Solar Radiation Prediction Based on TCN‐N‐BEATS Sequence Modeling

Abstract

Solar radiation prediction research is a key area of interest in the realm of solar energy utilization and has garnered significant attention in recent times. In order to realize accurate prediction of solar radiation and make solar radiation prediction better serve photovoltaic (PV) power generation, this study proposes a solar radiation prediction method based on sequence model, which integrates two kinds of neural networks, namely, temporal convolutional network (TCN) and neural basis expansion analysis (N‐BEATS). First, the dataset is preprocessed using Pearson’s correlation coefficient, outlier detection, and normalized to obtain valid and relevant data; second, the features of TCN feature extraction and N‐BEATS flexible extension are integrated to predict the solar radiation; then, the model’s hyperparameters are fine‐tuned using the grid search algorithm to ensure precise predictions; and last, the correctness of the method is verified by comparing the error metrics and the running time. Empirical findings indicate that the TCN‐N‐BEATS sequence model has high prediction accuracy and short time overhead, and it has certain application value in solar radiation prediction, and the model could offer valuable insights for predicting solar radiation.