Title: Online soft sensing method based on improved weighted Gaussian model
Authors: Xiaoqing Che; Weili Xiong
Addresses: School of Internet of Things Engineering, Jiangnnan University, Wuxi, Jiangsu Province, China ' School of Internet of Things Engineering, Jiangnnan University, Wuxi, Jiangsu Province, China; Key Laboratory of Advanced Process Control for Light Industry (Ministry of Education), Jiangnan University, Wuxi, Jiangsu Province, China
Abstract: Aiming at the problems of delay, nonlinearity and multi-mode and noise pollution in chemical process, a novel soft-sensor modelling method based on improved weighted Gaussian model is proposed. The moving grey relational analysis method is used to extract the process delay information, and the modelling data set is reconstructed to improve the accuracy of the model. The cumulative similarity factor is introduced into the selection rules of the model training set to improve the real-time performance of the model. When the new query sample arrives, an adaptive similarity threshold was used to determine criteria of updating local model of the current operating point, so as to reduce the model update frequency. The improved modelling method was applied to the debutaniser column process. The simulation results show that the model has high precision and real-time performance.
Keywords: variable time delay; moving grey relational analysis; MGRA; weighted Gaussian model; WGM; cumulative similarity factor.
International Journal of Modelling, Identification and Control, 2019 Vol.32 No.3/4, pp.244 - 250
Received: 14 Jan 2019
Accepted: 07 Mar 2019
Published online: 18 Nov 2019 *