Posted Tue, 17 Nov 2020 15:03:56 GMT by Natalia Gemza
Hello everyone,
I have a few issues revolving conducting simulations of SST and would appretiate any input from other users who maybe were confronted with similar problems.
I. Is there a way to conduct a long-term dynamic simualtion with alternating settling parameters of activated sludge - to mimic filamentous growth during winter months? Other then conducting a series of dynamic simulations and manually adjusting r_H, r_P or v0?
II. To simulate batch settling tests I wanted to declare a uniform initial concentration across all leyers of SST and then start dynamic simulation with no feed and no under- and overflow. And I failed in finding a way to impose a user-declared values for state variables. Is there a way to do that other then creating a new model?

With regards,
Posted Wed, 18 Nov 2020 07:43:39 GMT by Enrico Remigi WEST Product Owner
Hello Natalia,
what you are trying is of course perfectly possible in WEST, but it will require some (small) coding  ;)
[list type=decimal]
[li]you need to define a new CLASS in the library, in which r_H, r_P and v0 are defined as manipulated, interface variables[/li]
[li]you cannot set specific values for state variables because, by definition, state variables are computed. In particular, you cannot use a flow-through, constant volume model to mimic a batch model. Once again, you need to define a new CLASS in the library[/li]
Which library are you using? MSL or Modelica?
If you need help in writing these variants of base models, please let me know.
Posted Wed, 21 Apr 2021 14:10:31 GMT by Natalia Gemza
Hi Enrico,
model modifications I was mentioning in the first post were put on hold. I was working with MLS library but then at the end of a year decided to upgrade to WEST+ and work on Modelica from now on. Since there are some issues with SST model I want to work with in Modelica library, I decided to go back to MSL.
As I changed a bit research plans and I don't plan to perform long term simulations my first question is not valid anymore.
Coming back to the second question: I assume I have to declare a new class which would be a TSS concentration in all leyers at time O? Is that correct? If so, is this procedure similar in both MSL and Modelica (like placement of new parameter definition etc.).  Anyhow, I would definitely need help in creating those new models.

Additinally another question came along (similar in topic) - I have dynamic data for SST model validation (inflow and underflow - Q and TSS- and SBH). So I wanted to modify SST model so as an input I can have direct TSS values and not ASM fractions. I assume follwing code lines in SST models have to be changed:
in Modelica
[i]Help_X_In = [new defined input vector] [s]sum(Inflow[i] for i in zModWWTP.TComponentsTSS) * (if size(zModWWTP.TComponentsTSS,1) > 1  then F_TSS_COD else 1)[/s];
X_In = if (Q_In <= 0.0) then 0.0 else Help_X_In / [s]Q_In ;[/s][/i]
in MSL (base.clarifier file)
[i]state.Help_X_In = [new defined input vector][s](SUMOVER Comp_Index IN {IndexOfFirstTSSComponent .. IndexOfLastTSSComponent}:

    {state.X_In = IF (state.Q_In == 0)
          THEN 0
          ELSE state.Help_X_In [s]/ state.Q_In * parameters.F_TSS_COD;[/s][/i]
And of course similarly for X_Out.
Am I even close to the right solution?

Posted Tue, 27 Apr 2021 12:58:06 GMT by Enrico Remigi WEST Product Owner
Hello Natalia,
and apologies for the late reply.
Yes, it looks ok.
For consistency in nomenclature, I would change Help_X_Layer, into "X_Layer_init" or something similar - but that's of course a detail

You must be signed in to post in this forum.