A Machine Learning-Based Approach for PV Power Forecasting

The contribution of Solar Photovoltaic (PV) power in the energy sector has steadily grown over the past decades. However, the intermittent nature of PV power poses challenges to grid stability. Therefore, it is of significance for electrical utilities to obtain an accurate prediction of PV output power at a small time-scale so as to perform grid scheduling. In this paper, we propose a model based on machine learning to predict hourly-ahead PV output power. The model includes the effect of the sun's position, the tilt angle of the array, PV array aging effect and the cloud cover. Simulations are performed to ascertain the efficacy of the model and to verify the accuracy of the model. Using the proposed model, nth-hour ahead PV output power forecasting is also performed.