[1]吴兴惠,吴 迪,周玉萍,等.基于机器学习算法的稀土元素掺杂TiO2光催化活性分析[J].南京师范大学学报(工程技术版),2017,17(03):087.[doi:10.3969/j.issn.1672-1292.2017.03.013]
　Wu Xinghui,Wu Di,Zhou Yuping,et al.Photocatalytic Activity Prediction of Rare Earth Doped TiO2Based on Machine Learning Algorithm[J].Journal of Nanjing Normal University(Engineering and Technology),2017,17(03):087.[doi:10.3969/j.issn.1672-1292.2017.03.013]
点击复制

基于机器学习算法的稀土元素掺杂TiO2光催化活性分析
分享到：

南京师范大学学报（工程技术版）[ISSN:1006-6977/CN:61-1281/TN]

卷:
17卷
期数:
2017年03期
页码:
087
栏目:
计算机工程
出版日期:
2017-09-30

文章信息/Info

Title:
Photocatalytic Activity Prediction of Rare Earth Doped TiO2Based on Machine Learning Algorithm
文章编号:
1672-1292(2017)03-0087-06
作者:
吴兴惠1; 吴 迪2; 周玉萍1; 史载锋2
1.海南师范大学信息科学与技术学院,海南 海口 571158)(2.海南师范大学化学与化工学院,海南 海口 571158)
Author(s):
Wu Xinghui1; Wu Di2; Zhou Yuping1; Shi Zaifeng2
(1.School of Computer Science and Technology,Hainan Normal University,Haikou 571158,China)(2.School of Chemistry and Chemical Engineering,Hainan Normal University,Haikou 571158,China)
关键词:
稀土掺杂二氧化钛; 数据分析; 机器学习; 光催化活性预测
Keywords:
rare earth doped TiO2; data analysis; machine learning; photocatalytic activity prediction
分类号:
TP181
DOI:
10.3969/j.issn.1672-1292.2017.03.013
文献标志码:
A
摘要:
机器学习是人工智能及机器学习领域的共同研究热点,其理论和方法已被广泛应用于解决工程应用和科学领域的复杂问题. 在化学领域,稀土金属掺杂二氧化钛光催化剂提高光催化活性的研究已有大量研究,但掺杂机理一直存在争论,基于第一性原理的算法复杂且存在误差. 为了探索能够不需要化学结构数学模型,以元素电子结构等基础数据为先验知识,通过计算机算法准确预测光催化剂活性,采用线性回归、高斯过程回归、支持向量机回归、K-最近邻法对稀土TiO2光电性质进行预测研究,并通过实验验证基于逐步回归分析的关键因素分析的有效性与基于关键因素的k-NN回归模型的优势. 结果表明,使用逐步回归分析得到的关键特征所得的预测性能更好. 对比不同的回归方法的预测性能可知,k-NN的预测性能最好.
Abstract:
Machine learning is one of the hot points of the research in the area of artificial intelligence and pattern recognition,and its theories and methods have been widely used to resolve complex problems in science and engineering applications. In the chemical area,rare earth doped TiO2 photocatalysts for improving photocatalytic activity has been massively studied and got many important results. However,the doping mechanism is not fully clear,even though the first-principle algorithm is applied to mechanism simulation,the calculation process is complicated and there are many inevitable errors. To solve the problem,it is expected that the simulation can be done without complicated chemical model,the prediction accuracy can be improved through a simply computer algorithm only based on element’s essential data as the prior knowledge. In the present paper,the linear regression,Gaussian process regression and support vector machines regression,k-nearest neighbor algorithm are used to predict the photocatalytic reaction rate constant of rare earth doped TiO2. Then,experiments are used to verify the key factor analysis effectiveness based on stepwise multiple regression and advantages of the key factor analysis based on k-NN regression model. Results show that the pridiction performance is better with key factors obtained from stepwise regression analysis. Compared with other regression methods,it may be found the prediction performance of k-NN is the best.

