Accuracy of simulated data for bifacial systems with varying tilt angles and share of diffuse radiation
Graphical abstract
Introduction
Currently bifacial technology attracts considerable interest in the PV community and a constantly increasing share of bifacial capacity is also expected for the future (VDMA-PV, 2019). Due to technical progress, such as improved bifacial cell concepts or the availability of thin solar glass, the technology gets increasingly attractive (Libal and Kopecek, 2018) (Nussbaumer et al., 2018) (Osborne, 2017). Moreover, most of the advanced solar cell technologies, which are currently implemented in industrial production, allow a comparatively simple adaption to a bifacial layout (Romijn, 2017) (Veschetti, 2016). The interest in “peak shaving” and customized solutions for specific applications, further supports the development towards bifacial technology (Faturrochman et al., 2018) (Kreutzmann, 2017) (Guerrero-Lemus et al., 2016) (Soria et al., 2016) (Obara et al., 2014) (Lim et al., 2014) (João, 2013) (Nordmann et al., 2012).
The potential for an increased power output of bifacial modules was demonstrated by simulations and measurements on single modules or installations in various orientations (Libal and Kopecek, 2018) (Stein et al., 2017) (Appelbaum, 2016) (Ishikawa, 2016) (Reise and Schmid, 2015) (Yusufoglu et al., 2015) (Van Aken et al., 2014) (Guo et al., 2013) (Sugibuchi et al., 2013) (Kreinin et al., 2010). Nevertheless, potential investors in bifacial technology are often deterred by the uncertainty of the yield predictions (Kopecek and Libal, 2018) (Meydbray, 2018), which is caused by the considerably more complicated conditions compared to monofacial standard installations.
For freestanding bifacial modules, the optimum orientation is a trade-off between the front and rear side irradiance and the efficiency is dependent on factors such as the ground reflectance or the installation height. In real, extended systems, the arrangement of multiple modules has additional effects, such as direct shading by modules in adjacent rows or indirect shading of the modules surrounding, which results in a reduced effective albedo. Data of larger systems are rare and the results are linked to a specific mounting and/or the respective orientation. Accordingly, the generalization of field data from specific bifacial installations is difficult and the optimal installation conditions or module orientations are not known with sufficient accuracy.
The simulation of bifacial systems is more complicated as compared to monofacial installations. While the use of simulation tools is state-of-the-art and widely accepted to calculate the yield of projected monofacial standard systems, their adaption for bifacial systems is still ongoing and their reliability needs to be proven by comparison with measured data.
Several institutes and companies currently work on the development of suitable models, algorithms and software for bifacial applications (Chudinzow et al., 2019) (Chiodetti et al., 2018) (DiOrio and Deline, 2018) (Janssen et al., 2018) (Mermoud and Wittmer, 2018a) (Berrian et al., 2017) (Dassler, 2017) (Gali, 2017) (Hansen et al., 2017) (Kunath, 2017) (Castillo-Aguilella and Hauser, 2016) (Lindsay et al., 2016b) (Shoukry et al., 2016) (Solarworld, 2016) (Reise and Schmid, 2015) (Wang et al., 2015) (Yusufoglu et al., 2015).
These models use different approaches to simulate the amount of irradiance reaching the rear side of a bifacial module, such as view factors, ray tracing, and empirical modelling (Liang et al., 2019) (Pelaez et al., 2019) (Chiodetti et al., 2018) (Deline et al., 2017) (Marion et al., 2017) (Castillo-Aguilella and Hauser, 2016) (Hansen et al., 2016) (Lindsay et al., 2016a) (Van Aken, 2016). The different approaches and algorithms vary in complexity and the prediction accuracy may show a differing dependency on the relevant factors concerning the ambient and the installation conditions.
The presentation of data obtained on a test array for the systematic measurement of bifacial modules (Nussbaumer et al., 2019) (Baumann et al., 2017) (Klenk, 2017) sparked the interest of groups, which work on corresponding simulation tools. In this work, the prediction accuracy of these models is tested by a comparison of measured and simulated data. The focus of this work is not the comparison of long-term measurement data with simulation results, but the simulation of specific conditions. Days with different light intensity and share of diffuse radiation were chosen to determine the impact of the irradiation conditions on the simulation accuracy. Front side and rear side effects are respectively analysed and discussed. Due to the properties of the test rig, the simulation validation is respectively possible for a tilt angle range from 0° to 90°. Accordingly, it is possible to reveal irradiation and tilt angle related dependencies of the simulation tools.
Section snippets
Measurement set-up
The data aquisition is done on the BIFOROT (Bifacial Outdoor Rotor Tester), which enables a continuous tilt angle variation. The BIFOROT is located on the roof of the ZHAW (Zurich University of Applied Science) in Winterthur, Switzerland. It is a 3x3-module array for the systematic measurement of bifacial systems with varying mounting conditions (Klenk, 2018) (Klenk, 2017) (Baumann et al., 2017). This array is based on commercially available, 60-cell modules (Megacell, MBA-GG60-270) with a
Simulation tools
The measured data is compared to computations carried out with the commercial simulation software PVsyst (V6.8.1) and simulation tools developed at ISC Konstanz and ECN.TNO. These three tools have in common that they are able to model the energy yield of both monofacial and bifacial PV systems for a fixed tilt or if mounted on trackers. The simulation tools need as input the weather data (ambient temperature and wind speed) as well as irradiance data i.e. global (GHI) and diffuse (DHI)
General aspects
In this study, measured and simulated data are compared in order to determine the resulting accuracy and to reveal specific trends. Three days from October and November 2017 with differing insolation intensity and share of diffuse radiation were chosen (Fig. 4). While the 10/15/2017 represents almost clear sky conditions, the 11/08/2017 is a completely overcast day with negligible direct insolation. As an example for a day with mixed conditions, the 11/02/2017 was selected. The aim of this
Simulation results
The output of bifacial modules is combined from the front and rear side contribution. It is therefore of interest to consider the respective contributions if the accuracy of simulations is investigated. The BIFOROT enables such an analysis by its setup.
The results that will be discussed here are
- ▪
Calculated front side irradiance compared to data of a pyranometer on the module axis (M2). The front side irradiance is also compared to the Isc measured in M3.
- ▪
The rear side irradiance as determined by
Conclusions
The prediction accuracy of three simulation tools was tested with the BIFOROT test rig at varying irradiation conditions and tilt angles. The aim of this work is not a comparison of long-term measurement and simulation data, but the analysis of characteristic trends and dependencies at specific conditions.
The simulated front side irradiance is as good as the irradiance data enables. At days with predominantly diffuse light, the results are very sensitive to the small difference between the
Funding sources
The contribution of ISC Konstanz has been funded by the EC (Horizon 2020) and by the German BMWi (FKZ 0324088A) within the Solarera.net project ”Bifalo”. The funding within this project was given by the BMWi (contract nummer given) and ressources from the European Commission.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (64)
Bifacial photovoltaic panels field
Renew. Energy
(2016)- et al.
Simulating the energy yield of a bifacial photovoltaic power plant
Solar Energy
(2019) - et al.
Improvement and validation of a model for photovoltaic array performance
Solar Energy
(2006) - et al.
Maximizing annual yield of bifacial photovoltaic noise barriers
Solar Energy
(2018) - et al.
Bifacial solar photovoltaics – A technology review
Renew. Sustain. Energy Rev.
(2016) - et al.
Vertically mounted bifacial photovoltaic modules: A global analysis
Energy
(2013) - et al.
Aspects of bifacial cell efficiency
Energy Procedia
(2017) - et al.
Outdoor performance of bifacial modules by measurements and modelling
Energy Procedia
(2015) - et al.
Temperature effects of bifacial modules: hotter or cooler?
Solar Energy Mater. Solar Cells
(2018) - et al.
Design and evaluation of passive concentrator and reflector systems for bifacial solar panel on a highly cloudy region – A case study in Malaysia
Renew. Energy
(2014)