基于机器学习的多孔碳材料吸附CO_(2)的关键因素  

Study on the key factors of CO_(2) adsorption by porous carbon materials based on machine learning

作　　者：许大伟 杨榛[1,2] XU Dawei;YANG Zhen(State Key Laboratory of Chemical Engineering,College of Chemical Engineering,East China University of Science and Technology,Shanghai,200237,China;Key Laboratory of Special Functional Polymer Materials and Related Technologies(Ministry of Education),College of Chemical Engineering,East China University of Science and Technology,Shanghai,200237,China)

机构地区：[1]华东理工大学化工学院,化学工程国家重点实验室,上海200237 [2]华东理工大学化工学院,特殊功能高分子材料及相关技术教育部重点实验室,上海200237

出　　处：《环境化学》2024年第8期2646-2657,共12页Environmental Chemistry

基　　金：国家自然科学基金(21978097);中央高校基本科研业务费基金(JKD01211701)资助.

摘　　要：多孔碳材料具有性质稳定、高孔隙率和高比表面积等优点,被广泛应用于气体分离和存储.目前,针对多孔碳材料吸附CO_(2)的优化主要通过“炒菜”的模式,但对于各种结构特性、元素组成对吸附的作用尚不明确.机器学习算法通过构建特征与标签之间的模型从而量化其中的非线性关系,被逐渐运用于高性能多孔碳材料的筛选.为了探究多孔碳材料吸附CO_(2)的关键因素,建立了一个包含65个碳样本、1591个数据点的数据集,以7∶3的比例随机拆分为训练集和测试集,采用线性回归、反向传播神经网络、支持向量机、随机森林等5种不同机器学习模型训练和测试.结果表明:随机森林模型在测试集上具有最佳的泛化能力(R^(2)=0.954、RMSE=0.286).压力是多孔碳材料吸附CO_(2)最重要的影响因素,但随着压力和温度的提高,结构特性和元素组成逐渐成为主导因素.结构特性方面,比表面积和微孔是结构特性中影响CO_(2)吸附能力的关键因素;常压下CO_(2)吸附主要受微孔控制,高压下中孔对吸附也有重要作用.元素组成方面,氮元素是最重要的因素,其特征重要性在273 K、0.5—1 bar时可以达到15.27%.Porous carbon materials are widely used for gas separation and storage because of their stable properties,high porosity and high specific surface area.Currently,the optimization of CO_(2) adsorption on porous carbon materials is mainly through the"frying"mode,but the role of various structural properties and elemental composition on adsorption is not clear.Machine learning algorithms,which quantify the nonlinear relationships by constructing models between features and labels,are gradually being applied to the screening of high-performance porous carbon materials.In order to find the key factors of CO_(2) adsorption by porous carbon materials,a dataset containing 65 carbon samples with 1591 data points is established and randomly split into training and testing sets in the ratio of 7:3.Five different machine learning models,including linear regression,backward propagation neural network,support vector machine,and random forest,are used for training and testing.The results show that the random forest model has the best generalization ability on the test set(R^(2)=0.954,RMSE=0.286).Pressure is the most important influencing factor for CO_(2) adsorption by porous carbon materials,but with increasing pressure and temperature,structural properties and elemental composition gradually become the dominant factors.As for the structural properties,specific surface area and micropores are the key factors affecting CO_(2) adsorption capacity in structural properties;CO_(2) adsorption at atmospheric pressure is mainly controlled by micropores,mesopores also play an important role in adsorption under high pressure.As for elemental composition,nitrogen is the most important factor,and its characteristic importance can reach 15.27%at 273 K and 0.5-1 bar.

关 键 词：多孔碳材料 孔结构 二氧化碳吸附 机器学习. 

分 类 号：X-1[环境科学与工程] O6[理学—化学]