Short-Term Power Generation Forecasting of a Photovoltaic Plant Based on PSO-BP and GA-BP Neural Networks

With the improvement in integration of solar power generation, photovoltaic (PV) forecasting plays a significant role ensuring operation security and stability grids. At present, widely used backpropagation (BP) improved BP neural network algorithm short-term output prediction PV stations own drawbacks neglection meteorological factors weather conditions inputs. Meanwhile, existing traditional model lacks variety numerical optimization algorithms, such that error is large. Therefore, based on plant Lijiang, considering related influence as irradiance, environmental temperature, atmospheric pressure, wind velocity, direction, historical generation data station, three algorithms (i.e., BP, GA-BP, PSO-BP) are utilized respectively this work to construct output. Simulation results show GA-BP PSO-BP models both obtain high accuracy, which indicates GA PSO methods can effectively reduce errors contrast original model. In particular, owns better applicability than GA, further