Carbon dioxide emissions prediction framework based on deep learning, combination model and optimization

The prediction of carbon dioxide emissions can better understand the future trend and impact of carbon emissions, and provide scientific basis for the formulation of emission reduction policies and measures. In this study, the carbon emissions of China's industrial sector are predicted, and an ensemble learning model based on Grey Wolf Optimization algorithm (GWO) is proposed. Firstly, the original carbon dioxide emission sequence was decomposed into a series of sub-sequences by singular spectrum analysis, and then backpropagation neural network, recurrent neural network, long short-term memory and bidirectional long short-term memory models were used for prediction. Finally, the combined weights of the four models were optimized by GWO, and the prediction results of the four models were combined by the combined weights to obtain the prediction results. The experiment results show that the carbon dioxide emission prediction framework proposed in this paper has good forecasting performance and provides a good forecasting tool for the government.