Photovoltaic Power Prediction Based on Wavelet Analysis

The randomness and volatility of photovoltaic power have negative effect on its application, power prediction is the key to solve this problem, however, existing photovoltaic power prediction methods have the problem such as single model, insufficient parameters and large error. Based on 5G and Internet of things technology, real-time monitoring of photovoltaic equipment and weather conditions can be carried out, so as to provide data support for power prediction. According to the collected data, this paper proposes a hybrid photovoltaic power prediction model based on discrete wavelet transform, convolution neural network, Long Short-Term Memory and Numerical Weather Prediction (DWT-CNN-LSTM-NWP Model) to reduce the prediction error, then the effectiveness of the new model is verified by simulation.