tcsmolten_salt example failed in SDK

15 posts / 0 new
Last post
ozsolarwind
tcsmolten_salt example failed in SDK

Hi

I might be jumping the gun. I was attempting to run the tcsmolten_salt model with the SAM SDK. I receive an error:

exec fail(tcs_moltensalt): Connections and value settings haven't been programmed for this module yet! (MJW 2104.11.21 time -1

Any timeframe for when it might be available? Meantime I'll use another TCS model

Kind regards

AnGus

tcsmolten_salt example failed
ssc version = 41
ssc build info = Unix 64 bit GNU/C++ Feb 2 2015 04:18:49
Log Error: exec fail(tcs_moltensalt): Connections and value settings haven't been programmed for this module yet! (MJW 2104.11.21 time -1
Error [ 0 ]: exec fail(tcs_moltensalt): Connections and value settings haven't been programmed for this module yet! (MJW 2104.11.21

Paul Gilman

Dear AnGus,

The tcs_moltensalt module should be working correctly in SDK 2015-1-30. Would you mind emailing me a copy of the program or script that generated that message?

Thanks,
Paul.

ozsolarwind

Paul

On it's way

Kind regards
AnGus

ozsolarwind

Hi

Still encountering this problem with the 2015-6-30 version. I'm trying to "bootstrap" using this module so it may just be a problem with variable values. Any clues would help.

Kind regards

AnGus

Paul Gilman

Hi AnGus,

From my reading of the source code, it looks like this error message is generated when you set the value of receiver_type to a value other than zero. The value of that variable in your CSV file is set to one -- changing it to zero should fix the problem.

That variable is a placeholder for a cavity receiver, which is not currently implemented for the molten salt power tower SSC module -- it can only model an external receiver.

Sorry it took me so long to get back to you on this.

Best regards,
Paul.

ozsolarwind

Hi Paul

Great. I've just checked and yes receiver type = 1 was the culprit. Now I can progress with fabricating suitable defaults for other variables. No worries on the delay - I have plenty to do :-)

Kind regards

Angus

ozsolarwind

Paul

Next question with tcsmolten_salt. I now encounter the following error:

tcsmolten_salt error [ 2 ]: unit 3 () type 'sam_mw_pt_type222' failed at initialization

I notice a post from 2012 (re receiver convective losses) that refers to the source for sam_mw_pt_Type222.f90 being available in the sam installation folder but can't find it in my current versions (maybe the packaging has changed).

Kind regards

AnGus

ozsolarwind

Paul

I've obtained the source from a Windows version so will look at the code to see if I can work it out.

Kind regards

AnGus

Paul Gilman

Hi AnGus,

The TRNSYS source code files are available with SAM 2014.1.14 and older versions. Later versions no longer rely on the TRNSYS simulation engine, but do use essentially the same code (translated to C++) with a few modifications in some cases.

Best regards,
Paul.

ozsolarwind

Paul

I've progressed to where tcsmolten_salt is returning an array of (8760) 'nan' values for gen (hourly generation). I'll work through the variables I've set to see if I can solve the problem but any clues would help.

Kind regards
AnGus

Paul Gilman

Hi AnGus,

Would you mind emailing me a copy of the script that generates those errors? My colleague promises he won't wait as long as last time to look at it.

Thanks,
Paul.

ozsolarwind

Paul

On it's way

Kind regards

AnGus

Paul Gilman

Hi Angus,

We were able to replicate the NaN results from your Python script. I suspect that there is a problem one or more of the input variables in the script. Instead of trying to find those variables, we recreated a Python script from the LK script generated by the molten salt tower model in SAM that works with your weather file. I tested it in both the latest version of the SDK (2015-6-30 r1) and the previous version (2015-6-30) and it generated numbers for the hourly results instead of NaN. 

You can download our test files here: ZIP 108 KB

To run that script, extract the files from the zip archive and put them in the SDK \languages\python folder

Note that the results from SSC/Python are different from the SAM default molten salt tower model because the costs used by the field optimization routine are different. The Python script sets those values (tower_fixed_cost, rec_ref_cost, etc.) to default values that are internal to the tcs_moltensalt module, while SAM takes those values from the default values of the single_owner financial model.

Thanks,
Paul.

ozsolarwind

Paul

Great! Thank you.

I suspected a variables problem but not easy for me to "debug". I've run the Python script and it works fine. Now it will just be a simple case of fitting the variables back into our simulation (I might report back what the differences are FYI).

I'm not too concerned by the financials at this stage but being able to add molten salt tower to our simulation is great. Steve, our team leader is off on leave tomorrow so will be pleased this has been (nearly) sorted before he goes. Thanks for the quick turnaround.

Kind regards

AnGus

ozsolarwind

Paul

I'm embarrassed to say the incorrect variable was a pretty obvious error. design_eff (Power cycle efficiency at design) was set to 0. I'll use your variable values as defaults where I can as the results produced can be very different. Now on to biomass :-)

Thanks, and Kind Regards

AnGus

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer