Elsevier

Solar Energy

Volume 197, February 2020, Pages 6-21
Solar Energy

Accuracy of simulated data for bifacial systems with varying tilt angles and share of diffuse radiation

https://doi.org/10.1016/j.solener.2019.12.071Get rights and content

Highlights

  • Appraisal of the prediction accuracy of simulation tools for bifacial systems.

  • Comparison of three tools from PVsyst, ECN.TNO and ISC Konstanz.

  • Simulations compared to measurements obtained with test rig for bifacial modules.

  • Analysis of characteristic dependencies for varying irradiation and tilt angle.

  • Front side and rear side effects are respectively analysed and discussed.

Abstract

Potential investors in bifacial PV systems are still deterred by the uncertainty of energy yield predictions. Validation of the prediction methods and establishing an accuracy range is therefore a central prerequisite concerning the bankability of the bifacial technology.

Because of the sensitivity on multiple additional factors compared to standard monofacial installations, the yield simulation of a bifacial PV array is far more complicated. The available simulation tools for bifacial systems still have to prove their reliability, also for predicting the effect of varying ambient and installation conditions. Moreover, bifacial devices enable new types of applications, such as vertical installations.

In this work, we compare the results of simulation tools developed at ISC Konstanz and ECN.TNO, as well as the output of the commercial PVsyst simulation software, to measured data. The measured data is obtained by a test rig that carries out a continuous and automated variation of the tilt angle. The test rig is not a single stand-alone module but an array in order to include shading effects of real extended systems. In addition, the measurements focus on a central module, which represents the conditions of a typical device in an array. Days with differing light intensity and share of diffuse radiation were chosen to determine the impact of the insolation conditions on the simulation accuracy. General trends with regard to the sensitivity to tilt and insolation conditions are analyzed in order to evaluate the quality of state-of-the-art bifacial energy yield simulations.

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)

  • H. Nussbaumer et al.

    Energy yield prediction of a bifacial PV system with a miniaturized test array

    Solar Energy

    (2019)
  • S. Obara et al.

    Analysis of output power and capacity reduction in electrical storage facilities by peak shift control of PV system with bifacial modules

    Applied Energy

    (2014)
  • R. Perez et al.

    Modeling daylight availability and irradiance components from direct and global irradiance

    Solar Energy

    (1990)
  • R. Perez et al.

    A new simplified version of the perez diffuse irradiance model for tilted surfaces

    Solar Energy

    (1987)
  • R. Perez et al.

    All-weather model for sky luminance distribution—Preliminary configuration and validation

    Solar Energy

    (1993)
  • I. Shoukry et al.

    Modelling of bifacial gain for stand-alone and in-field installed bifacial PV Modules

    Energy Procedia

    (2016)
  • S. Wang et al.

    Bifacial photovoltaic systems energy yield modelling

    Energy Procedia

    (2015)
  • T. Baumann et al.

    Illumination homogeneity of bifacial systems – Outdoor measurements with systematically varied installation conditions

  • D. Berrian et al.

    Performance of bifacial PV arrays with fixed tilt and horizontal single-axis tracking: comparison of simulated and measured data

    IEEE J. Photovolt.

    (2019)
  • D. Berrian et al.

    MoBiDiG simulations and LCOE

    Presented at the 4th bifi PV workshop, Konstanz, Germany

    (2017)
  • Burgers, A.R., Janssen, G.J.M., Van Aken, B.B., 2018. BIGEYE: Accurate energy yield prediction of bifacial PV systems....
  • J.E. Castillo-Aguilella et al.

    Multi-variable bifacial photovoltaic module test results and best-fit annual bifacial energy yield model

    IEEE Access

    (2016)
  • M. Chiodetti et al.

    Predicting Yields of Bifacial PV Power Plants – What Accuracy Is Possible?

    35th European Photovoltaic Solar Energy Conference and Exhibition

    (2018)
  • Dassler, D., 2017. Bifacial gain simulations of modules and systems under desert conditions. Presented at the 4th bifi...
  • Deline, C., Marion, B., MacAlpine, S., Stein, J., Toor, F., Ayala, S., 2017. Bifacial PV Performance Models: Comparison...
  • DiOrio, N., Deline, C., 2018. Bifacial simulation in SAM. Presented at the 5th bifi PV workshop, Denver,...
  • Gali, R.S.R., 2017. Energy Yield Model for Bifacial PV Systems: A study and analysis of temperature & rear irradiance...
  • C.W. Hansen et al.

    A detailed performance model for bifacial PV modules

  • C.W. Hansen et al.

    Analysis of irradiance models for bifacial PV modules

  • Ishikawa, N., 2016. World First Large Scale1.25MW Bifacial PV Power Planton Snowy Area in Japan. Presented at the 3rd...
  • G.J.M. Janssen et al.

    How to Maximize the kWh/kWp Ratio: Simulations of Single-Axis Tracking in Bifacial Systems

  • G. João

    Testing bifacial PV cells in symmetric and asymmetric concentrating CPC collectors

    Engineering

    (2013)
  • Cited by (0)

    View full text