Research on industrial carbon emission prediction method based on CNN–LSTM under dual carbon goals
Xuwei Xia, Dongge Zhu, Jiangbo Sha, Rui Ma, Wenni Kang
Abstract
Abstract In order to achieve the dual carbon goal, a prediction method of industrial carbon emissions based on CNN–LSTM was studied. The extended Kaya identity is used to measure the emissions, and the LMDI decomposition method is used to determine the influencing factors. The model inputs historical emission data, extracts spatial features through CNN, and then makes time series prediction by LSTM, and finally outputs the prediction results. Experiments show that this method can effectively predict carbon emissions in different scenarios and provide support for the goal of double carbon.
Topics & Concepts
Dual (grammatical number)Carbon fibersComputer scienceArtificial intelligenceReinforced carbon–carbonPattern recognition (psychology)AlgorithmLinguisticsPhilosophyComposite numberAir Quality Monitoring and ForecastingVehicle emissions and performanceEnvironmental Impact and Sustainability