The role of neural networks in the optimisation of rolling processes

Abstract Applications of neural networks in the rolling of steel are reviewed. The first papers on the topic were published in 1991 and since then the number of publications has steadily increased. In most applications today, so-called back propagation networks are used. After briefly reviewing the various neural network types, the results of two case studies at Rautaruukki cold strip mill are presented. In the first case an efficiency model for tandem cold rolling was developed. By using the model it is possible to study whether a new product with a given width, strength or thickness can be produced, and the optimised mill settings can then be determined. A 1.8% improvement in efficiency was obtained with the model. The second case concerns the prediction of the mechanical properties of steel strips and temper rolling force by using neural network modelling and measured process data. The location of the coils in annealing stacks and their vanadium content were found to explain the deviation in mechanical properties. The temper rolling force could be predicted with good accuracy, which can be exploited in determining mill pre-settings.

[1]  David M. Skapura,et al.  Neural networks - algorithms, applications, and programming techniques , 1991, Computation and neural systems series.

[2]  Yong-Taek Im,et al.  Fuzzy-control simulation of cross-sectional shape in six-high cold-rolling mills , 1996 .

[3]  Teuvo Kohonen,et al.  Self-Organizing Maps , 2010 .

[4]  Yasuo Morooka.  Shape control of rolling mills by a neural and fuzzy hybrid architecture , 1995, Proceedings of 1995 IEEE International Conference on Fuzzy Systems..

[5]  R. Pradhan,et al.  Developments in the annealing of sheet steels , 1992 .

[6]  Dorel Aiordachioaie,et al.  Pre-processing of acoustic signals by neural networks for fault detection and diagnosis of rolling mill , 1997 .

[7]  J. Larkiola,et al.  Development of prediction model for mechanical properties of batch annealed thin steel strip by using artificial neural network modelling , 1996 .

[8]  A. Fouarge,et al.  Modélisation et pilotage industriel optimisé du recuit base , 1995 .

[9]  Sungzoon Cho,et al.  Reliable roll force prediction in cold mill using multiple neural networks , 1997, IEEE Trans. Neural Networks.

[10]  Yasunori Katayama,et al.  Fuzzy control algorithm and neural networks for flatness control of a cold rolling process , 1992 .

[11]  Olli Simula,et al.  The SOM Based Data Mining in Hot Rolling , 1998 .

[12]  Amir F. Atiya,et al.  Application of the recurrent multilayer perceptron in modeling complex process dynamics , 1994, IEEE Trans. Neural Networks.

[13]  Prasad K. Yarlagadda,et al.  Neural network approach to flow stress evaluation in hot deformation , 1995 .

[14]  Antti Korhonen,et al.  Prediction of rolling force in cold rolling by using physical models and neural computing , 1996 .

[15]  Zhenyu Liu,et al.  Prediction of the mechanical properties of hot-rolled CMn steels using artificial neural networks , 1996 .

[16]  Yong-Taek Im,et al.  Development of fuzzy control algorithm for shape control in cold rolling , 1995 .

[17]  Andreas Kugi,et al.  Neural network for identification of roll eccentricity in rolling mills , 1996 .