KPIs and Solar Metrics to Track in Utility-Scale Solar

KPIs and Solar Metrics to Track in Utility-Scale Solar

By: Adam Baker

Calculating solar metrics to improve solar performance.

Key performance indicators (KPIs) help quantify if your solar asset is on track to meet plant performance. Too many owners and operators don’t have a clear picture of their solar metrics to evaluate performance and improve issues. Part of the problem is the sheer amount of useless metrics out there.

I wanted to provide a very transparent, precise outline of what your critical and secondary solar KPIs should be. Oh, and I also want to point out a few of those BS KPIs you should stop tracking immediately.

 Solar Metrics

Critical KPIs

How do you ultimately measure success in utility-scale solar? I hope you’re ready for this very short list of solar metrics that I consider critical. Ready?

  • KWh

That’s it. The only thing you get paid for is energy, so that’s really the only thing that matters. Every other KPI in this article supports this one critical KPI.


Indicative KPIs

This set of solar metrics help you identify if there are any underperformance issues that would ultimately affect your #1 KPI: KWh.


Uptime/downtime per inverter

Inverters are designed to run all the time. They don’t need to be taken offline for periodic maintenance and lubrication, they just run. You can run them for a year without shutting them off. If an inverter stops, you’re not producing energy, therefore not making money.

Are you able to tell if your inverters are running well or not?

You can, by taking a look at uptime/downtime over time. What is the total number of generating hours vs. the total number of hours the inverter stops for reasons other than weather (e.g., maintenance, site tripped offline, blown fuse)? I would expect 4 hours of downtime per year per inverter. Anything higher, and you’ve got some root cause investigation to do.


Theoretical output (via irradiance & ambient temperature)

By looking at total sunlight received and temperature of the modules, you should be able to calculate the theoretical output of your plant. With a simple equation calculated using the module manufacturer’s standard testing conditions, irradiance variables, and temperature variables, you can analyze if your plant is in line with its theoretical output.

In practice, one met station is not perfectly representative of the whole site, but it should be representative of overall conditions. So, if you have watt-hours of irradiance, and the temperature correction factor from the module data sheet, you have theoretical energy.

There is one problem with using theoretical output as one of your solar metrics. It doesn’t take clipping into account. During clipping, your theoretical output will be higher than your equipment’s limit. At that point you’ll just have to normalize the excess DC calculated out of the total, as the inverter’s Pmax - the maximum you can produce.

Related: How to increase DC health and avoid early/late clipping

For some reason, very few owners track theoretical output. In fact, I only know of one. That’s likely because most system integrators integrating solar SCADA systems don’t fully understand solar energy production, and have no idea how to form this calculation in SCADA…or that its even desirable.


Clipping ratio per inverter

Your clipping ratio is calculated by taking the time of maximum power + the clipping time before and after maximum power (Tmp + Tc1 + Tc2), and correlating it with your DC/AC ratio.

Why should you track clipping ratio as one of your main KPIs? Say your site is designed with a 1.3 DC/AC ratio. If yesterday was clear skies and you calculated the clipping ratio on every inverter on your site, everything should be running at 1.3. If an inverter is running at 1.1, you’ve got an issue.

Like most of these KPIs, you can’t calculate clipping ratio in real time. You don’t have the data to calculate what the afternoon will look like when watching morning data. My best practice is to calculate this value for each day for each inverter, and find the previous month’s best day for each inverter. If it ran all day, any inverter running below its designed DC/AC ratio is under suspicion for DC health issues.

Tracking the daily, monthly, and yearly clipping ratio of a site is yet another rare solar metric. Hardly any system integrators understand how to do the math to find this KPI, and very few even know what to ask.


Average current per string

Instead of displaying combiner box current (which is one of the least helpful metrics to ever grace an HMI), every site should track average current per string. An apples to apples report of string current provides a comparison of all inputs to the inverter to pinpoint outliers.

