Shade Modeling for Utility-Scale PV Plants: The PV Tech Webinar Q&A

Why do so many utility-scale PV plants chronically underperform? One of the main reasons is a lack of deep understanding of the true impacts of intra-array shading, especially on sites with sloped and uneven terrain. Hundreds of megawatt-hours in lost energy generation and tens of thousands of dollars in potential revenue are left on the table every day because shade losses are not well-quantified in the yield estimations, leading to a gap between expected and actual energy production. To minimize the impact of array shading, better modeling and mitigation technologies are needed.

Nextracker and DNV recently collaborated on a PV Tech webinar devoted to the issue of shade modeling. Director of Product Management, Software at Nextracker, Aron Dobos talked about the root causes of shade loss, the solar industry’s shade modeling state of play, and the extent of the shade loss problem. Neelesh Umachandran, Solar Performance Engineer at Nextracker, then discussed the evolving shade modeling best practices and offered examples of mitigation solutions such as Nextracker’s TrueCapture yield optimization software that can help recover much of the shade-induced energy losses. DNV’s MinWah Leung also presented an overview of the independent engineering firm’s approach to tracker terrain modeling, including its SolarFarmer energy modeling software that is incorporated into its independent energy assessments.

At the end of their presentations, there was a healthy Q&A discussion, but time ran out before many of the questions submitted could be answered. We have compiled some of the most interesting questions in this blog post, with answers provided by Aron and Neelesh.

How do terrain shade losses differ for bifacial PV modules, and does terrain modeling work in PVsyst for bifacial modules?

With respect to modeling in PVsyst, north-south slopes cannot be taken into consideration for monofacial or bifacial modules at this time. Also, with varying east-west slopes/row heights considered, PVsyst considers diffuse shading effects both on the front- and backside of a tracker system within the 3D field.

What is the step by step processes for estimating TrueCapture row-to-row and diffuse gains using tools like PVsyst?

We have published a step-by-step guide for modeling terrain shade loss in PVsyst for customers and IE partners. Please send us an email request: marketing@nextracker.com.

Some tracker companies mitigate terrain losses during the commissioning process by adjusting the tracker parameters. What is Nextracker’s view on this approach?

Manually adjusting tracker parameters during commissioning to avoid shade can mitigate shade losses in some specific cases, but usually results in reduced annual yield. Changing tracker parameters might reduce shade loss during one particular season, but the settings would need to be changed over the course of the year to account for seasonal differences. One set of parameter values for minimizing shade loss during summer commissioning might result in worse than baseline performance in winter.

Nextracker’s view is that ground-cover ratio (GCR) tuning is not a robust and easily verifiable approach that introduces a significant risk of reducing long-term performance. Ongoing manual configuration and verification work can be completely avoided and optimized by applying TrueCapture control and optimization software, with the addition of diffuse light optimized gains and specialized row-to-row algorithms that can be optimized for either full-cell or half-cell modules.

How bankable is TrueCapture? Can the yield boost for a particular site be built into the power purchase agreement (PPA), or does Nextracker offer a guarantee of the uplift?

All major independent engineering (IE) firms (DNV, Black & Veatch, Leidos, ICF, Luminate, RINA, and others) can provide TrueCapture gain estimation as a part of due diligence on energy models for developers. The site-specific gains estimated and quoted by Nextracker together with the (IE) analysis provide bankable yield boosts supported by in-field measurement and verification. This TrueCapture bankability translates into the customer being able to claim ITC against the purchase.

How do we approach shade loss modeling with north-south axis slopes, and trackers that can follow terrain features?

PVsyst supports only an average north-south axis slope. We suggest a median slope be assigned during modeling to represent north-south effects in aggregate. Nextracker Recently published work presented at EU PVSEC 2021 shows the effects of average north-south slopes combined with convex, concave, and rolling terrain geometries.

Performance modeling for terrain following trackers should consider the differences between convex, concave and rolling terrain. (Nextracker: EU PVSEC 2021)

Are construction tolerances accounted for typically by limiting backtracking angles? How can shade loss from construction variance be estimated in modeling?

From a TrueCapture product standpoint, construction tolerances are accounted for by utilizing our proprietary deployment methods. These as-built heights are then used as the primary input for deploying the TrueCapture custom row-to-row algorithm that determines optimal backtracking angles. From a modeling standpoint, we could create contour layers based on the as-built heights to create contour layers that can be used to create shade scenes for PVsyst.

To what extent is it more cost-effective to adapt to terrain (meaning accepting shade and mitigating it with TrueCapture) versus cut-and-fill grading to reduce shade losses due to terrain?

Optimization of site design with respect to terrain grading and cut and fill, terrain-following tracker systems, and shade loss and mitigation with software algorithms is a very complex problem and highly site/location-specific.

How do the terrain shade loss modeling results compare with actual measured field operational data?

We have seen a good correlation of PVsyst-based terrain shade loss results for split-cell modules with both our internal raytracing-based model for TrueCapture estimates that is field validated by long-term TrueCapture A/B type testing data and validation. We expect for full-cell modules PVsyst may tend to overpredict the losses due to its simplistic model for shade loss, which largely ignores nonlinear partial shading effects on arrays.

Which is better, the 1P or 2P tracker architecture, when considering shading losses and the ability to mitigate them on complex terrain?

It’s true that 2P systems, due to fewer piers per unit tracker surface area and shorter rows, have some advantages on constrained terrains and complex soils. But 1P systems, with longer but more flexible rows, have more potential to achieve significant terrain following in practice. The answer for 1P vs. 2P warrants a holistic view of all customer site-specific conditions and costs, and project – by geography specifications of a project.

Related Resources:

A PV Tech feature article – Tracker terrain loss: the elephant in the room and the low hanging fruit