Predicting solar power output using complex fuzzy logic

Photovoltaic (PV) power is one of the most promising renewable energy sources. However, it is also intermittent, and thus short-term forecasts of PV power generation are needed to integrate PV power into the electricity grid. This article compares two existing machine-learning approaches for forecasting (ANFIS and radial basis function networks) against a new approach based on complex fuzzy logic (ANCFIS). The proposed approach was more accurate in predicting power output one minute in advance on a simulated solar cell.