Similar articles

Abstract: soft sensors are widely used to predict quality variables which are usually hard to measure. It is necessary to construct an adaptive model to cope with process non-stationaries. In this study, a novel quality-related locally weighted soft sensing method is designed for non-stationary processes based on a Bayesian network with latent variables. Specifically, a supervised Bayesian network is proposed where quality-oriented latent variables are extracted and further applied to a double-layer similarity measurement algorithm. The proposed soft sensing method tries to find a general approach for non-stationary processes via quality-related information where the concepts of local similarities and window confidence are explained in detail. The performance of the developed method is demonstrated by application to a numerical example and a debutanizer column. It is shown that the proposed method outperforms competitive methods in terms of the accuracy of predicting key quality variables.

基于含隐变量的贝叶斯网络质量相关局部加权的非平稳过程软测量方法

[1]Al-Jlibawi A, Othman MLB, Al-Huseiny MS, et al., 2019. Efficient soft sensor modelling for advanced manufacturing systems by applying hybrid intelligent soft computing techniques. Int J Simul Syst Sci Technol, 19(3):15.1-15.7.

[2]Atkeson CG, Moore AW, Schaal S, 1997. Locally weighted learning. In: Aha DW (Ed.), Lazy Learning. Springer, Dordrecht, p.11-73.

[3]Ben‐Gal I, 2008. Bayesian networks. In: Ruggeri F, Kenett RS, Faltin FW (Eds.), Encyclopedia of Statistics in Quality and Reliability. John Wiley & Sons, Chichester, p.1.

[4]Bidar B, Sadeghi J, Shahraki F, et al., 2017. Data-driven soft sensor approach for online quality prediction using state dependent parameter models. Chemom Intell Lab Syst, 162:130-141.

[5]Bishop CM, 1998. Latent variable models. In: Jordan MI (Ed.), Learning in Graphical Models. MIT Press, Cambridge, p.371-403.

[6]Cain C, 2016. Modelling latent variables for Bayesian networks. Undergraduate Research Symp. Available from https://digitalcommons.morris.umn.edu/urs_2016/2/

[7]Chang SY, Baughman EH, McIntosh BC, 2001. Implementation of locally weighted regression to maintain calibrations on FT-NIR analyzers for industrial processes. Appl Spectrosc, 55(9):1199-1206.

[8]Chen GF, Yu HL, 2007. Bayesian network and its application in maize diseases diagnosis. Proc Int Conf on Computer and Computing Technologies in Agriculture, p.917-924.

[9]Fortuna L, Graziani S, Rizzo A, et al., 2007. Soft Sensors for Monitoring and Control of Industrial Processes. Springer, London, UK.

[10]Frank E, Trigg L, Holmes G, et al., 2000. Technical note: naive Bayes for regression. Mach Learn, 41(1):5-25.

[11]Frank E, Hall M, Pfahringer B, 2002. Locally weighted naive Bayes. Proc 19^th Conf on Uncertainty in Artificial Intelligence, p.249-256.

[12]Ge ZQ, 2016. Supervised latent factor analysis for process data regression modeling and soft sensor application. IEEE Trans Contr Syst Technol, 24(3):1004-1011.

[13]Ge ZQ, 2018. Process data analytics via probabilistic latent variable models: a tutorial review. Ind Eng Chem Res, 57(38):12646-12661.

[14]Ge ZQ, Song ZH, Ding SX, et al., 2017. Data mining and analytics in the process industry: the role of machine learning. IEEE Access, 5:20590-20616.

[15]Geiger D, Heckerman D, 1994. Learning Gaussian networks. In: de Mantaras RL, Poole D (Eds.), Uncertainty in Artificial Intelligence: Proceedings of the Tenth Conference (1994). Morgan Kaufmann Publishers, California, USA, p.235-243.

[16]Jiang LX, Cai ZH, Zhang H, et al., 2013. Naive Bayes text classifiers: a locally weighted learning approach. J Exp Theor Artif Intell, 25(2):273-286.

[17]Kadlec P, Gabrys B, Strandt S, 2009. Data-driven soft sensors in the process industry. Comput Chem Eng, 33(4):795-814.

[18]Kadlec P, Grbić R, Gabrys B, 2011. Review of adaptation mechanisms for data-driven soft sensors. Comput Chem Eng, 35(1):1-24.

