In the quest for sustainable energy solutions, solar power has emerged as a frontrunner, but its intermittent nature poses significant challenges. A recent study published in the *Journal of Engineering Sciences* offers a promising approach to predict photovoltaic (PV) power generation more accurately, considering various uncertainties. The research, led by Ebenezer Narh Odonkor from the Department of Electrical Engineering at the Pan African University Institute for Basic Sciences, Technology and Innovation (PAUSTI) in Kenya, could have substantial implications for the energy sector.
The study focuses on predicting PV power generation using an Adaptive Neuro-Fuzzy Inference System (ANFIS), a hybrid model that combines neural networks and fuzzy logic. This approach is particularly valuable because it accounts for uncertainties such as solar irradiation, wind speed, and shading effects from trees, which can significantly impact solar power output.
“Solar power is intermittent, and this intermittency can lead to inefficiencies in power generation and distribution,” Odonkor explains. “By predicting PV power generation more accurately, we can better manage these uncertainties and optimize the use of solar energy.”
The research involved collecting weather data and modeling a 150kW PV system using the System Advisor Model (SAM) software. The data was then used to train the ANFIS model in MATLAB software. The results were impressive, showing that ANFIS, due to its adaptive nature, is highly suitable for predicting the performance of solar PV systems under different climate conditions.
Under no shading conditions, the annual AC energy harvested was around 56,020kWh, with an energy yield of 962kWh/kW and a DC capacity factor of 11%. When shading from trees was considered, the annual AC energy harvested decreased to 51,000kWh, with an energy yield of 930kWh/kW and a DC capacity factor of 9.5%.
The findings suggest that ANFIS could be a game-changer in the energy sector, particularly for solar power plants. By providing more accurate predictions, it can help energy companies optimize their operations, reduce costs, and improve the reliability of solar power as an energy source.
“This research is a significant step forward in the field of renewable energy,” says Odonkor. “It shows that by leveraging advanced technologies like ANFIS, we can overcome some of the key challenges associated with solar power and make it a more viable and reliable energy source.”
The study also highlights the importance of considering various uncertainties when predicting PV power generation. By doing so, energy companies can better plan and manage their operations, ultimately leading to a more efficient and sustainable energy sector.
As the world continues to shift towards renewable energy sources, research like this is crucial. It not only advances our understanding of solar power but also paves the way for more innovative solutions in the future. With further optimization techniques incorporated into the ANFIS prediction approach, the potential for even more accurate predictions and improved solar power generation is immense.