Results 1 to 8 of 8

Thread: Infowater multi-trace

  1. #1

    Join Date
    Jun 2017
    Posts
    5

    Infowater multi-trace

    I have the model of a service district with four pump stations and one tower. How can I use the trace tool to determine the travel of a contaminant that occurred at one pump station? The purpose is to build maps of influence for each facility so we could contain a contamination that could occur at any time of the day and for any duration.
    Thank you.
    Lelia

  2. #2
    Forum Moderator

    Innovyze Employee



    Innovyze Employee



    Join Date
    May 2015
    Posts
    419
    Leila,

    In InfoWater you can use both the source trace and multi-trace functions. These allow you to identify a "source" usually a reservoir, but can be a single junction or a tank. All water that passes through that element would be assigned a 100% from this source. Other junctions would show a percentage of that source they contain.

    You set this up in the simulation options under the Quality tab.
    Source Trace:
    (Click if need larger image)
    Single Trace setup.png
    Multi Trace:
    (Click if need larger image)
    Multi-trace-options.png

    The multi-trace report is found here:
    (Click if need larger image)
    Multi-trace-output table.png

    The multi-trace results show up in a separate report for all junctions, while the trace report shows up in the model explorer and output reports.

    As with any water quality analysis, make sure you run the model long enough that it reaches a repeatable steady state condition and you may wish to consider averaging the results of the final 24-48 hours after it stabilizes. You can easily do that by changing the report start to only start at the last 24-48 hours of the model run. The averaged output result statistics will only be for the period reported and you can use this to get a reasonable average.

    Please feel free to contact us at support@innovyze.com if you need further assistance as we can do a WebEx with you if it would assist you.

    Patrick Moore

  3. #3

    Join Date
    Jun 2017
    Posts
    5
    Thank Patrick.
    With water age and chlorine residual the results at hour 24 are copied as initial quality for all nodes in the system for the next 24 hour run. Repeat the process until no change is observed. Is this what I have to do with multi-trace? What time step do you recommend?
    Lelia

  4. #4
    Forum Moderator

    Innovyze Employee



    Innovyze Employee



    Join Date
    May 2015
    Posts
    419
    Leila,

    For most systems you would need to run much longer than a single day to reach a steady state repeating stabilized patter of any water quality parameter. While one certainly can take the end results and use them as initial conditions to do this, you would generally need to do it for a high number of runs unless the water quality results were essentially run every day and the previous results used, but very few people are modeling water quality that consistently to do so. But if you are comfortable with this process, you certainly can use it for a multi-trace.

    But if you are modeling a new parameter like a source trace, you would essentially be starting anew with a new parameter and would have to run your model for long enough that the water quality parameter evaluated reached a steady and repeating pattern in the run and would need to run the model until it reached a steady repeating pattern. Most people just run the model for an extended EPS run and view the WQ parameter withing the storage tanks as a good guide of how long to run the model for. Run times needed to reach a steady condition are very system specific but for larger systems can easily take 2-4 weeks of model run length or longer.

    Time steps used are also very model specific, but the most commonly used value is 1 hour for the hydraulic timestep and the quality timestep (and rule timestep) per EPANET recommendations should be adjusted to be 1/10th of your hydraulic timestep. If the actual water quality values predicted are desired to be of the highest accuracy, it is recommended to half your hydraulic timestep and (adjust Quality and rule time-steps accordingly) and re-run the model to steady state and compare the results to the previous values. If the results are not within your desired tolerance (in other words they do not match within a reasonable value) it is recommended to repeat the procedure and half the values again and repeat the comparison. Once the values match reasonably you can generally be fully confident in using the lager of the two values as your timestep values.

    This is a good quality control step to do periodically to verify your system water quality predictions are as accurate as possible. The increase in confidence in the results generally associated with better predictions of the system hydraulics which the water quality results are based upon. The downside to the lower hydraulic timestep is that it will increase the model runtime needed to run, but this is important if you are trying to get water quality results as accurate as possible.

    Since it is often easier to run a long EPS run that to copy 24 hour results over and over, you may wish to potentially adjust your procedure to more easily run the checks, but it is certainly up to you. You can also do a hybrid method where you run 2-4 weeks take the final results as initial as this may reduce the number of iterations you may need to do, but the process you use is certainly up to you.

    Please feel free to reply to this post if you need further assistance.

    Patrick Moore

  5. #5

    Join Date
    Jun 2017
    Posts
    5
    Patrick,
    Thank you for the information. I reached the repeatability after 191hr. Now I can import results in GIS and create maps of influence. This task is done.
    The second part gets confusing: determine how far into the system a contaminant spreads. At one of the 3 pump stations I created a second reservoir & controls to simulate a contamination that starts at 5am and stops at 7am. The main reservoir is closed at 5am and opened at 7am. I was surprised to see traces from second reservoir before 5am. I use multi-trace simulation. What am I doing wrong?

  6. #6
    Forum Moderator

    Innovyze Employee



    Innovyze Employee



    Join Date
    May 2015
    Posts
    419
    Leila,

    This is a type of question that is somewhat difficult to answer without seeing your model. As you are a customer with an Active InfoCare contract, I will email you directly and see if we can't set up a support call or WebEx to look more deeply at your question.

    Patrick Moore

  7. #7

    Join Date
    Jun 2017
    Posts
    5
    Thank you Patrick. We can do this tomorrow anytime is convenient to you.

  8. #8
    Forum Moderator

    Innovyze Employee



    Innovyze Employee



    Join Date
    May 2015
    Posts
    419
    Just to follow up for the benefit of the forum. Using the Protector Suite tool (if you have a suite license) or running a single trace is the easiest way to see a contamination event coming from a specific pump station (like an algal bloom). By color coding junctions and pipes with a similar map display we were able to watch the contamination event unfold over the EPS simulation and the user could modify system operation to see the impact of changing operation had on the contamination event.

    If any other user has a similar question feel free to contact us at technical support at support@innovyze.com as it is easiest to assist with these type of questions this way.

    Patrick Moore

Posting Permissions

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