[19]Kano M, Fujiwara K, 2012. Virtual sensing technology in process industries: trends and challenges revealed by recent industrial applications. J Chem Eng Jpn, 46(1):1-17.

[20]Kim JS, Jun CH, 2013. Ranking evaluation of institutions based on a Bayesian network having a latent variable. Knowl-Based Syst, 50:87-99.

[21]Kim S, Okajima R, Kano M, et al., 2013. Development of soft-sensor using locally weighted PLS with adaptive similarity measure. Chemom Intell Lab Syst, 124:43-49.

[22]Li CQ, Jiang LX, Li HW, 2014. Naive Bayes for value difference metric. Front Comput Sci, 8(2):255-264.

[23]Liu F, Xu DX, Yuan C, et al., 2006. Image segmentation based on Bayesian network-Markov random field model and its application to in vivo plaque composition. Proc 3^rd IEEE Int Symp on Biomedical Imaging: Nano to Macro, p.141-144.

[24]Liu Q, Zhuo J, Lang ZQ, et al., 2018. Perspectives on data-driven operation monitoring and self-optimization of industrial processes. Acta Autom Sin, 44(11):1944-1956 (in Chinese).

[25]Liu ZW, Ge ZQ, Chen GJ, et al., 2018. Adaptive soft sensors for quality prediction under the framework of Bayesian network. Contr Eng Pract, 72:19-28.

[26]Masmoudi K, Abid L, Masmoudi A, 2019. Credit risk modeling using Bayesian network with a latent variable. Expert Syst Appl, 127:157-166.

[27]Mohammadi A, Zarghami R, Lefebvre D, et al., 2019. Soft sensor design and fault detection using Bayesian network and probabilistic principal component analysis. J Adv Manuf Process, 1(4):e10027.

[28]Montáns FJ, Chinesta F, Gómez-Bombarelli R, et al., 2019. Data-driven modeling and learning in science and engineering. Compt Rend Mécan, 347(11):845-855.

[29]Murphy KP, 2001. The Bayes net toolbox for Matlab. Comput Sci Stat, 33(2):1024-1034.

[30]Nie SQ, Zheng M, Ji Q, 2018. The deep regression Bayesian network and its applications: probabilistic deep learning for computer vision. IEEE Signal Process Mag, 35(1):101-111.

[31]Shao WM, Tian XM, Wang P, 2015. Supervised local and non-local structure preserving projections with application to just-in-time learning for adaptive soft sensor. Chin J Chem Eng, 23(12):1925-1934.

[32]Steurtewagen B, van den Poel D, 2020. Machine learning refinery sensor data to predict catalyst saturation levels. Comput Chem Eng, 134:106722.

[33]Tang H, Liu S, 2007. Basic theory of fuzzy Bayesian networks and its application in machinery fault diagnosis. Proc 4^th Int Conf on Fuzzy Systems and Knowledge Discovery, p.132-137.

[34]Vallejo M, de la Espriella C, Gómez-Santamaría J, et al., 2019. Soft metrology based on machine learning: a review. Meas Sci Technol, 31(3):032001.

[35]Wang JB, Shao WM, Song ZH, 2019. Semi-supervised variational Bayesian student’s t mixture regression and robust inferential sensor application. Contr Eng Pract, 92: 104155.

[36]Wu J, Wu B, Pan SR, et al., 2015. Locally weighted learning: how and when does it work in Bayesian networks? Int J Comput Intell Syst, 8(S1):63-74.

[37]Yao L, Ge ZQ, 2017a. Locally weighted prediction methods for latent factor analysis with supervised and semisupervised process data. IEEE Trans Autom Sci Eng, 14(1):126-138.

[38]Yao L, Ge ZQ, 2017b. Moving window adaptive soft sensor for state shifting process based on weighted supervised latent factor analysis. Contr Eng Pract, 61:72-80.

[39]Yuan XF, Ge ZQ, Song ZH, 2014. Locally weighted kernel principal component regression model for soft sensing of nonlinear time-variant processes. Ind Eng Chem Res, 53(35):13736-13749.

[40]Yuan XF, Huang B, Ge ZQ, et al., 2016. Double locally weighted principal component regression for soft sensor with sample selection under supervised latent structure. Chemom Intell Lab Syst, 153:116-125.

[41]Yuan XF, Huang B, Wang YL, et al., 2018. Deep learning-based feature representation and its application for soft sensor modeling with variable-wise weighted SAE. IEEE Trans Ind Inform, 14(7):3235-3243.