基于最小二乘支持向量机的电网企业供应链碳排放预测方法研究  

Supply Chain Carbon Emission Prediction for Power Grid Enterprises Based on Least Squares Support Vector Machine

作　　者：卞龙江 李俊颖 胡承鑫 徐友刚 周晓斌 Bian Longjiang;Li Junying;Hu Chengxin;Xu Yougang;Zhou Xiaobin(State Grid Shanghai Electric Power Company Materials Company,Shanghai 200030,China)

机构地区：[1]国网上海市电力公司物资公司,上海200030

出　　处：《环境科学与管理》2024年第2期16-21,共6页Environmental Science and Management

摘　　要：电网企业供应链碳排放的预测对推动产业链供应链绿色转型具有重要意义,为此提出基于最小二乘支持向量机的电网企业供应链碳排放预测方法。首先,利用4E平衡模型获取电网企业供应链碳排放数据;其次,利用PLS-VIP算法对碳排放数据实施数据筛选,得到有效的碳排放数据变量;最后,引入最小二乘支持向量机,构建碳排放预测模型,并且采用量子粒子群优化算法对其展开优化,实现电网企业供应链碳排放高精度预测。实验结果表明,所提方法在保证预测过程较高稳定性的同时,一定程度上提高了预测精度和预测效率。The prediction of supply chain carbon emissions of power grid enterprises is of great significance to promote the green transformation of industrial chain supply chain.Therefore,a method of supply chain carbon emissions prediction of power grid enterprises supply chain based on least square support vector machine is proposed.Firstly,the 4E balance model is used to obtain the supply chain carbon emission data of power grid enterprises.Secondly,PLS-VIP algorithm is used to screen carbon emission data and obtain effective carbon emission data variables.Finally,the least square support vector machine is introduced to build the carbon emission prediction model,and the quantum particle swarm optimization algorithm is used to optimize it,so as to realize the high-precision prediction of the supply chain carbon emission of power grid enterprises.The experimental results show that the proposed method not only guarantees the stability of the prediction process,but also improves the prediction accuracy and efficiency to a certain extent.

关 键 词：最小二乘支持向量机 4E平衡模型 PLS-VIP算法 数据筛选 碳排放预测模型 

分 类 号：X32[环境科学与工程—环境工程]