Improved Lazy Learning Based Dynamic Modeling Method on Combustion Process

Abstract:

This paper presents a new data-based method for the industrial combustion process. In this method, angle measure, which presents change trend of samples, is introduced in evaluating the similarity between the sample data and query data, which is not exploited in the previous word. And ARX method is used to build the local model. With the moving of working points, different models are set up to realize the accurate modeling for combustion process. An example of coke oven combustion process is presented to illustrate the modeling capability of the proposed method.

• Periodical: Applied Mechanics and Materials (Volumes 52-54)
• Pages: 680-685
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.52-54.680
• Online since: March 2011
• Authors: Qi Lei, Min Wu, Chun Sheng Wang
• Keywords: ARX, Combustion Process, Lazy Learning, Local Model

References

[1] B. C. Buss and H. R. Mccollum. Feedforward heating control for coke batteries. Ironmaking Conf. Proc., Vol. 67(1984), p.205–208.

[2] D. Wild, T. Kiefer, A. Kugi. A mathematical model of a slab reheating furnace with radioactive heat transfer and non-participating gaseous media. International Journal of Heat and Mass Transfer, Vol. 5(2010), pp.5933-5946.

DOI: 10.1016/j.ijheatmasstransfer.2010.07.029

[3] O. Nelles. Nonlinear System Identification. (Springer, Berlin 2001).

[4] W. Pedrycz, A. Amato, V, D, Lecce, V. Piuri. Fuzzy clustering with partial supervision in organization and classification of digital Images. IEEE Transactions on Fuzzy Systems, Vol. 16(2008), pp.1008-1016.

DOI: 10.1109/tfuzz.2008.917287

[5] M. E. Gegúndez, J. Aroba, J.M. Bravo. Identification of piecewise affine systems by means of fuzzy clustering and competitive learning. Engineering Applications of Artificial Intelligence, Vol. 21(2008), pp.1321-1329.

DOI: 10.1016/j.engappai.2008.03.014

[6] A. Lj. Juloski, W. P. M. H. Heemels, G. Ferrari-Trecate. Data-based hybrid modeling of the component placement process in pick-and-place machines. Control Engineering Practice, Vol. 12(2004), pp.1241-1252.

DOI: 10.1016/j.conengprac.2004.04.004

[7] A. Bemporad, A. Garulli, S. Paoletti, A. Vicino. A bounded-error approach to piecewise affine system identification. IEEE Transcations on Automatic, Vol. 50(2005), pp.1567-1580.

DOI: 10.1109/tac.2005.856667

[8] E. Bertolissi, M. Birattari, G. Bontempi, A. Duchateau, H. Bersini. Data-driven techniques for direct adaptive control: the lazy and the fuzzy approaches. Fuzzy Sets and Systems, Vol. 128(2002), pp.3-14.

DOI: 10.1016/s0165-0114(01)00179-8

[9] A. Lj. Juloski, W. P. M. H. Heemels, G. Ferrari-Trecate. Data-based hybrid modeling of the component placement process in pick-and-place machines. Control Engineering Practice, Vol. 12(2004), pp.1241-1252.

DOI: 10.1016/j.conengprac.2004.04.004

[10] Bertolissi E, Birattari M, Bontempi G, Duchateau A, Bersini H. Data-driven techniques for direct adaptive control: the lazy and the fuzzy approaches. Fuzzy Sets and Systems, Vol. 128(2002), pp.3-14.

DOI: 10.1016/s0165-0114(01)00179-8

[11] C. G. Atkeson, A. W. Moore, S. Schaal. Locally weightedlearning. Artificial Intelligence Review. Vol. 11(1997), p.11–73.

[12] T. H. Pan, S. Y. Li, X. Wang. A Multi-model modeling approach to nonlinear systems based on lazy learning. Proceedings of the 24th Chinese Control Oonference. (2005), pp.268-274.