Medium and Long Term Carbon Emission Forecast Based on GM BP optimized by Beetle Antennae Search

Abstract

To better cope with global warming and fossil energy depletion, judge China's carbon emissions trends over the next decade. This paper proposes an improved BP neural network to predict carbon emissions from 2021 to 2030. Firstly, the GM(1, N) model is used to screen the factors affecting the total carbon emissions. Secondly, the weight threshold of the BP neural network is optimized by the Tennosaur algorithm, which avoids the shortcomings of slow local convergence of the BP neural network. Finally, the optimization model predicts China's total carbon emissions in the next decade. At the same time, by comparing the prediction results and error indexes of SVM and BP models, it is confirmed that the BAS-BP prediction model proposed in this paper is more accurate, which accurately reflects the trend of carbon emissions and provides reasonable suggestions for China's carbon emissions trends over the next decade. This paper proposes an improved BP neural network to predict carbon emissions from 2021 to 2030. Firstly, the GM(1, N) model is used to screen the factors affecting the total carbon emissions. Secondly, the weight threshold of the BP neural network is optimized by the Tennosaur algorithm, which avoids the shortcomings of slow local convergence of the BP neural network. Finally, the optimization model predicts China's total carbon emissions in the next decade. Meanwhile, by comparing the prediction results and error indexes of SVM and BP models, it is confirmed that the BAS-BP prediction model proposed in this paper is accurate and accurately reflected.