Bayesian network for dynamic variable structure learning and transfer modeling of probabilistic soft sensor. (April 2021)

• Record Type: Journal Article
• Authors: Zeng, Lingquan; Ge, Zhiqiang
• Is Part Of: Journal of process control. Volume 100(2021)
• Page Start: 20
• Page End: 29
• Publication Date: 2021-04
• Subjects: Dynamic Bayesian network -- Variable structure transferring -- Soft sensor -- Quality prediction; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2021.02.004
• Languages: English
• ISSNs: 0959-1524

Abstract

Abstract: Bayesian network is a frequently-used uncertainty reasoning method, which systematically describes relations between random variables. Dynamic Bayesian network is an extension of Bayesian network, which contains the relations between variables at different times. Soft sensor is an important industrial application, in which feature variables are selected to predict the value of the target variables. For industrial soft sensor applications, dynamics is still a tough problem, particularly together with the uncertain feature of process data. In this article, DBN is employed for dynamic variable structure learning and transfer modeling to some strong regression models to build a soft sensor for dynamic industrial processes. At the beginning, a series of dynamic Bayesian networks are constructed on the training set with a sliding window. From these network structures we can find variables related to the quality variables. Then, in each time period, a sequence of data is compared with training data to select the most similar sequence by Dynamic Time Warping. Therefore, the structure of variables is built, i.e., the related feature variables of the quality variables can be determined by the network structure. In the regression step, the dynamic variable structure is transferred to some strong regressors, like Support Vector Regression and Adaboost for further regression. In case study, we use the debutanizer process and a low temperature transformer case to confirm the quality of the soft sensor method. The result reveals that, the prediction accuracy of the new method is much higher than the original commonly-used regression methods. Highlights: A soft sensor based on Dynamic Bayesian Network is proposed for industrial processes. A series of DBNs are developed to describe the dynamics of the process. Dynamic Time Warping is used for similarity selection for online data sequences. Variable structure is transferred to regression models for soft sensor development.