Photovoltaic Power Prediction Based on Improved Long and Short-Term Memory Neural Network Model

Photovoltaic power generation is affected by weather and geographical environment, which presents fluctuation and random multi-interference, so predicting the output power is crucial to optimize the grid-connected operation of photovoltaic power generation and reduce the influence of uncertainties. Based on the deep learning theory, this paper proposes an improved long short-term memory neural network photovoltaic power prediction method. Irradiance, temperature, pressure, humidity data and historical photovoltaic power are selected as the characteristic data of the model, which is used as the input of the neural network to predict the power value of photovoltaic power at the next moment, and introduce the RMSProp algorithm to obtain the optimal loss function value of the model, ensuring the best prediction results. The simulation results demonstrate that the model has good prediction accuracy.