Rolling Force Prediction Algorithm Based on Bayesian Regularization Neural Network

DOI

10.2991/icence-16.2016.146How to use a DOI?

Keywords

Hot continuous rolling, Rolling force prediction, Neural network, Bayesian regularization.

Abstract

For obtaining relative accurate rolling-mill model is difficulty by the simple mathematical method, due to the complexity of the actual production scene and the non-linear relationship between variables, this paper firstly proposes an improved Bayesian regularization neural network model according to these measured data of 1580 production line. In this model, the paper constructs the improved Bayesian neural networks by the introduction of bound terms that represents the network complexity in the objective function. At last, the simulation result proves the effectiveness and validity of the model and the prediction accuracy of the model algorithm is superior to the traditional model.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xiaodan Zhang
AU  - Lu Yao
AU  - Zhenxiong Zhou
PY  - 2016/09
DA  - 2016/09
TI  - Rolling Force Prediction Algorithm Based on Bayesian Regularization Neural Network
BT  - Proceedings of the 2nd International Conference on Electronics, Network and Computer Engineering (ICENCE 2016)
PB  - Atlantis Press
SP  - 790
EP  - 795
SN  - 2352-538X
UR  - https://doi.org/10.2991/icence-16.2016.146
DO  - 10.2991/icence-16.2016.146
ID  - Zhang2016/09
ER  -