Long-term prediction of industrial furnace by Extended Sequential Prediction method of LOM

Masatoshi Ogawa*, Yichun Yeh, Syou Kawanari, Harutoshi Ogai

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)


Recently, attention has been drawn by the local modeling techniques of a new idea called "Just-In-Time (JIT) modeling" or "Lazy Learning". To apply "JIT modeling" to a large amount of database online, "Large-scale database-based Online Modeling (LOM)" has been proposed. LOM is such a technique that makes the retrieval of "neighboring" data more efficient by using "stepwise selection" and quantization. This paper reports an Extended Sequential Prediction (ESP) method of LOM with the local regression model. The ESP method is able to predict process variables over a long period by modeling the operator and the plant based on LOM, the approach is to repeat a process that predicts the process variables of the next step by using the predicted variables of the previous step. The method is applied to a dynamic industrial furnace with several deeply-intertwined physical phenomena; practical effectiveness of the method is verified. As a result, the method has predicted the process variables with satisfactory accuracy.

Original languageEnglish
Title of host publicationProceedings of SICE Annual Conference 2010, SICE 2010 - Final Program and Papers
PublisherSociety of Instrument and Control Engineers (SICE)
Number of pages4
ISBN (Print)9784907764364
Publication statusPublished - 2010 Jan 1

Publication series

NameProceedings of the SICE Annual Conference


  • ESP
  • Industrial furnace
  • JIT modeling
  • LOM
  • Operation support
  • Sequential Prediction

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Computer Science Applications
  • Electrical and Electronic Engineering


Dive into the research topics of 'Long-term prediction of industrial furnace by Extended Sequential Prediction method of LOM'. Together they form a unique fingerprint.

Cite this