Short-Term Solar Photovoltaic Generation Forecasting by Ensemble Learning Method

The development of renewable energy has become a necessary measure for countries to reach the goal of environment sustainable development. As a result of this technical trend, solar photovoltaic (PV) generation has paid more attentions as an alternative energy resource. In spite of many advantages could be provided from PV, the unpredictable intermittent nature of the solar irradiance can significantly affect the use of PV energy. This directly impacts the operation of the electric power systems. To overcome this challenge, many different techniques recently have been applied for PV power forecasting. This paper therefore proposes an ensemble learning method based on Bidirectional Long Short-Term Memory (BILSTM) for short-term PV power forecasting method. Field measurements of PV power generations from a campus PV site are collected for modelling, testing, and validation. Results show that proposed forecasting method presents the well performance when compares to real measurements.