Analysis And Forecast Of Carbon Emission In Electricity Coal Supply Chain

Conserving resources and protecting the environment is China's basic state policy and an important measure to build a beautiful China.As a major coal consuming industry,the amount of carbon dioxide produced by the power industry accounts for a large proportion of the total national carbon emissions.It is of great significance to actively promote the energy conservation and emission reduction of thermal coal for the cause of environmental protection in China.Therefore,this paper takes the whole thermal coal supply chain as the research object to analyze and forecast the carbon emissions produced by it,in order to provide reference for the control of carbon emissions in the future.Due to electricity coal supply chain from coal mining to eventually all link in power plants is consumed,and the uneven distribution of coal producing area of supply chain link,route length,so the need for electricity coal supply chain structure and the flow of the relationship between each node were analyzed,and find out the key chain nodes,and the entire supply chain of carbon emissions is calculated.This paper for the coal supply chain running complex problems,by using intuitionistic fuzzy Petri net theory of electricity coal supply chain logistics operation mode of the main structure and each node is analyzed,respectively,with the library says said chain node,change relations of fuzzy reasoning,by setting the fuzzy reasoning rules and their confidence in the supply chain,build up the coal supply chain model,Combined with TOPSIS method,the key chain nodes of the supply chain are selected.On the basis of determining the key chain nodes,the carbon emission source and calculation formula of each key chain node are analyzed,and the carbon emission of the whole supply chain is calculated.Then in the database collection affect carbon emissions data,combined with data to calculate the carbon emissions,building predict carbon emissions of GA-BP neural network,through training,testing,predictive value of the carbon emissions,and the predicted values and the actual value of the last two years to compare relative error,assess the accuracy of the model prediction.Finally,the results obtained by this algorithm are compared with other prediction algorithms(partial least squares regression analysis(PLS),extreme learning machine(ELM)).RMSE,MAE and MAPE indexes are used to analyze the errors between the real value and the predicted value,and the correlation analysis is carried out to illustrate the rationality and accuracy of this algorithm.It is instructive to take reasonable measures to control carbon emission in the future.