Abstract:Photovoltaic power generation is one of the emerging clean energy power generation methods. However, its efficiency in electricity generation is severely influenced by light intensity in the external environment and products unstable electricity to the power grid. So, it is very important to predict the trend of power generation through the collection and analysis of the external environment for stability of the power grid. To address this prediction problem, we propose a higher-precision dual-depth neural network optical power prediction method based on BPNN and LSTM-based discriminator, which are combined by genetic algorithms. We first introduce the principles of BPNN, LSTM and genetic algorithm, and then give the specific method of constructing a dual-depth neural network discriminative model. Finally, we perform a verification experiment. Experimental results show that the method has higher discrimination accuracy than the single neural network model.