Climate warming is a hot topic of common concern all over the world and it has had a significant impact on climate, oceans and human life. The increase in the concentration of carbon dioxide in the atmosphere has become a significant factor in climate warming. In recent years, the concentration of carbon dioxide in the atmosphere has been mostly anthropogenic emissions. Accurate forecasting of carbon dioxide emissions will effectively propose solutions to the problem of global warming and then improve the environment in which we live. In our work, first of all, we use the new information priority accumulation method to optimize the weight of the new information in the prediction. Then we use the numerical integration method to optimize the background value of the grey model to achieve more accurate forecast. Application case results show that our proposed model is superior to other grey models in predicting carbon dioxide emission in India and Bangladesh.

In order to achieve accurate prediction of new energy related data, a fractional grey support vector regression model based on nested cross-validation is proposed. In order to verify the superiority of the new model, China’s wind energy consumption data from 2001 to 2014 were selected, and a fractional grey prediction model, a support vector regression model and a fractional support vector regression combination model were established, and wind energy consumption in China was predicted from 2015 to 2018. Numerical experimental results show that the newly proposed combined prediction model has higher prediction accuracy.