Probabilistic density-based regression model for soft sensing of nonlinear industrial processes. (September 2017)

• Record Type: Journal Article
• Authors: Yuan, Xiaofeng; Wang, Yalin; Yang, Chunhua; Gui, Weihua; Ye, Lingjian
• Is Part Of: Journal of process control. Volume 57(2017)
• Page Start: 15
• Page End: 25
• Publication Date: 2017-09
• Subjects: Soft sensor -- Quality prediction -- Gaussian mixture model -- Weighted Gaussian regression -- Locally weighted learning; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2017.06.002
• Languages: English
• ISSNs: 0959-1524

Abstract

Highlights: A Probabilistic density-based regression is proposed for soft sensing of nonlinear industrial processes. A weighted Gaussian model is first constructed to model the joint distribution of input and output variables around the query sample. The query output can be predicted by taking the conditional distribution P(y|x) from the local weighted Gaussian model. Feasibility and efficiency of the developed method is tested through a numerical example and an industrial process. Abstract: Process nonlinearity is a challenging issue for soft sensor modeling of industrial plants. Traditional nonlinear soft sensing methods are not achieved through the probabilistic manner, which only give single point estimation for output variables but do not provide the prediction uncertainty. To meet the probabilistic soft sensor requirement, a novel density-based regression method, which is called weighted Gaussian regression (WGR), is proposed in this paper. By taking the weights of training samples into consideration, a local weighted Gaussian model (WGM) is first built to model the joint density P ( x, y ) of input and output variables around the query sample. Then, the output variables can be estimated by taking the conditional distribution P ( y | x ). The new method can successfully approximate the nonlinear relationship between output and input variables. Moreover, WGR can provide more detailed information of uncertainty for the prediction. The effectiveness and flexibility of WGR are validated through a numerical example and an industrial debutanizer column process.