Enrico Remigi WEST Product Owner

[quote author=Moses link=topic=2465.msg5763#msg5763 date=1433322597]
.. Y_Min >measured value (Y_M) then you allow bypass (u_on=1) ..
[/quote]
That's not exactly what I said/meant :)
In the [b]standard[/b] "OnOffBand" controller, the condition Y_M < Y_Min is 'associated with' the signal u_On.
The actual value that u_On is to take is entirely up to you, i.e. the type of control you've implemented.
Assuming you have:
[list]
[li]used a 2-way fraction splitter (as the by-pass valve)[/li]
[li]connected the "out_2" terminal of the splitter to the by-pass flow (and therefore "out_1" goes to your plant)[/li]
[/list]
then the signal of the controller ("u") will set the percentage of by-pass flow:
[list]
[li]u = 0 (no by-pass, all to the plant) for Y_M<Y_Min --> u_On = 0[/li]
[li]u = 1 (by-pass, no flow to the plant) for Y_M>Y_Max --> u_Off = 1[/li]
[/list]

However, as I mentioned earlier, [b]this may not be the best solution[/b]: because a) you probably want to cut-off flow rate in excess of a certain threshold (rather than divert the whole incoming flow); and b) you may not want a "band" (Y_Min - Y_Max) but rather that particular threshold.

So the "OnOffIdeal" model may be a better option (solves point b).
You still need to vary "u" in function of Y_M, e.g.:
u = Y_M - Y_Threshold

You can use (rather: start from) any fractionation model, to create you own variation: the ones provided with WEST in InputOutput\misc are only "suggestion" for the respective categories.
But one can [b]create alternative models[/b] (graphically in the Fractionation  page of the Influent Tool) and save them for further use.

All the models (and the categories) are illustrated in the Models Guide.
Please refer to it for differences ASM2dMod vs. ASMG2d.

Yes: if the incoming interface link is between the signal of the probe (m3/d) and "Y_M" of the controller, then "Y_M" as well as "Y_Max" and "Y_Min" are in m3/d.

How will it work? Assuming that you used the standard OnOff model (I guess), it will work as follows (also have a look at the [b]Models Guide[/b]; or [b]F1[/b] when the block is selected on the canvas):
[list]
[li]Y_M < Y_Min: u = u_On = 1[/li]
[li]Y_M > Y_Max: u = u_Off = 0[/li]
[li]Y_Min < Y < Y_Max: u does not change[/li]
[/list]

I don't believe this is the kind of behaviour you would like from a by-pass ...

That is perfectly possible.
But the default ASM2 fractionation model in WEST does not use those parameters: you will need to modify the model first, in order for the same parameters (e.g. i_P_BM) to be available both for the influent and ASU blocks.
I guess, you could add a variable ("V", algebraic state var) "organic P", computed on the basis of those fractions.

Then, yes, you can create a set of top-level parameters linked to the influent and to the ASU models.

I see: so if it only happens occasionally, but generally it remains positive, it is most likely related to the integrator.
You should be able to completely get rid of that (and related warning messages in the log - which, by the way, slow down the simulation) by adjusting the integrator settings.

Do you get a stable [b]negative concentration[/b]? or does it happen occasionally and for very limited time intervals, during the simulation?
In the latter case, you may want to check integrator and integrator settings.

If it's not a typo and you're actually using a [b]negative K_P value[/b] for a standard PI aeration control (DO->kLa), then of course you will get a negative aeration (!) - in fact, no aeration at all, because the output of the controller is limited to zero.
So, I guess your DO in the aerated tank will be effectively zero (a very small negative value e.g. 10E-09) provided the conversion model (its rate expressions) are protected against negative values of the substrates ... which is certainly the case, if you use one of the standard categories of WEST.

[quote author=E.Ramin link=topic=2477.msg5739#msg5739 date=1432888356]
The new ASM model can be used in a WEST project with no problem. Just that it cannot be observed through the Gujer Matrix if it is not made through the Gujer MAtrix.
Maybe this is a limitation of the software and can be fixed in the next release?
[/quote]
No, this is not actually true.
The [b]Gujer Matrix Editor reads the BlockLib.xml file[/b] (not the underlying .msl source files).
So if something is not displayed, it means that it is not properly stored in the BlockLib.xml.

Are you sure you have 1) created the new Instance, 2) associated it to the new Category, and 3) saved the BlockLib.xml ?
I have tested the situation you describe and the conversion model for my new test category is correctly displayed in the Gujer Matrix Editor.

If you would like me to have a look at your library, please send it to our tech support: mikebydhi@dhigroup.com

[quote author=E.Ramin link=topic=2477.msg5731#msg5731 date=1432809746]
My new ASM model has already an istance from the time I copied it from an existing ASM through the Gujer matrix (Step 2).
[/quote]
Do you mean "your block library (BlockLib.xml) has already an instance" that uses your new category "ASM_X" ?

Anyway, whether you have added the new category through the Model Editor or manually, the easiest way to [b]add a new instance[/b] to the block library is as I have explained in my previous reply (i.e. through the BlockLib Explorer pane) because it automatically adds a number of necessary settings.

Another way can be via a text editor, by:
[list]
[li]copying/pasting an existing instance as template[/li]
[li]rename it[/li]
[li]set the appropriate Category DisplayName and Value properties[/li]
[/list]

If I understood well, and you do have an instance that is to use your new category, then you can simply do the last step above.

[quote author=Moses link=topic=2480.msg5729#msg5729 date=1432808075]
Please note that I didn't have TSS influent flow data and there is hanging outgoing variable (X_TSS), I didn't know what to do with it.
[/quote]
That's indeed the problem. And I am actually surprised you could go through the influent tool (and generate the influent model successfully).

You may not have an actual measurement of the TSS in your influent, but, being [b]X_TSS[/b] a [b]component of ASM2[/b], you are to provide a value (in fact, a time series) for it in the influent file.
This is not different to any other components of the component vector (the green blocks in the fractionation layout).
In fact, this is precisely what the [b]fractionation model[/b] is all about: by using your measurements, combined with a set of fractionation parameters and perhaps realistic assumptions, one describes the influent according to the specific ASM- component vector.

I see you have measurements for both soluble- and particulate COD.
You could derive X_TSS from the measured particulate COD; or a the sum (weighted!) of the components that contribute to TSS, i.e. X_I, X_S, X_H, X_AUT, X_PAO, etc.

[quote author=E.Ramin link=topic=2477.msg5725#msg5725 date=1432738059]
.. However, I am not able to view the code modifications in the Gujer matrix ..
[/quote]
This is most likely because you have not 'regenerated' the [b]BlockLib.xml[/b]

The steps you describe are essentially correct, but let me first clarify one or two details.
Once you've modified the code (I assume you have done so through the Matrix Editor, right?) and before you check the library: have you saved the changes?
E.g. [i]by closing the Matrix Editor, you will be asked to save; since you had already created a new [b]category[/b] "ASM_X", you do not need to rename[/i].
Step 4 (library check) has no relevance, until the block library has actually 'acquired' your new category - which does not happen automatically in W2014.

So, after your step 3 (and after you saved the conversion model):
[list]
[li]add a new [b]instance[/b] to the block library (from the Block Library Explorer menu)[/li]
[li]the new instance does not need to be called "ASM_X", although this has been generally the case so far[/li]
[li]associate the new instance to your new category (cf. Category DisplayName and Value properties)[/li]
[li]make sure you also provide a PaletteLib for the new instance - otherwise you will not be able to use it to create layouts in WEST[/li]
[li]save the Block Library[/li]
[/list]

Now, you should be able to view the Gujer Matrix for your new category.