAP..

    appreciate if you could give me a suggestion to solve the issue..

    Kind regards,
    Indra

  2. #2
    Forum Moderator

    Innovyze Employee



    Join Date
    May 2015
    Posts
    386
    Indra,

    Several things I would check for sure would be the following:

    1) Are you using InfoWorks Pro for EPANET or strait InfoWorks Pro? (How did you Import the model? or are you actually using InfoWater SA by chance)
    2) Did your H20MAP Water model have features that are not supported in EPANET that may not come across in an EPANET export:
    a) Variable Speed pumps (implicit VSP controls) (You would need to recreate these as the pumps may now be running as full speed pumps)
    b) Flow Control Valve with a Pattern - This may require a different way of modeling this within EPANET or InfoWorks WS Pro
    c) Float Valves and Vacuum Breaker valves are also not in EPANET


    If you are using a feature that does not import directly to EPANET you would need to potentially modify the operation of that element when you bring it into another software package.

    If you can verify what package you were using to import to and if your old model had any non-EPANET features within it it is likely the differences are related to things that did not export directly to EPANET and may require some modifications to how they are modeled to replicate the model behavior.

    We would also recommend you contact us at support@innovyze.com for more detailed assistance.

    Patrick Moore

  3. #3

    Join Date
    Mar 2018
    Posts
    2
    Hi Patrick,

    The way I am imported the model is, first from H2OMAP exported to EPANET and from EPANET imported to Infowork WS Pro.

    FYI, my model is gravity system (not using pump), the available control are pattern in the source and 4 PRV to control the pressure. 2 PRV with static pressure output and 2 using patterns.
    I have been analysed the issue why the result is different and the issue is in PRV with the patterns. The output from the PRV is not match with the patterns (FYI, I inserted manually the pattern in the Infoworks since is not imported directly from EPANET). in the figure below is the comparison result in the same location (node after PRV) from H2OMAP and Infoworks (H2OMAP result is similar with the PRV pattern)
    Capture.JPG



    Kind regards,
    Indra

  4. #4
    Forum Moderator

    Innovyze Employee



    Join Date
    May 2015
    Posts
    386
    Indra,

    I just ran a test and verified that control of a PRV with a pattern is also a non-EPANET exportable feature and would explain why there are differences in your EPANET initial export.. This means you would have to recreate the pressure control pattern within InfoWorks WS Pro to have this feature again, but this would certainly explain why there are such differences as the pattern controlling the PRV in H20MAP Water would not get exported to EPANET and would have been noted as such in the Message board when exporting the file to EPANET.

    One major key point here though is why is your PRV setting varying in the real valve? Is this a two PRV valve and the flow through the valve causes headloss such that it is causing the valve to fall below the setpoint for the valve? Or by chance is this a PRV/PSV combination valve and is the PSV causing the valve to restrict the flow and thus not meet the PRV setting ? It certainly is highly unusual to see a PRV setting drop or vary like this unless there is something else really going on in the real world. I would highly recommend looking into this as as the modeler you want to try to avoid "forcing" a behavior in a model that is not really the cause of the behavior.

    Coincidentally if a PRV is 2 inch in size (very small) and you have a model PRV that has no minor loss coefficient assigned to it, mathematically the valve in EPANET will allow very large flows through the valve while th real world valve would not. But most PRV valves should have a Minor Loss coef of around 5 assigned to them in the model to better reflect the actual headloss that occurs through the valve once the flow exceeds a high value and the velocity creates high headloss. If there were 115 ft of head available to push a flow through a valve, this is the maximum head available to push flow. If minor loss Headloss is = K * V^2/2G where G is gravity , V is velocity, and K is the ML Coeff, if you assign a ML coef to a valve when the velocity^2/2G term times the K exceeds the head available the flow will not be able to increase because that would induce more headloss that is available to drive the flow. In the real world this is what generally happens in a PRV. The 1st small PRV covers the lower end flows and the next valve covers the larger flows. The smaller valve in the US is usually set at 5 psi higher than the next valve. As flows increase the first valve at some point would no longer be able to maintain the pressure setpoint due to the headloss in the valve and the pressure would drop. Once the second valve pressure setpoint (usually 5 psi below the small valve setpoint)is reached the second valve begins to modulate the pressure.

    Is it possible your PRV pressure you are seeing here is simply a valve whose flow is exceeding what the small PRV is capable of doing, but the pressure remains above the second valve setpoint? Or as noted is this a PRV/PSV combo and the PSV is controlling during the periods the valve drops below 53 psi such that the PRV is not really in control, but the PSV is?

    It would be wise to take a closer look at the real world valve here to determine what it actually is doing as it is very unusual for a PRV to have variable settings like this.

    Either way, if you need to induce more headloss in the valve itself, try adding a minor loss coef to your InfoWorks PRV as needed to replicate the drop in pressure. This may solve your issues and get model results closer to what you are expecting. By the way pressures within plus or minus 5 psi is generally considered by many as sufficient for most water models and your pressures even without changes appear pretty close to this.

    Patrick Moore

    PS: Please note that the InfoWorks engine and the EPANET engine are also different, so that you may still get differences even after addressing things you may see as obvious potential issues. Since they are not identical in how they "solve" they may not produce identical results based on how the model is currently setup.


    .
    Quote Originally Posted by Indra View Post
    Hi Patrick,

    The way I am imported the model is, first from H2OMAP exported to EPANET and from EPANET imported to Infowork WS Pro.

    FYI, my model is gravity system (not using pump), the available control are pattern in the source and 4 PRV to control the pressure. 2 PRV with static pressure output and 2 using patterns.
    I have been analysed the issue why the result is different and the issue is in PRV with the patterns. The output from the PRV is not match with the patterns (FYI, I inserted manually the pattern in the Infoworks since is not imported directly from EPANET). in the figure below is the comparison result in the same location (node after PRV) from H2OMAP and Infoworks (H2OMAP result is similar with the PRV pattern)
    Capture.JPG



    Kind regards,
    Indra

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •