SSA-CNN-LSTM-based Power System Node Carbon Emission Factor Prediction

Accurately predicting the carbon emission factors of power system nodes is of key significance in promoting the realization of low-carbon power systems. In this paper, a prediction model is proposed, which combines Sparrow Search Algorithm (SSA), Convolution Neural Network (CNN) and Long Short-term Memory Network (LSTM). The model firstly utilizes the powerful feature extraction capability of CNN to accurately capture the complex nonlinear relationship between carbon emission factors and their influencing factors; subsequently, the long-term dependence characteristics in the sequence data are effectively handled by LSTM, so as to realize the accurate prediction of the future trend of carbon emission factors. In this process, SSA is used to optimize the configuration of key parameters of the CNN-LSTM model to ensure the optimal model performance. Through the close collaboration of SSA, CNN and LSTM, this prediction model fully exploits and utilizes the hidden information in the data, which significantly improves the prediction accuracy. The experimental data fully verified the excellent performance of the model in the field of predicting the carbon emission factors of power system nodes, providing solid technical support and data guarantee for the low-carbon operation of the power system.