基于水泥企业电-碳关系的碳排放监测

Carbon emission monitoring based on analysis from“electricity-carbon”relationship of cement enterprises

基于电力数据的实时性、精确性、覆盖面广等优势,构建水泥企业的电碳指数,运用机器学习算法研究水泥企业生产净购入电量与二氧化碳排放总量的关系,分别建立单样本日度电碳监测模型和多样本年度电碳监测模型,结果表明,9种回归方法中表现最佳的Lasso模型拟合优度(R^(2))为0.915和0.831,水泥企业电碳指数的关键影响因素主要为熟料生产电力排放因子和熟料工段耗电量在全厂用电量的占比,验证了电力相关数据在碳排放监测模型中的重要性,模型应用将有效提高水泥企业的碳排放监测效率和降低碳排放监测成本.

Based on the advantages of electrical data-instantaneity,accuracy and wide coverage,this paper establishes the electricity-carbon index for cement enterprises.By exploring the relationship between purchased electricity and total carbon dioxide emissions using the machine learning method,we formulate and investigate both single-sample daily monitoring models and multiple-sample annual monitoring models.The numerical study demonstrates that the Lasso model outperforms the other nine regression models deployed in the monitoring models,with a goodness of fit(R^(2))of 0.915 and 0.831.The results indicate that the critical factors influencing the electricity-carbon index of cement enterprises are the electricity emission factors in the clinker production and the proportion of electricity consumption in the clinker section,thus underscoring the importance of electrical data in carbon emission monitoring.The proposed model has the potential to significantly improve the efficiency and reduce the cost of carbon emission monitoring for cement enterprises.