基于CF-EEMD-LSSVR算法的铅冶炼系统温室气体排放的评估与预测  

Evaluation and prediction of greenhouse gas emission of lead smelting system based on CF-EEMD-LSSVR

作　　者：罗曦 王洪才[2] 李玉强 

机构地区：[1]中南大学建筑与艺术学院,湖南长沙410083 [2]中南大学能源科学与工程学院,湖南长沙410083

出　　处：《中南大学学报（自然科学版）》2018年第1期15-21,共7页Journal of Central South University:Science and Technology

基　　金：国家自然科学基金资助项目(51478470)~~

摘　　要：利用碳足迹理论建立铅冶炼系统生命周期内各工序的投入产出模型,对单位产品温室气体排放进行评估。针对温室气体排放时间序列的非线性,建立1个基于集合经验模态分解法与最小二乘支持向量回归机相结合的预测模型。集合经验模态分解法首先将温室气体排放时间序列分解成一系列相对比较平稳的本征模函数分量,然后利用最小二乘支持向量回归机对各分量分别预测,最后进行叠加求和,将铅冶炼系统温室气体排放量的预测结果与实际结果进行对比。研究结果表明:预测结果与实际结果均方根误差为2.896 1%,所提出的方法可实现铅冶炼系统温室气体排放的精确评估与预测。Input-output （I-O） model for each step of lead smelting system to evaluate greenhouse gas emission per unit product was established based on carbon footprint （CF） theory. Due to the nonlinear characteristic of greenhouse gas emission data, a prediction model was developed based on the combination of ensemble empirical mode decomposition （EEMD） and the least square support vector regression （LSSVR）.The procedures were as follows： the data of greenhouse gas emission of leads melting system was firstly decomposed into a series of relatively stable intrinsic mode functions （IMF）, and then they were separately predicted by LSSVR. The predicted values were compared with the real results. The results show that the root mean square error of the predicted values and the real results is 2.896 1%, which verifies that the proposed method can realize the accurate evaluation and prediction of greenhouse gas emission of lead smelting system.

关 键 词：铅冶炼系统 温室气体排放 碳足迹 集合经验模态分解 最小二乘支持向量回归机 

分 类 号：TF8[冶金工程—有色金属冶金]