摘要
目的精确预测天然气净化厂尾气处理装置烟气中二氧化硫(SO_(2))排放质量浓度。方法利用某天然气净化厂2018—2023年每小时44000条尾气处理日报数据构建数据集,进行数据处理,并利用重要性分析方法提取27个重要特征。针对烟气中SO_(2)排放质量浓度的预测任务,采用了随机森林(Random Forest)、梯度提升(Gradient Boost)和极值梯度提升(XGBoost)3种集成学习算法,以及基于径向基(RBF)内核的支持向量机(SVM)替代仿真模型进行建模。结果3种集成学习模型比SVM单模型的预测效果更为精准,而Random Forest模型展现出最佳性能,决定系数为0.89,均方误差为1250.59,相对于8800个真实测试集样本数据,其预测偏差为9.86%,相比于Random Forest模型(数据未处理),其决定系数提高了61.82%。结论Random Forest模型在准确预测尾气处理装置SO_(2)排放质量浓度方面具有实际生产应用价值,可为后续尾气处理装置的工艺参数优化提供可靠的模型支持。
Objective The aim is to accurately predict the emission mass concentration of sulfur dioxide(SO_(2))in the flue gas of the tail gas treatment unit of natural gas purification plants.Method The data set was constructed using 44000 hourly tail gas treatment daily report data from a natural gas purification plant from 2018 to 2023.Data processing was conducted,and 27 important features were extracted using importance analysis methods.Aiming at the prediction task of SO_(2)emission mass concentration in flue gas,three ensemble learning algorithms—namely,Random Forest,Gradient Boost,and XGBoost—and a Support Vector Machine(SVM)based on a Radial Basis Function(RBF)kernel were used to model the process instead of simulation models.Result The prediction accuracy of the three ensemble learning models was higher than the SVM single model.Among them,the Random Forest model exhibited the best performance,with a coefficient of determination of 0.89 and a mean square error of 1250.59.Relative to a data set containing 8800 real test set samples,its prediction deviation was 9.86%.Compared to the Random Forest model without data treatment,its coefficient of determination increased by 61.82%.Conclusion The Random Forest model has practical production application value in accurately predicting SO_(2)emission mass concentration of the tail gas treatment unit and can provide reliable model support for the subsequent process parameter optimization of the tail gas treatment unit.
作者
张宝东
杜支文
闫昭
侯磊
ZHANG Baodong;DU Zhiwen;YAN Zhao;HOU Lei(PetroChina Changqing Oilfield Company,Xi'an,Shaanxi,China)
出处
《石油与天然气化工》
北大核心
2025年第1期9-17,共9页
Chemical engineering of oil & gas
关键词
天然气净化
硫磺回收
尾气处理
二氧化硫排放
预测模型
集成学习算法
natural gas purification
sulfur recovery
tail gas treatment
sulfur dioxide emission
prediction model
ensemble learning algorithm
