In this paper we consider an application of data mining technology to the analysis of time series data from a pilot circulating fluidized bed (CFB) reactor. We focus on the problem of the online mass prediction in CFB boilers. We present a framework based on switching regression models depending on perceived changes in the data. We analyze three alternatives for change detection. Additionally, a noise canceling and a state determination and windowing mechanisms are used for improving the robustness of online prediction. We validate our ideas on real data collected from the pilot CFB boiler.
Description
Handling outliers and concept drift in online mass flow prediction in CFB boilers
%0 Conference Paper
%1 Bakker2009
%A Bakker, J.
%A Pechenizkiy, M.
%A Zliobait\.e, I.
%A Ivannikov, A.
%A Kärkkäinen, T.
%B Proceedings of the Third International Workshop on Knowledge Discovery from Sensor Data
%C New York, NY, USA
%D 2009
%I ACM
%K boiler cfb datamining massflow prediction
%P 13--22
%R 10.1145/1601966.1601972
%T Handling outliers and concept drift in online mass flow prediction in CFB boilers
%U http://doi.acm.org/10.1145/1601966.1601972
%X In this paper we consider an application of data mining technology to the analysis of time series data from a pilot circulating fluidized bed (CFB) reactor. We focus on the problem of the online mass prediction in CFB boilers. We present a framework based on switching regression models depending on perceived changes in the data. We analyze three alternatives for change detection. Additionally, a noise canceling and a state determination and windowing mechanisms are used for improving the robustness of online prediction. We validate our ideas on real data collected from the pilot CFB boiler.
%@ 978-1-60558-668-7
@inproceedings{Bakker2009,
abstract = {In this paper we consider an application of data mining technology to the analysis of time series data from a pilot circulating fluidized bed (CFB) reactor. We focus on the problem of the online mass prediction in CFB boilers. We present a framework based on switching regression models depending on perceived changes in the data. We analyze three alternatives for change detection. Additionally, a noise canceling and a state determination and windowing mechanisms are used for improving the robustness of online prediction. We validate our ideas on real data collected from the pilot CFB boiler.},
acmid = {1601972},
added-at = {2011-03-22T10:39:41.000+0100},
address = {New York, NY, USA},
author = {Bakker, J. and Pechenizkiy, M. and \v{Z}liobait\.{e}, I. and Ivannikov, A. and K\"{a}rkk\"{a}inen, T.},
biburl = {https://www.bibsonomy.org/bibtex/2462fd54527e90e303834050b8ad0d98e/pejowart},
booktitle = {Proceedings of the Third International Workshop on Knowledge Discovery from Sensor Data},
description = {Handling outliers and concept drift in online mass flow prediction in CFB boilers},
doi = {10.1145/1601966.1601972},
groups = {public},
interhash = {f703d42fca624b5a9ccad1a5bc92f678},
intrahash = {462fd54527e90e303834050b8ad0d98e},
isbn = {978-1-60558-668-7},
keywords = {boiler cfb datamining massflow prediction},
location = {Paris, France},
numpages = {10},
pages = {13--22},
publisher = {ACM},
series = {SensorKDD '09},
timestamp = {2012-06-20T10:05:24.000+0200},
title = {Handling outliers and concept drift in online mass flow prediction in CFB boilers},
url = {http://doi.acm.org/10.1145/1601966.1601972},
username = {pejowart},
year = 2009
}