Neural networks for process control in steel manufacturing

Schlag, M.; Broese, Einar; Granckow, O.; Jansen, M.; Pappe, T.; Schäffner, Clemens; Sörgel, Günter

doi:10.1109/icassp.1997.599582

Cited by 3 publications

(2 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Neural network is another one of most popular algorithms used in steel industry, which has stronger nonlinearity and capability of adaptive information processing. Some researchers have successfully adopted ANNs to predict heat transfer coefficient, temperature or water flow [25][26][27]. Valentina introduced an ANN to find correlations between model parameters and process variables [28].…”

Section: Introductionmentioning

confidence: 99%

Temperature distribution prediction in control cooling process with recurrent neural network for variable-velocity hot rolling strips

Chen

Zhang

et al. 2022

Int J Adv Manuf Technol

View full text Add to dashboard Cite

Control cooling is essential method for microstructure and mechanical properties control in hot rolling strip making. It is vital to realize high precises temperature distribution prediction and control in cooling process to ensure the industrial production. In this paper, a traditional mechanism model based on finite-difference method and combining with online cycle velocity calculation strategy was introduced as one of estimating temperature distribution baseline method. However, considering calculation time, variable-velocity rolling makes it difficult to rapidly realize temperature and water distribution modifying of each segment in cooling zone. Herein, a temperature distribution prediction method based on recurrent neural network was proposed, instead of only final cooling temperature prediction. And temperature distribution prediction performance of model with different recurrent cell and time steps were evaluated.The results indicated that the proposed model could realize temperature distribution prediction and the model based on bi-LSTM and 48 timesteps has the highest determination coefficient value of 0.976 and lowest root mean square error of 8.03 and mean absolute error of 5.7. Furthermore, compared with baseline model, the proposed model retained lower computational cost, making it applicable in industrial application by providing real-time temperature distribution prediction.

show abstract

Section: Introductionmentioning

confidence: 99%

Temperature distribution prediction in control cooling process with recurrent neural network for variable-velocity hot rolling strips

Chen

Zhang

et al. 2022

Int J Adv Manuf Technol

View full text Add to dashboard Cite

show abstract

“…Starting with a wrong set of parameters, it can be tedious work to optimize the performance of the SSC. Usually, this procedure relies on expert knowledge, heuristic methods, and methods that are based on neural networks . There are only a few approaches that compute SSC curves based on a finite element simulation model of the rolling process .…”

Section: Introductionmentioning

confidence: 99%

Model‐Based Optimization of Short Stroke Control in Roughing Mills

Werner

Kurz²,

Stingl

et al. 2017

steel research int.

View full text Add to dashboard Cite

Automatic width control is one key issue in hot strip rolling and the main actuator for width control is the vertical draft (edging) applied to slabs in the roughing sequence. In order to compensate for the occurring strip necking, a variable edger gap is applied in the edging process. This paper introduces a comprehensive mathematical optimization approach for the fully automatic generation of the associated edger roll trajectories, which is based on a nonlinear dynamic simulation model for the material flow in the roughing mill. Due to the complexity of this simulation model, gradient-based methods are applied to solve the resulting optimization problem in reasonable time. Hence, derivatives of cost functional and constraints must be provided, which entails computing sensitivity information from the underlying simulation model. Finally, numerical results for real-world examples are discussed to illustrate the capability of the presented approach. In particular, this paper examines the influence of characteristic parameters such as the slab width or the maximum edger screw in velocity onto the optimization results.[ Ã ] Dr.

show abstract

Temperature Distribution Prediction in Control Cooling Process with Recurrent Neural Network for Variable-Velocity Hot Rolling Strips

Chen

Zhang

et al. 2022

Preprint

View full text Add to dashboard Cite

Control cooling is essential method for microstructure and mechanical properties control in hot rolling strip making. It is vital to realize high precises temperature distribution prediction and control in cooling process to ensure the industrial production. In this paper, a traditional mechanism model based on finite-difference method and combining with online cycle velocity calculation strategy was introduced as one of estimating temperature distribution baseline method. However, considering calculation time, variable-velocity rolling makes it difficult to rapidly realize temperature and water distribution modifying of each segment in cooling zone. Herein, a temperature distribution prediction method based on recurrent neural network was proposed, instead of only final cooling temperature prediction. And temperature distribution prediction performance of model with different recurrent cell and time steps were evaluated. The results indicated that the proposed model could realize temperature distribution prediction and the model based on bi-LSTM and 48 timesteps has the highest determination coefficient value of 0.976 and lowest root mean square error of 8.03 and mean absolute error of 5.7. Furthermore, compared with baseline model, the proposed model retained lower computational cost, making it applicable in industrial application by providing real-time temperature distribution prediction.

show abstract

Neural networks for process control in steel manufacturing

Cited by 3 publications

References 3 publications

Temperature distribution prediction in control cooling process with recurrent neural network for variable-velocity hot rolling strips

Temperature distribution prediction in control cooling process with recurrent neural network for variable-velocity hot rolling strips

Model‐Based Optimization of Short Stroke Control in Roughing Mills

Temperature Distribution Prediction in Control Cooling Process with Recurrent Neural Network for Variable-Velocity Hot Rolling Strips

Contact Info

Product

Resources

About