Carbon Emission Prediction Model of Power Industry Based on CEEMD-SSA-ELM Method

When forecasting the carbon emissions of the power industry, due to the lack of analysis on the comprehensive effect of different influencing factors, there is a large deviation between the prediction results and the actual situation. Therefore, a research on the carbon emissions prediction model of the power industry based on CEEMD-SSA-ELM method is proposed. The logarithmic average Dixon index method is used to analyze the factors affecting carbon emissions in the thermal power industry. From the perspective of production and consumption, the composition of specific factors affecting carbon emissions is comprehensively analyzed. With the help of kaya identity, the first decomposition model of carbon emissions is constructed to achieve a comprehensive analysis of the factors affecting carbon emissions in the thermal power industry. In the stage of building the prediction model, CEEMD was used to decompose the original carbon emission influencing factor data, and SSA was used to comprehensively calculate the action intensity of each influencing factor of carbon emissions. Finally, ELM was used to calculate the comprehensive value of carbon emissions. In the test results, the prediction results of the design model for carbon emissions under the baseline scenario, low-carbon scenario and enhanced low-carbon scenario are highly consistent with the measured values, and there is no significant error.