relationship between dc_array_power and ac_inverter_power

I am new to the solar performance modeling and certainly lack the engineering knowledge behind all the conversions. I was playing around the PVWatts module and found something not intuitive in the output. The model set up is the standard PVWatts no financial model, using "USA TX Abilene (TMY2).csv" as the weather data, using all defaults value on the system design page, in which the Inverter Efficiency is 96%. In the output, I look at the hourly dc_array_power and ac_inverter_power, the first two days results are listed below for reference. it seems to me the hourly ratio between the AC and DC is not uniform. And it seems the ratio is much small when the sun just rises or is about to set. is it true that I shouldn't expect to see a uniform AC/DC ratio in line to the Inverter Efficiency in the PVWatts module? Can you explain what other factors affect the hourly AC/DC ratio?

Thanks for your help.


DC array power (W) AC inverter power (W) AC/DC ratio
1/1/2015 0:00 0 0
1/1/2015 1:00 0 0
1/1/2015 2:00 0 0
1/1/2015 3:00 0 0
1/1/2015 4:00 0 0
1/1/2015 5:00 0 0
1/1/2015 6:00 0 0
1/1/2015 7:00 37.3887 14.4476 38.64%
1/1/2015 8:00 460.69 429.238 93.17%
1/1/2015 9:00 1479.86 1421.65 96.07%
1/1/2015 10:00 1883.83 1812.57 96.22%
1/1/2015 11:00 2645.51 2545.85 96.23%
1/1/2015 12:00 2160.55 2079.55 96.25%
1/1/2015 13:00 2686.79 2585.45 96.23%
1/1/2015 14:00 2363.34 2274.77 96.25%
1/1/2015 15:00 1581.85 1520.48 96.12%
1/1/2015 16:00 985.946 941.81 95.52%
1/1/2015 17:00 207.482 181.304 87.38%
1/1/2015 18:00 0 0
1/1/2015 19:00 0 0
1/1/2015 20:00 0 0
1/1/2015 21:00 0 0
1/1/2015 22:00 0 0
1/1/2015 23:00 0 0
1/2/2015 0:00 0 0
1/2/2015 1:00 0 0
1/2/2015 2:00 0 0
1/2/2015 3:00 0 0
1/2/2015 4:00 0 0
1/2/2015 5:00 0 0
1/2/2015 6:00 0 0
1/2/2015 7:00 50.5411 27.3586 54.13%
1/2/2015 8:00 274.548 247.026 89.98%
1/2/2015 9:00 604.482 569.791 94.26%
1/2/2015 10:00 852.423 811.734 95.23%
1/2/2015 11:00 507.946 475.449 93.60%
1/2/2015 12:00 1002.5 957.926 95.55%
1/2/2015 13:00 900.274 858.367 95.35%
1/2/2015 14:00 661.711 625.682 94.56%
1/2/2015 15:00 703.534 666.51 94.74%
1/2/2015 16:00 412.599 382.191 92.63%
1/2/2015 17:00 60.1188 36.7595 61.14%
1/2/2015 18:00 0 0
1/2/2015 19:00 0 0
1/2/2015 20:00 0 0
1/2/2015 21:00 0 0
1/2/2015 22:00 0 0
1/2/2015 23:00 0 0

I corrected a few typos.

Hello,
The inverter efficiency input in PVWatts is a nominal value, which is the inverter's nominal AC power divided by the nominal DC power. That is what you would expect the inverter's DC-to-AC conversion efficiency to be when the output power is the same as the inverter's rated power. For example, if a manufacturer claims that the nominal conversion efficiency of a 5 kW inverter is 96%, you would expect the efficiency to by 96% when the output power is 5 kW.
There are a number of factors that affect the actual conversion efficiency of an inverter at any given moment. That means that the actual conversion efficiency is going to change throughout the day. Inverters are less efficient at lower power levels than at higher power levels, so on a clear day, a properly sized inverter will be most efficient in the middle of the day when the PV array output is at its highest. Inverters are also less efficient at higher temperatures than lower temperatures. Most inverter data sheets show an efficiency curve that shows the relationship between the conversion efficiency and the inverter's output power and temperature.
By "nominal," I mean the rated values that appear on the data sheet provided by the inverter's manufacturer. Because there is not a standard for the information that appears on the data sheets, different manufacturers provide different information, but they usually provide enough information to determine the invereter's nominal efficiency. If the nominal efficiency (or nominal AC and DC power) does not appear on the data sheet, you can use the efficiency curve to determine the nominal efficiency.
During simulations, PVWatts uses a quadratic equation to calculate the inverter efficiency. The equation was empirically derived from inverter performance data measured from thousands of inverters in the California Energy Commission database. For details, see Chapter 12 of the PVWatts Reference Manual, which you can download from the Performance Model Documentation page of the SAM website.
Best regards,

Paul.