Research On Industrial Carbon Emission Prediction Based On Improved PSO-LSSVM Model In Hebei Province

Hebei Province is a major industrial province and a carbon emitter in China.Industrial enterprises consume a large amount of fossil fuels such as coal and emit a large amount of carbon dioxide every year,causing serious damage to the environment of China.As the international status of China and the responsibilities is increasing in recent years,in order to actively cooperate with China in fulfilling its commitments at international climate change conferences,Hebei Province should take action for emission reduction and sustainable economic development.Hebei need to develop green and low-carbon economy,reduce fossil fuel consumption,promote economic restructuring,only in this way can truly achieve energy conservation,emission reduction and sustainable development.Because Hebei has more industrial enterprises,so industrial carbon emissions accounts for high proportion of carbon emissions of Hebei.Therefore,this article uses Hebei’s industrial carbon emissions as a research object to explore the factors affecting industrial carbon emissions of Hebei and predict industrial carbon emissions in Hebei.Finally the author make recommendations according to the result,which have certain practical significance for energy saving and emission reduction of Hebei.This paper firstly conforms industrial carbon emissions in Hebei Province as the research object.First of all,it introduces the background of the article and the current research status of carbon emissions at home and abroad,then studies the relevant carbon emission theories,as well as the introduction and learning of the PSO algorithm and the LSSVM algorithm.Based on the factor analysis,a factor index system that affects industrial carbon emissions in Hebei Province was constructed.Data was collected and processed in accordance with the factor index system to obtain the energy consumption and carbon emissions of the industrial terminals of Hebei province over the years.Then the grey correlation degree method was used to analyze the influence degree of each impact factor on industrial carbon emissions.The results show that each impact factor has a high correlation coefficient with carbon emissions,indicating that all selected indicators are reasonable and have an obvious impact on industrial carbon emission of Hebei.For the characteristics of particle swarm optimization and least squares support vector machine,this article introduces an adaptive weight adjustment scheme to promote the PSO-LSSVM prediction model,further improving the premature and local optimal problems in the model optimization process.The improved PSO-LSSVM prediction model is constructed and the training set and test set are substituted into the improved PSO-LSSVM model to obtain the optimal parameters and prediction results.The research results obtained through modeling and prediction are as follows: The analysis results of the gray correlation method show that the most important factor affecting industrial carbon emissions of Hebei is the total energy consumption,and the lowest degree is government revenue.The prediction results show that the prediction accuracy of the improved PSO-LSSVM model is higher than that of the LSSVM model and the unimproved PSO-LSSVM model.It shows that the adaptive weight adjustment scheme can effectively improve the prediction accuracy of the PSO-LSSVM model.The scenario analysis method sets the development state of the next five years and substitutes it into the improved model to get the predict results.And the results show that the carbon emissions of Hebei’s industry will decline first and then rise in the next five years.After the prediction results are obtained,countermeasures are proposed for the factors that affect industrial carbon emissions in Hebei Province.It is of certain reference for Hebei Province to develop a green and low-carbon economy.