Research On Scenario Prediction Of Carbon Emissions In Guangdong Province Based On CSO-FLN Model

Reducing carbon emissions and developing low-carbon economy are important measures to address climate change and build a community with a shared future for mankind,which has become a global consensus.As the world’s second largest economy and the largest carbon emitter,China has put forward the goal of reaching carbon emissions peaking by 2030 and the vision of striving to achieve carbon neutrality by 2060 in the form of agreements.Facing the severe situation of emission reduction,exploring the evolution path of provincial carbon emissions is crucial for realizing China’s " double carbon" goal.Gaungdong is the most economically developed province in China with the largest population,and one of the pilot provinces and cities of carbon trading,which plays the role of window and experimental field in economy,culture and technology.Analyzing the current situation and the mechanism of influencing factors of carbon emissions in Guangdong,and predicting carbon emissions under different scenarios scientifically will help to explore the optimal path for the realization of the "dual carbon" goal and provide reference for the transformation of green and low-carbon development in other provinces and cities.On the basis of studying and summarizing the existing literatures on carbon emission measurement,influencing factor analysis and change trend prediction,this paper determines the content and method of carbon emission research in Guangdong Province.Firstly,based on the data of various carbon sources and carbon sinks in Guangdong,the carbon emissions from 1995 to 2019 are calculated,and the current situation of carbon emission is comprehensively analyzed by three indicators: total carbon emission,net carbon emission intensity and per capita carbon emission.Secondly,population size,affluence,industrial structure,technical level,energy consumption structure and opening to the outside world are selected as the influencing factors of carbon emissions,and the change characteristics of each factor are analyzed qualitatively.The significant correlation between each influencing factor and carbon emissions is verified by regression analysis with the extended STIRPAT model.A Fast Learning Network model(FLN)based on Chicken Swarm Optimization(CSO)is constructed,and the data of carbon emissions and influencing factors from 1995 to2019 are used as training samples to verify the feasibility and superiority of CSO-FLN model.Finally,according to the relevant policies and planning objectives of Guangdong,the positive and negative factors of carbon emission are combined to form nine scenarios,and the carbon emissions under nine scenarios from 2020 to 2060 are predicted by the CSO-FLN model.The main conclusions of this study are included:(1)From 1995 to 2019,the total carbon emissions and per capita carbon emission in Guangdong generally showed an increasing trend,and the change curves are similar.The carbon emission intensity showed a downward trend,while the carbon sinks fluctuated steadily without obvious changes.(2)Population size,affluence,and industrial structure play a positive role in the growth of carbon emissions,while technological level,energy structure,and opening to the outside world play an inhibitory role,among which industrial structure has the most significant impact.(3)The fitting effect of CSO-FLN model is better than Fast Learning Network(FLN)and Extreme Learning Machine(ELM),and the prediction accuracy is also higher than a single prediction model,which can be used as a feasible method to forecast carbon emissions under various scenarios.(4)Under nine scenarios,the carbon emissions of Guangdong all show an "inverted U-shaped" change characteristic of first rise and then decline,and the carbon peak time is between2032 and 2035;only in scenarios 7、8、9,can Guangdong achieve carbon neutrality by 2060 or before.