The utility solar industry has a dark secret, well at least a shady one, but the word is getting out: A significant portion of solar power plants are underperforming. The recently issued “Solar Generation Report 2021” from kWh Analytics (one of the industry’s most comprehensive energy validation studies taken from asset owner data) revealed the following: “More than 30% of non-residential systems in the U.S. compared actual production against financed P50 estimates (target production) from 2011–2020 and found that systems on average underperformed by 5-13% in any given year, even after adjusting for weather.”
Although these statistics are troubling, they don’t necessarily mean that solar power plants aren’t working, but that the gap between expected and measured performance has widened. The problem is related to how those P50 estimates are determined. The models are not as accurate as they should be, thus giving asset owners a less-than-realistic expectation of what the actual plant performance should be. Thankfully, there are ways to close the “expected versus measured” gap—and to mitigate some of the underlying reasons for the underperformance.
One pathway to better align modeled expectations to real in-field performance is to address the issue of terrain shade losses. The root causes of utility-solar shade loss can be divided into several buckets:
• Undulating terrain
• Construction tolerances resulting in nonplanar tracker pier heights among rows
• Near shading from trees, nearby mountains, and other features at the site boundary
Like any uncomfortable truth, one must first admit there is a problem. Do utility solar plants experience shading, and does it impact energy yields? Yes, nearly all solar plants have shade losses. These shade issues can end up resulting in several percentage points of annual energy yield loss—and significantly undercut the expected financial performance of the asset.
Standard modeling tools have not traditionally quantified these shade issues well (if at all) in yield estimations. Shade loss is often ignored or categorically derated, there’s a misguided perception of so-called “flat sites” that may have a variety of localized slopes, and the usual modeling methods are tedious and have a long way to go to properly model terrain shade loss.
The good news is, better shade modeling is possible with the latest industry-accessible tools, and the modeled versus measured gap can be narrowed. It’s possible to make informed decisions on shade mitigation technologies by having an accurate understanding of the shade loss and the “real” baseline.
Nextracker has been working closely with our customers and partners on the “elephant in the room” of terrain loss, by dialing in realistic bankable models and then fine-tuning the model to account for the effectiveness of our TrueCapture™ energy yield enhancement and control software platform.
Our recommended practice for shade modeling is to combine AutoCAD and PVsyst—which now supports file formats exported from CAD-based tools to model performance impacts of the terrain for single-axis trackers—which then provides a more accurate and detailed picture of the shade loss. In Nextracker’s utility-scale field testing, this modeling practice is tracking close to the actual production loss numbers.
Nextracker is working with owners, developers, and independent engineering firms to standardize the terrain shade modeling process to create accurate and bankable production expectations. Using site topographic information in PVsyst for the purpose of estimating shade losses before construction should be a key step in accurately setting performance expectations for a plant from the very beginning.
While shade (loss) happens on sites with traditional backtracking, energy yield enhancement and control software platforms like TrueCapture paired with an independent-row tracker architecture can mitigate those losses. With a TrueCapture–enhanced system, construction tolerances are taken into account by utilizing our proprietary deployment methods that accurately determine the as-built height of each row. These as-built heights are then used as the primary input for deploying the TrueCapture row-to-row algorithm that determines optimal tracking angles. From a modeling standpoint, we can create contour layers based on the as-built heights to create shade scenes for PVsyst.
There is a growing body of knowledge that’s been drawn on in this brief overview of plant underperformance and terrain shade loss. A deeper understanding can be attained by taking advantage of these resources.
Take a look at what has been shared and discussed, such as our findings during a recent PV Tech webinar and contributed article, plus a follow-up blog post answering many of the questions posed during the webinar. Other resources addressing the topic of plant underperformance and specifically tracker terrain challenges include a Solar Builder article from Black & Veatch published earlier this year, DNV’s SolarFarmer PV plant design software for modeling complex terrain (which considers terrain modeling a standard best practice), kWh Analytics’ above-mentioned study, and our own Defne Gun’s blog post on our Split Boost mode on TrueCapture that optimizes split-cell, large-format PV module energy production. DNV also came out with two IEEE PVSC papers in 2020 and 2021 on tracker terrain loss.
If you want to learn more about how to mitigate tracker terrain shade loss and optimize the performance of your utility solar plant using TrueCapture, contact email@example.com.
Aron Dobos, Director Software Product Management, Nextracker</p.
Aron has over 13 years of experience in the solar industry. He joined Nextracker in May 2021, to help further build out Nextracker’s market-leading digital tracker technologies, including TrueCapture and NX Navigator. He holds a BS in engineering from Swarthmore College, an MS in electrical engineering from University of Washington, and a PhD in
mechanical engineering from Colorado State University.