Carbon Emission Prediction Using CNN-LSTM

In order to achieve energy structure change and green lowcarbon transition as soon as possible, the schedule and roadmap of total carbon emission control need to be formulated as early as possible and continuously improved in the process of transition development. However, the ability to obtain prediction models with high accuracy from the temporal characteristics of historical data needs to be improved urgently. In this paper, a CNN-LSTM-based CO2 emission prediction model is constructed and empirically tested using Chinese data from 2009 to 2019 to calculate and predict the carbon emission intensity and per capita carbon emission in 2023 and 2033, and the validity and feasibility of the model is confirmed by comparing with the carbon emission intensity in 2010.