PVWatts with net metering, almost no PV and batteries and flat rate, yields LCOE < 0.

[frohro]

Hi, I'm puzzled by the SAM version 2018.11.11's output when I do a PVWatts net metering simulation with batteries and almost no PV. I get negative LCOE, which seems strange, though my understanding of LCOE may be incorrect. I'm enclosing a .sam file where I made the rate flat, zeroed incentives, and interest/inflation to simplify things. In order to upload the .sam file, I'm putting a .log as the extension. You will need to change it back to .sam. negative_LCOE.log Is this a bug, or is my understanding incorrect? Thanks, Rob

[GridTechs]

There is a selection I believe asking, Are the batteries going to be charged off the grid.

On Wed, Jan 29, 2020 at 7:36 PM Rob Frohne notifications@github.com wrote:

Hi, I'm puzzled by the SAM version 2018.11.11's output when I do a PVWatts net metering simulation with batteries and almost no PV. I get negative LCOE, which seems strange, though my understanding of LCOE may be incorrect. I'm enclosing a .sam file where I made the rate flat, zeroed incentives, and interest/inflation to simplify things. In order to upload the .sam file, I'm putting a .log as the extension. You will need to change it back to .sam. negative_LCOE.log https://github.com/NREL/SAM/files/4132063/negative_LCOE.log Is this a bug, or is my understanding incorrect? Thanks, Rob

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/NREL/SAM/issues/239?email_source=notifications&email_token=AMOWX3JICVOFSKGTEKGV6CTRAJDNNA5CNFSM4KNPKJXKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IJW5KLA, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMOWX3L623PYYJJYR47XYNTRAJDNNANCNFSM4KNPKJXA .

[frohro]

This is the simple PVWatts model. I can't find anything like that. Any pointers?

Regardless of where the batteries are charged, they should not improve the power bill. They are lossy, and the best you can do with almost zero PV generation is to disconnect them, so they don't waste energy, when the price of energy is the same at every time.

Thanks for the response!

Rob

[frohro]

Thanks to GridTechs, I found in the detailed PV simulation there is a section where you can check where the batteries change from. I set up a simulation (shared here as .log, which needs to be changed to .sam to run) in the detailed PV model similar to what I did here in the PVWatts model. When I have only charge the batteries from the PV system checked it acts as I expect, but when I check charge from only the grid, I get negative LCOE, and furthermore, it reduces my electric bill, even though the rates are constant as a function of time. This should not be, because batteries are lossy, and that means that if I charge from the grid, and then give it back to the grid at the same rate, what I give back is less than what I used to charge the batteries, and it costs me money. This simulation shows negative energy production, which is expected if you charge and discharge batteries from/to the grid, but it shows the electric bill going down with this negative production (claiming I save money by this).
surprising_results.log This may be related to a similar bug in the 2020.1.14 beta. (number 240).

[frohro]

Another funny thing with the simulation I uploaded above is that the DC module modeled loss is negative.

[frohro]

I think the problem is the AC connected battery loss. It is -305.82%, which makes it a generator.

[frohro]

I should have said that I changed it so it charges from the system instead of the grid to git the -305.82% for the AC connected battery loss. Maybe that is because the PV array is charging the battery, and that is calculated as a negative loss. If I swap it back to charging only from the grid, the AC connected battery loss is 1731.22%. I can't find a definition for AC connected battery loss.

[GridTechs]

Hi I am Brent. I haven’t started at the beginning of your discussion please Excuse me for chiming In. Not sure how I was included on this thread. what are you trying to accomplish, maybe I can help? If you pay a higher rate at night at night during peak hours. The batteries will allow you to store energy during the day when costing electricity is lower. This enables you to take advantage of peak shaving. The powered you stored during the day you can use at night for a lower cost and avoid high peaks demand charges.

Hope this helps

Brent

On Sun, Feb 2, 2020 at 8:36 PM Rob Frohne notifications@github.com wrote:

I think the problem is the AC connected battery loss. It is -305.82%, which makes it a generator.

— You are receiving this because you commented.

Reply to this email directly, view it on GitHub https://github.com/NREL/SAM/issues/239?email_source=notifications&email_token=AMOWX3PEP4APAV32BYJHLS3RA6NMTA5CNFSM4KNPKJXKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEKSOWIY#issuecomment-581233443, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMOWX3JASLI3MFQ7BBN2YT3RA6NMTANCNFSM4KNPKJXA .

[GridTechs]

Also if you have current transformers In place to monitor energy consumption and solar production for your home. It is critical to have these in the correct position. It’s possible the monitoring is reading your solar production head home consumption.

On Sun, Feb 2, 2020 at 8:49 PM Rowan Built rowanbuilt@gmail.com wrote:

Hi I am Brent. I haven’t started at the beginning of your discussion please Excuse me for chiming In. Not sure how I was included on this thread. what are you trying to accomplish, maybe I can help? If you pay a higher rate at night at night during peak hours. The batteries will allow you to store energy during the day when costing electricity is lower. This enables you to take advantage of peak shaving. The powered you stored during the day you can use at night for a lower cost and avoid high peaks demand charges.

Hope this helps

Brent

On Sun, Feb 2, 2020 at 8:36 PM Rob Frohne notifications@github.com wrote:

I think the problem is the AC connected battery loss. It is -305.82%, which makes it a generator.

— You are receiving this because you commented.

Reply to this email directly, view it on GitHub https://github.com/NREL/SAM/issues/239?email_source=notifications&email_token=AMOWX3PEP4APAV32BYJHLS3RA6NMTA5CNFSM4KNPKJXKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEKSOWIY#issuecomment-581233443, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMOWX3JASLI3MFQ7BBN2YT3RA6NMTANCNFSM4KNPKJXA .

[frohro]

Thanks Brent, Yes, I understand if the rate is different at some time of the day, then batteries can save you some money, but I have the rate flat. I only have one rate for every time of the day. :-) I want to use SAM to model a system with batteries, and I want to believe the results. When I started getting what appear to me to be strange things, I started reporting them here, making examples that show the problem better than my actual simulations. Probably the last thing I wondered about above is ok. I'm just a little spooked, by what appears to be an error in the simulations I'm doing. I'm doing a feasibility study and I need to be able to trust these results. At this point I don't.

Thanks for the reply. I don't know how you ended up getting these issue reports, but I appreciate your response. I'm on Guam, and it is Sunday back in the States, though it is Monday here. So I was surprised to get a response. :-) Again, thanks!

Rob

[dguittet]

Thanks for your detailed and informative comments. I've been looking at the issues you've mentioned in the last few days, and directed my fixes to the closest appropriate issue.

The bug is that the Utility rate calculator is NOT pricing the energy purchased from the grid. At the moment, the output from the PV simulation that gives a time series of electricity purchased from the grid for the battery is not being added to the load which is how the utility rate calculator determines how much energy has been purchased (as load - gen given load > gen).

[dguittet]

Regarding the ac connected battery loss, when the battery operates by adding or reducing power in a single hour, it increases or decreases the system output. The ac connected battery loss is the (annual system output before battery - annual system output after battery).

In the case where the battery can charge from the grid, the second term will be higher, resulting in a negative loss.

The loss can also be very small and negative if the battery can't charge from the grid but only discharges throughout the year.

[frohro]

Thank you so so much Darice! It did indeed seem like charging the batteries might just be free. :-)