Semisupervised Bayesian Method for Soft Sensor Modeling with Unlabeled Data Samples

2012-03-23 Vistors:1189

Ge, Zhiqiang; Song, Zhihuan

AICHE JOURNAL Volume : 57   Issue: 8  2011,    Pages: 2109-2119    

In practical soft sensor modeling, may only obtain the output data for a small portion of the whole dataset, and have much more input data samples. In this paper, a semi-supervised method is proposed for soft sensor modeling, which can successfully incorporate the unlabeled data information. In order to determine the effective dimensionality of the latent space, the Bayesian regularization method is introduced into the semi-supervised model structure. The feasibility and efficiency of new developed soft sensor are evaluated through a debutanizer column case study.

Contact US
Email : fit@zju.edu.cn
Tel:86-571-87951772
FAX : 0086-571-87951077
Addr:No. 38 ZheDa Road