Excessive production from storage?

[jammyc]

Hi,

I am comparing a grid connected system feeding an existing load with zero export to one with an additional small storage, and am getting an unrealistic increase in ESolar in the latter.

For the solar-only option, the scheme is defined with a self-consumption defined from a CSV and scaled. A 1kW grid limit is applied (for some reason this cannot be set to zero), specified in active power (the system runs at 0.95PF) at the injection point.

For the storage option, the strategy is set to self-consumption, with the same scaled CSV. Allow to export to the grid is unchecked. I have to deselect the grid limit in Energy Management for the simulation to work.

All other parameters are the same.

When I run them, the annual ESolar with storage seems to be far too high given the storage capacity relative to the array. Looking at the hourly figures, for some reason the array stops generating with losses shunting to IL_PMax at certain times (even though it's below the inverter rated capacity) for the non-BESS but in the storage scenario it continues to generate.

Any ideas what's going on?

[jammyc]

Just to add: The systems do have a heavy DC/AC ratio, but not ridiculously so, and the IL_PMax is less than the inverter rating (even allowing for power factor).

[André Mermoud]

There are indeed some simulation problems with a very low grid limitation threshold, especially when you have several sub-arrays.

For simulating a system in self-consumption without injection into the grid, you should use the "Stand-alone" option.  Or for a big or complex system, you can simply use the normal Grid-connected system, and consider the E_Grid as an unused energy rahter than an enegy injected into the grid. This is quite equivalent.

Indeed, the grid limitation is not compatible with self-consumptions systems with storage.

[jammyc]

Hi Andre, thanks for getting back.

Quote

There are indeed some simulation problems with a very low grid limitation threshold [...] grid limitation is not compatible with self-consumptions systems with storage

Do you mean there are general problems with zero-export sites? That's a major limitation on a common design requirement... Can you expand on this / is it documented anywhere? There are several options in the application that appear to specifically provide for this

Quote

especially when you have several sub-arrays

Just one homogenous sub-array in this case.

Quote

For simulating a system in self-consumption without injection into the grid, you should use the "Stand-alone" option. Or for a big or complex system, you can simply use the normal Grid-connected system, and consider the E_Grid as an unused energy rahter than an enegy injected into the grid

Understood, but... for the avoidance of further confusion, in either case which output parameter should I use to determine the solar energy supplied to the user, including any storage? E_Solar? Is this accurate or does it affect inverter operation during curtailment?

[jammyc]

Hi Andre,

Thanks for the advice earlier in the week; eventually got things working.

1) Please can you expand on the issue with "zero-export" (grid export limit = 1kVA) sites. Do you know why it occurs, and what the trigger points in the design are?

2) I have noticed that, for grid connected PV with storage in self-consumption mode, the parasitic auxiliary losses (e.g. night losses) are recorded in different places to simple grid-connected sites. This means that E_grid and E_Solar are not directly comparable.