PV Solar Power Output Forecasting using Machine Learning Algorithms

Researchers from Universiti Tenaga Nasional have published a report on the use of machine learning algorithms for PV solar power output forecasting, highlighting the potential of the Artificial Neural Network (ANN) algorithm. The research, which was published in the Journal of Engineering Applications of Computational Fluid Mechanics, used a dataset from the National Renewable Energy Laboratory (NREL) to compare the performance of five algorithms: Artificial Neural Network (ANN), Decision Tree (DT), Extreme Gradient Boosting (XGB), Long Short-Term Memory (LSTM), and Random Forest (RF). The results showed that the ANN algorithm produced the best mean absolute error (MAE), root mean squared error (RMSE), and coefficient of determination (R2) with values of 0.4693, 0.8816, and 0.9988, respectively.

Key Takeaways:

• The Artificial Neural Network (ANN) algorithm is the most reliable and applicable for PV solar power output forecasting.
• The ANN algorithm produced the best mean absolute error (MAE), root mean squared error (RMSE), and coefficient of determination (R2) with values of 0.4693, 0.8816, and 0.9988, respectively.
• The research used a dataset from the National Renewable Energy Laboratory (NREL) to test the performance of the five algorithms.
• The study aims to address the research gap in PV solar power output forecasting by determining the best-performing algorithm.

Statistics:

• The ANN algorithm produced the best performance with a mean absolute error (MAE) of 0.4693.
• The ANN algorithm produced the best performance with a root mean squared error (RMSE) of 0.8816.
• The ANN algorithm produced the best performance with a coefficient of determination (R2) of 0.9988.

Sources:

• Engineering Applications of Computational Fluid Mechanics. (2022,16(1):2002-2034).
• doi: 10.1080/19942060.2022.2126528.
• Yusuf Essam, Department of Civil Engineering, College of Engineering, Universiti Tenaga Nasional, Selangor, Malaysia.