Bifacial Photovoltaic Yield Simulation as a Function of the Albedo

Bifacial photovoltaic (PV) technology represents a significant advancement in the field of solar energy. Unlike conventional monofacial PV modules, which only capture irradiance from the front, bifacial modules can also convert irradiance that reaches the rear side of the module, which is affected by the albedo, which refers to the reflectivity of the ground. Accurate and accessible simulation software is essential for spreading the technology and optimizing the design and performance of bifacial PV systems. The simulation of bifacial PV systems can be challenging due to the variety of models available, especially for the irradiance, albedo and electrical simulation, which can lead to a high variability in the simulation time and accuracy. Not all available models are free to use and have already been adapted from monofacial to bifacial PV simulation. One important factor to consider in PV simulation is albedo. This factor is not present in monofacial PV simulation and it can vary depending on factors such as light spectrum, ground type, and humidity. However, it is currently uncertain which simulation models are best suited for a comprehensive simulation of bifacial PV systems for different use cases, and there is no holistic open-source simulation tool available. This thesis presents the development, validation and publication of a holistic open-source simulation tool for bifacial PV (BifacialSimu), accompanied by further analysis of different simulation models. BifacialSimu makes the simulation of bifacial PV more accessible through a user-friendly graphical user interface with open source documentation and application examples. The tool includes a selection of simulation models for each step of the process and has been validated for various use cases, including agricultural PV and the impact of different albedo models and values. All simulation methods demonstrated a high level of accuracy. The selection between irradiance simulation on the front and rear side, using view factors or the ray-tracing model, has the highest computational impact. For accurate hourly power output trends, the ray-tracing simulation consistently outperforms other methods. Therefore, it should be prioritised for detailed temporal analysis. For calculating cumulative yield, either the pure view factor simulation or ray-tracing with the cumulative sky method is more appropriate. Furthermore, the importance of hourly albedo values, especially in areas with snowfall, has been demonstrated. To achieve a fast and accurate simulation using several weeks of data, an irradiance simulation is proposed using the view factor model for the front, the ray tracing model for the rear, a variable albedo model, a Typical Meteorological Year weather file and the simplified one-diode model for the electrical simulation. The simulation’s precision is enhanced by incorporating local weather data, such as albedo values. The simulation results of BifacialSimu are compared to those of the commercial simulation software PVSyst. BifacialSimu demonstrates higher accuracy than PVSyst in terms of energy yield prediction, as well as in terms of the coefficients of determination, with a value of 0.98 compared to 0.94. This is particularly evident on days with strong diffuse irradiance. In summary, this thesis provides insights into the advantages and disadvantages of various simulation models for bifacial PV simulation based on albedo. It also offers guidance on selecting the most accurate models for specific simulation purposes and describes the application of the open-source software BifacialSimu.