This requires that for each combiner box, the current is divided by a number from the engineering drawings, so it’s a little more time consuming to find, but with the work comes the benefit. The strings with the lowest average show the likeliest location of DC health issues. 8.5 amps per string? Great, looks like these strings are working properly. Under 8 amps in similar conditions? Might be a problem with a MC4 connector, fuse, or cracked cell. This metric is tricky on cloudy days, but so are the rest of the metrics at a solar site.

This is yet another KPI that very few people track. Probably because this level of detail used to exclusively require string monitoring solutions installed on new construction solar sites. Luckily, temporary solutions are now available that allow non-invasive insight into string current.


Nonsense KPIs

There are some metrics in the solar industry that just won’t die. Why? I hold system integrators who don’t understand the solar industry ultimately responsible. Often in the world of SCADA HMI development, the people putting together SCADA don’t deeply understand the application. The mentality of, “This device gave me a value…it must be important, so I’ll pass it on” provides little value to the end user.

I see a long list of monitored points in every single solar SCADA that bug me. Here are a few of my least favorites...

  • Frequency: Every inverter and every meter at a site will report the same frequency value, and you can’t do a single thing to change frequency. Putting frequency as a monitored point on every screen is a waste, but it’s still a good thing to track. If your site relay protection trips your site offline, you can dig through frequency deviation data to identify if that was the cause. But it doesn’t need to be everywhere.
  • Combiner box current: As a value, combiner box current is useless. 150 amps tells me nothing about solar performance. However, if you normalized this data, you’d be left with average current per string. And that metric is extremely important (see Indicative KPIs).
  • Vab – Phase to phase voltages are less obvious metrics than line to neutral voltage.
  • Inverter inlet air temperature – Trust the (precise) met station for atmospheric conditions.
  • Any metric that totalizes for ‘today’ – Inverter kWh so far today is painful for me to see on a screen. There’s not enough context to know what to do with this information. Similar (but more annoying) is kVARh today.
  • Inverter bridge current – If inverter kW output is good, then life is good. These parameters are probably important if you are on the phone with inverter tech support people troubleshooting an issue, but they do nothing for my ability to monitor the site’s performance.

I could go on, but you get my point (I hope). KPIs are, by definition, performance indicators. They’re key to understanding the asset’s operating performance. Just because a parameter is available in the Modbus map of the installed device does not mean that parameter is important to performance. (Troubleshooting being a whole different can of worms…).


Use Solar Metrics to Create Useful KPI Performance Reports

Using each of these KPIs, you can create a very useful performance report that provides data you can use to investigate issues. After you have a report, I suggest walking through each KPI like so:

  1. Determine if actual output is approximately equal to theoretical output.
  2. If not, check the inverter clipping ratio. Over the course of the last month, are there underperforming inverters based on your DC/AC ratio?
  3. If yes, look at combiner box currents for that inverter(s). Investigate strings with the lowest average amps.
  4. If everything at the string level is approximately equal, investigate the AC side of the inverter. You might have a bad crimp connector, which can be found with infrared.

All things considered in the case of underperformance, the most likely scenario is DC health issues. At the string level, there are a lot of opportunities for bad connections.


Adam Baker - Affinity EnergyAdam Baker is Senior Sales Executive at Affinity Energy with responsibility for providing subject matter expertise in utility-scale solar plant controls, instrumentation, and data acquisition. With 23 years of experience in automation and control, Adam’s previous companies include Rockwell Automation (Allen-Bradley), First Solar, DEPCOM Power, and GE Fanuc Automation.

Adam was instrumental in the development and deployment of three of the largest PV solar power plants in the United States, including 550 MW Topaz Solar in California, 290 MW Agua Caliente Solar in Arizona, and 550 MW Desert Sunlight in the Mojave Desert.

After a 6-year stint in controls design and architecture for the PV solar market, Adam joined Affinity Energy in 2016 and returned to sales leadership, where he has spent most of his career. Adam has a B.S. in Electrical Engineering from the University of Massachusetts, and has been active in environmental and good food movements for several years.