Current and future development in neural computation in steel processing

Schlang, Martin; Lang, Bernhard; Poppe, Thomas; Runkler, Thomas A.; Weinzierl, К.

doi:10.1016/s0967-0661(01)00086-7

Cited by 41 publications

(27 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This is due primarily to the fact that these processes and sub-processes are repetitive, highly automated, and have a large number of well-known variables that define them [6][7][8][9][10][11][12] .…”

Section: Control Of the Annealing Process For The Steel Stripmentioning

confidence: 99%

Overall model of the dynamic behaviour of the steel strip in an annealing heating furnace on a hot-dip galvanizing line

Martínez-de-Pisón¹,

Pernía²,

Jiménez-Macías³

et al. 2010

REVMETAL

View full text Add to dashboard Cite

Predicting the temperature of the steel strip in the annealing process in a hot-dip galvanizing line (HDGL) is important to ensure the physical properties of the processed material. The development of an accurate model that is capable of predicting the temperature the strip will reach according to the furnace's variations in temperature and speed, its dimensions and the steel's chemical properties, is a requirement that is being increasingly called for by industrial plants of this nature. This paper presents a comparative study made between several types of algorithms of Data Mining and Artificial Intelligence for the design of an efficient and overall prediction model that will allow determining the strip's variation in temperature according to the physico-chemical specifications of the coils to be processed, and fluctuations in temperature and speed that are recorded within the annealing process. The ultimate goal is to find a model that is effectively applicable to coils of new types of steel or sizes that are being processed for the first time. This model renders it possible to fine-tune the control model in order to standardise the treatment in areas of the strip in which there is a transition between coils of different sizes or types of steel. KeywordsHot-dip galvanizing line; Data mining; Artificial intelligence; Modelling.Modelo global del comportamiento dinámico de la banda de acero en un horno de recocido de una línea de galvanizado Resumen La predicción de la temperatura de la banda de acero dentro del proceso de recocido de una planta de galvanizado continuo en caliente es importante para garantizar las propiedades físicas del material procesado. El desarrollo de un modelo preciso que sea capaz de predecir la temperatura que va a alcanzar la banda según las variaciones de temperaturas y velocidades del horno, y sus dimensiones y propiedades químicas del acero, es una necesidad cada vez más demandada por este tipo de plantas industriales. En el presente estudio se muestra una comparativa realizada entre diversos tipos de algoritmos de Minería de Datos e Inteligencia Artificial para el desarrollo de un modelo de predicción eficiente y global que permita determinar la variación de temperatura de la banda según las características físico-químicas de las bobinas a procesar y las fluctuaciones de temperaturas y velocidades que aparezcan dentro del proceso de recocido. El objetivo final es la búsqueda de un modelo que sea eficiente ante bobinas con nuevos tipos de acero o dimensiones que no hayan sido procesadas anteriormente. Con este modelo es posible optimizar los modelos de control para poder conseguir homogeneizar el tratamiento en zonas de la banda donde existe la transición entre bobinas con diferentes dimensiones o tipos de acero. Palabras claveLínea de galvanizado continuo en caliente; Minería de datos; Inteligencia artificial; Modelado.

show abstract

Section: Control Of the Annealing Process For The Steel Stripmentioning

confidence: 99%

Overall model of the dynamic behaviour of the steel strip in an annealing heating furnace on a hot-dip galvanizing line

Martínez-de-Pisón¹,

Pernía²,

Jiménez-Macías³

et al. 2010

REVMETAL

View full text Add to dashboard Cite

show abstract

“…The proposed approach, which couples NNs and a physical model, gives rise to a sort of so-called ''grey-box'', ''hybrid'' or ''semiphysical'' models, which have already been applied in the field of materials processing [18] and in particular for hot rolling of steel [19,20]. However, while in [19] NNs have been applied mostly to add a correction factor which improves the performance of physical/analytical models, our approach is more similar to one of the schemes proposed in [20] for scale breaker entry temperature prediction in a hot strip mill, where an NN has been applied to predict the optimal value of the coefficient of the fourth-order term in the heat transfer physical model that is exploited to predict the desired temperature value. In the present case not a single temperature value, but a whole cooling curve is predicted by correlating the values of the parameters A and k in Eq.…”

Section: Neural Network-based Prediction Of the Cooling Behaviourmentioning

confidence: 99%

“…11, by comparing the blue histogram and the red dashed line. However, the purpose here is not to find the function that best fits the actual error distribution, but to find a statistical representation of the error random variable which is accurate enough to distinguish standard and abnormal operating conditions, also according to the idea of ''model-based diagnosis'' that is presented in [19]. The Rayleigh distribution is adequate for this purpose and, thanks to its quite simple expression, it allows one to set up a fast iterative procedure (the SPRT), where a sequence of subsequent measurements of the error are jointly evaluated in order to decide if they all come from a normal or anomalous conditions.…”

Section: Statistical Control Of Cooling Stabilitymentioning

confidence: 99%

Diagnosis of the instability of the cooling behaviour of flat steel products through parametric characterisation, neural networks and statistics

2010

View full text Add to dashboard Cite

“…Because of the mentioned difficulties that are faced when smelting gold slime, we have decided to search for a new model to predict gold content in the slag in addition to the traditional nonlinear regression. On the other hand, the neural network has proven to be a powerful tool in many areas including industrial processes (Schlang, Lang, Poppe, Runkler, & Weinzierl, 2001), prediction of materials properties such as steel (Bahrami, Mousavi Anijdan, & Ekrami, 2005;Capdevila, Garcia-Mateo, Caballero, & Garcl´a de Andre's, 2006;Guo & Sha, 2004). In addition, there are many other reports that the neural network approach has used in material science-based research (Sha & Edwards, 2007).…”

Section: Introductionmentioning

confidence: 99%

Artificial neural network vs. nonlinear regression for gold content estimation in pyrometallurgy

Liu

Yuan

Liao

2009

Expert Systems with Applications

View full text Add to dashboard Cite

Current and future development in neural computation in steel processing

Cited by 41 publications

References 6 publications

Overall model of the dynamic behaviour of the steel strip in an annealing heating furnace on a hot-dip galvanizing line

Overall model of the dynamic behaviour of the steel strip in an annealing heating furnace on a hot-dip galvanizing line

Diagnosis of the instability of the cooling behaviour of flat steel products through parametric characterisation, neural networks and statistics

Artificial neural network vs. nonlinear regression for gold content estimation in pyrometallurgy

Contact Info

Product

Resources

About