Industrial Carbon Emission Prediction Model Based on Artificial Intelligence

Industrial carbon emissions have a serious impact on China’s ecological security and climate stability, and the accuracy of traditional forecasting methods is low. This paper intends to use artificial intelligence methods to model carbon emissions in China’s industrial sector in order to improve its prediction accuracy and interpretability. During the establishment of the model, this paper collected data related to industrial production, such as output, energy consumption, consumption of raw materials, types of production equipment, etc., pre-processed these data, and used artificial intelligence technology to improve the prediction accuracy and interpretability. The determination coefficient R-Squared was introduced to explore the average value of actual carbon emissions. The results showed that the prediction accuracy of traditional statistical method was 0.75, and the running time was 18 s; the prediction accuracy of random forest was 0.82, and the running time was 30 s; the prediction accuracy of SVM was 0.85, and the running time was 45 s; the measurement accuracy of this model was 0.92, and the running time was 15 s. The research shows that the prediction model of industrial carbon emissions based on artificial intelligence is helpful to predict the impact of industrial carbon emissions on the environment.