as published in PV Magazine International March 2019
A few years ago, nobody took bifacial seriously. It was just another attempt to improve module efficiency and returns from PV projects that little bit further. But as is so often the case with the PV sector, things can move more quickly than expected. With bifacial installations set for a big year in 2019, Ragna Schmidt-Haupt, partner at technical and commercial energy consultancy Everoze shines a light on the remaining challenges to making projects bankable.
In 2018, bifacial had already seized 10% of the new-build market, and according to some market researchers, bifacial is forecast to capture a whopping 40% of the global market within the next decade. Although this may sound like a relatively gradual trend, past experience has shown just how quickly the PV industry can prove new technologies bankable. How often has PV outperformed forecasts? Ask the IEA!
The key advantage for bifacial is that its manufacturing costs are similar to mono-facial modules. Although overall project costs are still higher, announcements from field tests around the world have reported energy gains of up to 20%, which can easily outweigh the additional costs. However, some investors still shy away from including bifacial technology in their portfolios. Why? Some demystifying of key technical concepts for bifacial is required, including; albedo, light and elevated temperature-induced degradation (LeTID), spectral content, and project design optimization.
What is albedo?
Bifacial modules can capture reflected irradiance that reaches their rear surface. Albedo is the percentage of sunlight reflected by the ground up to the modules. The albedo is affected by many factors, such as the nature of the soil, the position of the sun, or the height of the device. The higher the equipment is placed from the ground, the more irradiance is captured.
So far, albedo data are difficult to obtain and to refine. NASA’s Moderate Resolution Imaging Spectroradiometer (MODIS), for example, provides monthly albedo values from satellite imaging, that can be processed by standard PV simulation software. But the resolution is low (several square kilometres at best) and the actual albedo at the site may be quite different from the MODIS pixel value. Therefore, such methods only provide a very rough estimate of potential gains.
To achieve more accurate results, ground measurements are needed from an albedometer and other relevant sensors at the site for a period of 12 months. Shortening the period only makes sense in places where the albedo does not vary significantly, like in the Middle East for example, but is not appropriate in locations which experience heavy snowfall in winter. It’s also important to ensure that the ground where the measurement is taken is representative of longer-term conditions, also after construction. However, the typically short project time and budget pressures of PV mean that it can be challenging to adopt such a considered approach. Hence, improved bankability for bifacial will emerge from aggregated experience over years, rather than lengthy albedo studies for a single site.
Modeling a bifacial project?
There is to date no standard regarding PV simulation. The financial community has gotten comfortable with the PV industry aligning on modelling approaches, with only marginal deviations, without a proper standard. Modelling the reflected irradiance and its interaction with the back surface of a bifacial module is new, adding further complexity – particularly in combination with tracking systems. But the PV community is developing and validating new models to rise to this challenge. The bifiPV workshops of the PV Performance Modelling Collaborative and the U.S.-funded DoE initiative that supports R&D aimed at increasing the reliability of bifacial technology are both examples of efforts in this direction.
Many effects have to be taken into account when estimating energy yields using bifacial modules, including degradation, LeTID, the spectral content of the reflected irradiance, and snow shedding to name just a few. How these differ from conventional PV technology as well as their effect on standard monocrystalline modules are ongoing fields of research. PVSyst is already offering a bifacial tool as part of its modelling software, in an attempt to incorporate some of these topics. Validation of this and other forthcoming software is still a work in progress. Again, the key challenge here is to get as much field data as possible to the widest range of independent reviewers and to share their results with the broader community.
Choice of technology?
All leading manufacturers are developing glass-glass bifacial modules based on high efficiency cells, p-PERC being currently the most common. Costs for other technologies adapted to bifacial, such as PERT or hetero-junction, might be slightly higher, but would then also yield a higher bifacial factor. Some are promoting the use of transparent back-sheets, that have the benefit of reducing weight, but with potential stability shortcomings. Even bifacial applied to half-cut cells is already commercially available. The IEC plans to release a new standard for bifacial modules later in 2019. Until then, suppliers will be asked to show independent peak power test results from reputable labs.
Detailed design needs to take into account several factors which are specific to bifacial, both for rooftop and ground mounted systems. On flat roofs, a straightforward option is to install a white layer below the modules such as a waterproofing membrane. Increasing the albedo for a ground-mounted system is more challenging, due to additional costs associated with removal of dust or grass or for the maintenance of ground material, such as white gravel.
As shading on the back side of the modules from the racking system should be minimised, the equipment choice can no longer be solely based on costs – form and design must be considered too. Decisions on whether to install the modules at a higher elevation to capture more irradiation can have a negative impact on the cost of structures and cleaning.
In addition to the height, the optimal angle for fixed-tilt systems is likely to differ when installing projects with bifacial modules. Some developers have proposed even vertical installation of the modules, as this could additionally reduce snow losses in some regions. With trackers, back-tracking – a tracking strategy that aims to minimise panel-to-panel shading on the front side – may not be the optimal solution anymore. Optimal row spacing is also likely to differ from mono-facial systems.
Despite the increasing maturity of the PV industry, bifacial shows it hasn’t become commonplace just yet! The complexity might still feel overwhelming, but there are no real myths around bifacial that need busting. Most of the questions raised are only teething issues, readily solved with a little more experience. Additional development, design, or balance of system costs compared to mono-facial, are likely to be eroded or erased once a few years of experience have been accrued. If bifacial PV systems can increase output by 10% or more in a predictable way, wider adoption by investors is inevitable.
The key task for the industry and R&D institutes is a concerted effort to extract as much data from the field as possible, and to make results available to the PV community. Data are needed from a range of climatic zones, applying different cell technologies and design options.
The question with bifacial is not whether it will fly, but more how quickly it will take off and whether it will outperform the forecasts.