More Inverters = Less Production (?)

I recently modeled two systems via the PVWatts/simple analysis option. The system inputs were alike in every respect except for the dc to ac ratio—essentially, one system was modeled with 8 inverters (ILR = 1.14) and the other with 9 inverters (ILR = 1.02). Please see attached simulation reports.

Notably, the annual production for the 9-inverter system is modeled to be (slightly) less than that for the 8-inverter system. Something about that seems a bit counter-intuitive to me; I feel like I may be missing something fundamental about the way inverters operate at relatively low dc to ac ratios... Can someone offer some insight here? or point to any explanatory documentation?

Thanks,
CS

Hi Corey,

In the attached .sam file, I set up a parametric study on DC/AC ratio for a PVWatts case similar to yours (but using the weather file for Des Moines that comes with SAM). To see it, click Parametrics under the Simulate button. The following plot of the results, which I exported from the Parametrics page to Excel to create the graph, shows that as you increase the DC/AC ratio from 1 to 1.5, the total annual output increases slightly up to a point and then decreases sharply.



PVWatts uses the efficiency curve below (Figure 4 from the PVWatts Technical Manual ) to model the inverter so that the inverter is slightly more efficient at higher load fractions.



As you increase DC/AC ratio for a given array capacity, the inverter size decreases, which causes the inverter to operate more often at a higher load fraction, causing a slight increase in the total annual output of the system. However, beyond a certain point, decreasing the size of the inverter causes more of the PV array power to be "shed" by the inverter causing the decrease in annual output.

Best regards,
Paul.