Development of neural network-based models to predict mechanical properties of hot dip galvanised steel coils

In the industrial arena, artificial neural networks are among the most significant techniques in system modelling because of their efficiency and simplicity. In this paper, we present an application of artificial neural networks, along with other techniques stemming from data mining, to model the yield strength, tensile strength, elongation, strain hardening coefficient and the Lankford’s anisotropy coefficient of galvanised steel coils, according to the manufacturing process data. In particular, we propose the use of these models to improve the current control systems of hot-dip galvanising lines since an open loop control strategy must be adopted because the mechanical properties of hot-dip galvanising coils are not directly measurable.

[1]  Ken-ichi Funahashi,et al.  On the approximate realization of continuous mappings by neural networks , 1989, Neural Networks.

[2]  Weiru Chen,et al.  Discovering exclusive patterns in frequent sequences , 2010, Int. J. Data Min. Model. Manag..

[3]  G. McLachlan.  Discriminant Analysis and Statistical Pattern Recognition , 1992 .

[4]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[5]  Joaquín B. Ordieres Meré,et al.  Estimation of mechanical properties of steel strip in hot dip galvanising lines , 2004 .

[6]  Wang Dan-min,et al.  Quality prediction model and control of the metallic mechanical properties , 2007 .

[7]  Subhas Ganguly,et al.  Designing High Strength Multi-phase Steel for Improved Strength-Ductility Balance Using Neural Networks and Multi-objective Genetic Algorithms , 2007 .

[8]  Richard Bellman,et al.  Adaptive Control Processes: A Guided Tour , 1961, The Mathematical Gazette.

[9]  Zhang Pi-jun,et al.  Software of Predicting Mechanical Properties of Strip Steel by Using BP Networks , 2007 .

[10]  Ana González-Marcos,et al.  Steel annealing furnace robust neural network model , 2005 .

[11]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[12]  Ill-Soo Kim,et al.  A study on genetic algorithm to select architecture of a optimal neural network in the hot rolling process , 2004 .

[13]  Michael K. Ng,et al.  SMART: a subspace clustering algorithm that automatically identifies the appropriate number of clusters , 2009, Int. J. Data Min. Model. Manag..

[14]  J. Watton,et al.  Prediction and validation of through coil final mechanical properties of high strength hot rolled coil using artificial neural networks , 2006 .

[15]  John Norrish,et al.  Artificial neural networks for modelling the mechanical properties of steels in various applications , 2005 .

[16]  Manuel Castejón Limas,et al.  Outlier Detection and Data Cleaning in Multivariate Non-Normal Samples: The PAELLA Algorithm , 2004, Data Mining and Knowledge Discovery.

[17]  Yu Hui,et al.  Diagonal Recurrent Neural Network and Its Application in Mechanical Property Prediction of 82B Steel , 2007 .

[18]  Dilip Kumar Pratihar,et al.  Modelling of electrical discharge machining process using regression analysis, adaptive neuro-fuzzy inference system and genetic algorithm , 2010, Int. J. Data Min. Model. Manag..