Prediction and Control of Rolling Quality by Just-in-time Method  

Oleh:

Zheng, Qiubao ; Kimura, Hidenori

Jenis: Article from Article
Dalam koleksi: Final Program and Book of Abstracts: The 4th Asian Control Conference, September 25-27, 2002 (Sep. 2002), page 922-927.
Topik: Prediction; Control Rolling; Just in Time; QBN
Fulltext: AC021596.PDF (234.29KB)

Isi artikel

This paper introduces a variant of Just-in-time methods, called “k Bipartite Neighbors with Quasi-nonlinear local models (k-QBN)”, for use in function prediction. In order to deal with complex nonlinear systems well and keep local models as simple as possible, k-QBN transforms the chosen neighbors with pre-chosen nonlinear functions first, and then uses linear local models to model the transformed complex nonlinear plants. Applied to several real measured datasets about hot plate rolling, k-QBN outperforms typical k-BN and other comparative alternatives.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali