Please use this identifier to cite or link to this item:
Title: A New Strategy of Locality Enhancement for Justin-Time Learning Method
Authors: Su, Q.L.
Kano, M.
Chiu, M.-S. 
Keywords: Data-Based
Just-in-Time Learning
Local Model
Locality Enhancement
Issue Date: 2012
Citation: Su, Q.L.,Kano, M.,Chiu, M.-S. (2012). A New Strategy of Locality Enhancement for Justin-Time Learning Method. Computer Aided Chemical Engineering 31 : 1662-1666. ScholarBank@NUS Repository.
Abstract: Just-in-Time Learning (JITL) method has recently received increasing attention, particularly its application to control and soft sensing. Unlike the conventional JITL methods, which construct a local model directly based on the relevant data selected from reference database, a novel strategy is proposed by considering the local model as a Taylor series of the global model expanded in the vicinity of a reference point. This reference point could be chosen as the most relevant data or the query data. A comparative study using a benchmark nonlinear CSTR process showed the efficiency of the proposed strategy by achieving better prediction than its conventional counterparts. © 2012 Elsevier B.V.
Source Title: Computer Aided Chemical Engineering
ISSN: 15707946
DOI: 10.1016/B978-0-444-59506-5.50163-2
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.


checked on Jan 29, 2023

Page view(s)

checked on Feb 2, 2023

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.