A Short-term Forecasting Method For Photovoltaic Power Generation Based On The Tcn-ecanet-gru Hybrid Model

This article discusses a new hybrid model, TCN-ECANet-GRU, for short-term PV power forecasting. The model combines temporal convolutional networks, efficient channel attention networks, and gated recurrent units to improve prediction accuracy. Experimental results show that the proposed method outperforms other baseline models in all four seasons of the year. Multistep predictions also demonstrate the effectiveness of the model.