1547-5905/asset/AIC_left.gif?v=1&s=43a3d567c64d3d5d712c0af6c2cacb1e1bcc1a2b)
1547-5905/asset/AIC_right.gif?v=1&s=518efadeedca9ceeef271499f690fdebd2ed9164)
Process Systems Engineering
Identification of nonlinear parameter varying systems with missing output data
Article first published online: 12 MAR 2012
DOI: 10.1002/aic.13735
Copyright © 2012 American Institute of Chemical Engineers (AIChE)
Additional Information
How to Cite
Deng, J. and Huang, B. (2012), Identification of nonlinear parameter varying systems with missing output data. AIChE J., 58: 3454–3467. doi: 10.1002/aic.13735
Publication History
- Issue published online: 5 OCT 2012
- Article first published online: 12 MAR 2012
- Accepted manuscript online: 12 JAN 2012 11:43AM EST
- Manuscript Revised: 22 DEC 2011
- Manuscript Received: 10 APR 2011
Funded by
- Natural Sciences and Engineering Research Council of Canada
- Abstract
- Article
- References
- Cited By
Keywords:
- parameter identification;
- EM algorithm;
- missing data;
- particle filter;
- multiple model
Abstract
An identification of nonlinear parameter varying systems using particle filter under the framework of the expectation-maximizaiton (EM) algorithm is described. In chemical industries, processes are often designed to perform tasks under various operating conditions. To circumvent the modeling difficulties rendered by multiple operating conditions and the transitions between different working points, the EM algorithm, which iteratively increases the likelihood function, is applied. Meanwhile the missing output data problem which is common in real industry is also considered in this work. Particle filters are adopted to deal with the computation of expectation functions. The efficiency of the proposed method is illustrated through simulated examples and a pilot-scale experiment. © 2012 American Institute of Chemical Engineers AIChE J, 2012