Study on power prediction of photovoltaic power generation based on NGO-LSTM

Forecasting photovoltaic (PV) power is crucial for addressing climate change and supporting the energy transition. Although PV generation produces no carbon emissions, its output varies due to environmental factors, complicating accurate predictions. PV power forecasts are divided in to ultra-short-term, short-term, and medium-to long-term categories, with this study focusing specifically on ultra-short-term predictions. The approaches used include physics-based methods, statistical techniques, artificial intelligence, and hybrid strategies. Particularly, combining deep learning with swarm intelligence algorithms can enhance accuracy. This paper presents the NGO-LSTM model to improve PV power forecasting and assist with power system scheduling.