An improved neural ensemble framework for accurate PV output power forecast

Abstract

A significant role of renewable energy resource such as solar photovoltaic (PV) is substantially important for the smart grid. One of a major challenge for large scale integration of PV into the grid is intermittent and uncertain nature of solar PV. Therefore, developing a framework for accurate PV output power forecast is utmost important. In this research, a novel ensemble forecast framework is purposed. A novel feed forward neural network (FNN) ensemble based forecast framework is proposed and trained with particle swarm optimization (PSO). The wavelet transform (WT) technique is applied to handle the sharp spikes and fluctuations in historical PV output data. Correlated variables such as PV output power data, solar irradiance, temperature, humidity and wind speed are applied as inputs to forecast the PV output power precisely. The performance of proposed framework was analyzed for one day ahead load PV output power forecast of summer (S), autumn (A), winter (W) and spring (SP) days. The selected days from each season are a clear day (CD), partial cloudy day (PCD) and cloudy day (CLD). The proposed forecast framework provides higher forecast accuracy compared to persistence and backpropagation neural network (BPNN) model.