error - big generation on the first two hours of january the first

I don't know how to research this error.

I am working with northeastern Brazil. In the CSP technologies with storage (parabolic cylinder and solar tower), a large and strange generation appears, sometimes corresponding to the highest generation hours of the day, in the first two hours of January 1st.

Usually in the other months of the year the averages of the first two hours are negative, only in January are positive because of this record, only on the 1st. What could it be?

Thanks a lot, 

Denise.

Hi Denise,

Would you mind attaching a copy of your .sam file so I can investigate?

Best regards,
Paul.

Dear Paul,

this .SAM has three misteries for me, I already mentioned at the SAM FORUM:

1) a big generation at the first two hours of january the first (see PTCA_PETRO_COR);

2) an error when I try to simulate PTSA_RECIFE_COR with the altered version of the tmy file (the one is selected)

3) for some reason at the PTCA_PETRO_COR (it is happening in all cases of parabolic trough with storage, not happening at parabolic trough withouth storage), although I selected "uniform dispatch" at REVENUE window, in the last hours of the day, in some months, after sunset, the generation starts to reduce, then it rises around 20:00, 21:00, and then falls again. I cant explain why it rises after begining to fall...

Thans a lot for your cooperation.

Denise.

Hi Denise,

1. The PTCA_PETRO_COR case has storage, with the hot tank at 30% at the beginning of the simulation, so SAM dispatches energy from TES in the first couple of time steps until the TES is discharged. In general, you should be comfortable ignoring anomalies like this that only affect one or two hours of the 8760 hours in one year.

2. The simulation for the PTSA_RECIFE_CO using the corrected weather file for Recife was slow, but did complete for me.

3. As for the nighttime increase in output power, that seems to be triggered by the power cycle input temperature. When it drops below a certain point, the TES output increases until the power cycle input temperature is at the design point of 391 degrees C.

Best regards,
Paul.

Thanks for your cooperation,

regarding the second point, that the simulation for the PTSA_RECIFE_CO using the corrected weather file for Recife was slow, but you completed it, my configuration is: (is it enough?)



and relatively to the third point, shouldnt it happen to solar tower with storage, if it happens with parabolic trough? But it doesnt, and at night, in solar tower plants modeled in the same sites, it begins to use storage, the generation decrease without interruption  until it ends.

3. As for the nighttime increase in output power, that seems to be triggered by the power cycle input temperature. When it drops below a certain point, the TES output increases until the power cycle input temperature is at the design point of 391 degrees C.

Best regards,

Denise.

Hi Denise,

Your computer has enough resources to run SAM.

I may not have tested the correct combination of cases and weather files in my earlier tests. The PTSA_RECIFE_COR case with the "corrected" Recife weather file generates an error message about the piping model not converging. Please see the webinar recording "Updates to the Physical Trough Model in SAM 2018.11.11, Nov 2018" posted on the SAM CSP videos page at  sam.nrel.gov/concentrating-solar-power/csp-videos.html for a discussion of the pipe sizing algorithm that may help you troubleshoot this issue.

The physical trough and molten salt power tower models use the same TES control algorithm, so you should see similar results between the two.

Best regards,
Paul.