Prediction Of Carbon Emissions Of China’s Electric Power Industry Using BAG-SA

With the acceleration of China’s urbanization and industrialization,the increment of energy demand stimulates the rapid growth of carbon emissions.Therefore,carbon reduction pressure has become increasingly prominent.Electric power industry is regarded as the main battlefield of carbon emissions reduction due to carbon emissions accounting for about 40%of the total carbon emissions.The accurate carbon emissions prediction of electric power industry can offer a scientific basis for the formulation of the national power industry development strategy,and further promote the sustainable,healthy and rapid development of the national economy.In this paper,the carbon emissions prediction of China’s electric power industry can be divided into two steps:firstly,considering the factors of national economic development the future electricity demand is forecasted;secondly,the carbon emissions of electric power industry is predicted based on meeting future electricity demand and considering the production factors.In terms of electricity demand forecasting,the economic growth rate,the rate of change of population,the rate of urbanization and industrial structure change rate are firstly selected as the main influencing factors of electricity demand.Using the stationarity test,cointegration test and Granger causality test,it can be proved that there are some significant and stable relationships between those factors and electricity demand.In the light of that relationship,the multiple linear and nonlinear models are constructed as the primary model for prediction.Then,using the bat algorithm improved by simulated annealing algorithm and Gaussian perturbation algorithm to estimate the alternative model parameters,the results show that the improved algorithm not only greatly enhance the global search ability and local search ability of the classical bat algorithm,but also improve the convergence speed of the algorithm.Based on the evaluation criteria of the model accuracy（MAPE andR²）,in this paper the nonlinear equation is chosen as the final model for electricity demand forecasting because of its stronger fitting ability.Then,using scenario analysis method,the electricity demand can be obtained according to the economic growth rate and the change level of industrial structure,the results display that electricity demand of China will be 7490.7-7974.6 billion kWh and 10581.6-13489.7 billion kWh in 2020 and 2030respectively.Further,under the premise of the same population level and urbanization level,the rapid adjustment of industrial structure has great potential to reduce the total demand for electricity in the whole society.According to the prediction results of electricity demand,the carbon emissions of China in 2020 and 2030 can be further forecasted combined with the adjustment factors of power generation technology.The experimental results show that the carbon emissions intervals of2020 and 2030 are 4186-4516 and 5313-7003Mtc respectively;the carbon emission intensity are 558-566 and 502-516 gCO₂/kWh.Fulfilling the tasks that the carbon emission intensity in2020 and 2030 respectively decreased by 40-45%and 60-65%compared with 2005 is very arduous.Thus,in the future China’s carbon emission reduction should focus on accelerating the speed of power generation technology restructuring,and the improvement of the proportions of clean energy and large capacity,high efficiency power generation unit.