参考文献/References:

[1] SNELSON E,RASMUSSEN C E,GHAHRAMANI Z. Warped Gaussian process[C]//Proc of the NIPS 16. Vancouver,British Columbia,Canada,2004.
[2]NGUYEN T D,PETERS J. Incremental online sparsification for model learing in realtime robot control[J]. Neurocomputing,2011,74(11):1 859-1 867.
[3]PETELIN D,KOCIJAN J. Control system with evolving Gaussian process model[C]//Proc of IEEE Symposium Series on Computational Intelligence,2011.
[4]MUSIZZA B,PETELIN D,KOCIJAN J. Accelerated learning of Gaussian process models[C]//Proc of the 7th EUROSIM Congress on Modelling and Simulation. Praha,CZ,VCVUT,2010.
[5]SHI H X,ZHANG T Y,AN T C,et al. Enhancement of photocatalytic activity of nano-scale TiO2 particles co-doped by rare earth elements and heteropolyacids[J]. Journal of colloid and interface science,2012,380(1):121-127.
[6]DI S C,GUO Y P,Lü H W,et al. Microstructure and properties of rare earth CeO2-doped TiO2 nanostructured composite coatings through micro-arc oxidation[J]. Ceramics international,2015,41(5A):6 178-6 186.
[7]VIGNESH C B,DANIEL R R,JOHN N K. Assessment of mechanisms for enhanced performance of Yb/Er/titania photocatalysts for organic degradation:role of rare earth elements in the titania phase[J]. Applied catalysis B(environmental),2017,202:156-164.
[8]HüSNü A Y,MUHSIN ?. The effect of rare earth element doping on the microstructural evolution of sol-gel titania powders[J]. Journal of alloys and compounds,2017,695:1 336-1 353.
[9]YUAN W J,ZHU Q Y,DENG C J,et al. Effects of rare earth oxides additions on microstructure and properties of alumina-magnesia refractory castables[J]. Ceramics international,2017,43(9):6 746-6 750.
[10]AOIFE L,ANNA I,MICHAEL N. A first principles investigation of Bi2O3-modified TiO2 for visible light Activated photocatalysis:the role of TiO2 crystal form and the Bi3+ stereochemical lone pair[J]. Materials science in semiconductor processing,2014,25:59-67.
[11]LI S J,QIU H,WANG C D,et al. Highly efficient NaTaO3 for visible light photocatalysis predicted from first principles[J]. Solar energy materials and solar cells,2016,149:97-102.
[12]ZENG X Y,XIAO X Y,ZHANG W P,et al. Interfacial charge transfer and mechanisms of enhanced photocatalysis of an anatase TiO2(0 0 1)-MoS2-graphene nanocomposite:A first-principles investigation[J]. Computational materials science,2017,126:43-51.
[13]FU A M,WANG X Z,HE Y L,et al. A study on residence error of training an extreme learning machine and its application to evolutionary algorithms[J]. Neurocomputing,2014,146:75-82.
[14]ZHENG W B,FU X P,YING Y B. Spectroscopy-based food classification with extreme learning machine[J]. Chemometrics and intelligent laboratory systems,2014,139:42-47.
[15]WOUBISHET Z T,ESKO S. Machine learning for durability and service-life assessment of reinforced concrete structures:recent advances and future directions[J]. Automation in construction,2017,77:1-14.
[16]LIU Y,BI J W,FAN Z P. Multi-class sentiment classification:the experimental comparisons of feature selection and machine learning algorithms[J]. Expert systems with applications,2017,80:323-339.
[17]JOSé M M M,PABLO E M,CARLO B,et al. Prediction of the hemoglobin level in hemodialysis patients using machine learning techniques[J]. Computer methods and programs in biomedicine,2014,17(2):208-217.
[18]DI K C,LI W,YUE Z Y,et a;. A machine learning approach to crater detection from topographic data[J]. Advances in space research,2014,54(11):2 419-2 429.
[19]MA C,ZHANG H H,WANG X F. Machine learning for Big Data analytics in plants[J]. Trends in plant science,2014,19(12):798-808.
[20]TAYFUN K,CAMBAZOGLU B B,CEVDET A,et al. A machine learning approach for result caching in web search engines[J]. Information processing and management,2017,53(4):834-850.
[21]CYRIL V,GILLES N,SOTERIS K,et al. Machine learning methods for solar radiation forecasting:a review[J]. Renewable energy,2017,105:569-582.

备注/Memo

备注/Memo:
收稿日期:2017-03-18.
基金项目:海南省自然科学基金(20156242).
通讯联系人:史载锋,博士,教授,博士生导师,研究方向:化学. E-mail:zaifengshi@163.com
更新日期/Last Update: 2017-09-30