基于集成学习的钢铁(高炉)行业碳排放预测方法

Carbon Emission Prediction Method of Steel(Blast Furnace)Industry Based on Ensemble Learning

钢铁工业是能源消耗大、碳排放大的典型产业。中国的钢铁业是全球碳排放最多的行业之一。目前,由于缺乏监测数据,对钢铁行业企业碳排放预测的研究很少。为了有效解决这一问题,提出一种基于集成学习的钢铁(高炉)行业碳排放预测方法。选取具有较好计量手段的用电量作为碳排放影响因素,并结合高炉和电弧炉工艺,采用了三种机器学习模型,分别是反向传播(back propagation,简称BP)神经网络、支持向量机和随机森林,这些模型在机器学习领域具有广泛的应用前景。利用Shapley集成学习方法进行碳排放预测。通过仿真实验验证了本文所提的以用电量输入的钢铁行业碳排放预测模型具有理想的有效性与准确性。该结果可以为钢铁行业企业在减排和资源优化方面提供科学依据。

The steel industry is a typical industry with high energy consumption and amplified carbon emissions.China's steel industry is one of the industries with the highest carbon emissions in the world.At present,due to the lack of monitoring data,there is little research on predicting carbon emissions of steel industry enterprises.To effectively address this issue,a carbon emission prediction method for the steel(blast furnace)industry based on ensemble learning is proposed.It selects electricity consumption with good measurement methods as the influencing factor of carbon emissions,and combines blast furnace and electric arc furnace processes to adopt three machine learning models,namely Back Propagation(BP)neural network,Support Vector Machine,and Random Forest.These models have broad application prospects in the field of machine learning.Using Shapley ensemble learning method for carbon emission prediction.The effectiveness and accuracy of the proposed carbon emission prediction model for the steel industry based on electricity consumption input have been verified through simulation experiments.This result can provide scientific basis for steel industry enterprises in emission reduction and resource optimization.