The impact of carbon emissions on climate change is increasingly prominent and caused great concern around the world, in order to help save energy and reduce emissions, it is necessary to conduct research on the factors affecting carbon emissions and the forecasting model, in this paper, using the GM (1,1) forecasting model, Selecting the carbon emission data of a city in Jilin Province from 2017 to 2021, using Python program code to make short-term prediction of carbon emissions of a city, and predicting the increase value of carbon emissions based on the grey prediction model, correlation analysis of the data to predict the increase value of carbon emissions. Through comparative analyses, it is suggested that efficient ways to reduce industrial carbon emissions in a city are to prevent excessive growth of industrial output, optimise energy structure and reduce energy consumption intensity.