基于LASSO-GSWOA-KELM模型的石化行业碳排放预测研究
Prediction of carbon dioxide emission in petrochemical industry based on LASSO-GSWOA-KELM model
余博 1王尹 1柴俊松 1乔子恒 1孙野1
作者信息
- 1. 南京财经大学金融学院,江苏 南京 210023
- 折叠
摘要
石化行业是碳排放的重要来源,构建精确预测石化行业碳排放的模型对我国实现"双碳"目标具有重要意义.通过采用STIRPAT模型和LASSO回归筛选影响碳排放的关键因素,并利用全局搜索策略的鲸鱼优化算法优化KELM模型以提高预测精度,构建了 LASSO-GSWOA-KELM模型.实证结果显示,该模型预测精度超过其他模型,证明该模型为准确预测石化行业碳排放提供了有效工具.预测结果显示,我国石化行业碳排放将继续增长但增速放缓,预计在2029年达峰值.针对研究结果,提出了发展CCUS技术、淘汰落后产能、建立绿色金融体系等建议,为石化行业减排提供理论和决策支持.
Abstract
The petrochemical industry is a significant source of carbon dioxide emission.Constructing a model for accurately predicting carbon dioxide emission in the petrochemical industry is of significance for China to achieve"carbon dioxide emission peaking and carbon neutrality"goals.LASSO-GSWOA-KELM model is constructed through adopting STIRPAT model and LASSO regression to screen key factors affecting carbon dioxide emission,and optimizing KELM model by the whale optimization algorithm,which utilizes global search strategy,to improve prediction accuracy.Empirical results show that the prediction accuracy of this model surpasses those of other models,proving that this model provides an effective tool for accurately predicting carbon dioxide emission in the petrochemical industry.The prediction results show that the carbon dioxide emission in China's petrochemical industry will continue to grow but at a slowing pace,and is expected to peak in 2029.According to the study results,it is suggested to develop CCUS technology,eliminate outdated production capacity,and establish a green financial system,providing theoretical and decision-making support for emissions reduction in the petrochemical industry.
关键词
石化行业/碳排放预测/鲸鱼优化算法/KELM/机器学习Key words
petrochemical industry/carbon emission prediction/whale optimization algorithm/KELM/machine learning引用本文复制引用
出版年
2024
NETL
NSTL
万方数据