Search

Camber in hot-rolled plates significantly impacts product quality and rolling process stability, making accurate camber prediction crucial. However, it is challenging to measure asymmetric factors impacting camber in real production, hindering the ability of current models to predict and analyze the overall camber of rolled plates. This study proposes a method that combines asymmetric fluctuations of measurable variables with deep learning to predict camber. We developed a data analysis platform for plate processing and constructed a camber dataset using machine vision technology. During the modeling phase, a hot-rolled plate camber prediction model, BiLSTM-AM-Seq2Seq, was proposed, integrating a bidirectional long short-term memory network, an attention mechanism, and a sequence-to-sequence model. Additionally, an improved scheduled sampling method was also introduced to steer model training. Model performance was evaluated with real-world production data, demonstrating superior accuracy and stability. Specifically, the model achieved a mean absolute error of 12.29 mm and a root mean square error of 17.32 mm. Consequently, this model meets the demands of practical production and addresses the need for overall camber prediction of hot rolled plates. • Data-driven prediction model for overall camber shape of rolled plates is proposed. • Modeling using fluctuations of observable variables to address the absence of key asymmetric variables in real production. • Data analysis platform was developed and real data were used to demonstrate the superiority of the proposed model. [ABSTRACT FROM AUTHOR]

In the field of hot rolling big data, the presence of different steel types, specifications, and data heterogeneity poses significant challenges to the accuracy and stability of using single machine learning regression technology for prediction. Therefore, this study proposes a hot‐rolled strip crown prediction method that combines data clustering and fusion modeling. First, this article introduces a relevant mechanism for designing cluster strategies. The optimal clustering strategy is determined through comparative experiments using rolling process parameters, strip size, and main material components as the clustering features. Subsequently, the K‐Means++ algorithm is used to effectively cluster the training and testing datasets based on this strategy, generating multiple clusters for both datasets. Finally, this study establishes seven different training models to match the most suitable regression prediction model for each cluster, and matching between each cluster and the model is determined through rigorous testing. The evaluation of the fusion model shows an R2 value of 0.829 and a root mean square error value of 3.974. The experimental results show that the proposed method outperforms traditional methods in solving the challenges of multiclass classification and data heterogeneity, providing strong data support for the intelligent control of the hot‐rolled strip crown in the future. [ABSTRACT FROM AUTHOR]

To obtain the transverse temperature distribution of running strips under C‐type edge induction heaters during the hot‐rolling process, a coupled electromagnetic–thermal finite‐element model is established using COMSOL, in which the electromagnetic and temperature distributions regularity with various process parameters are systematically obtained. It is shown in the results that the temperature increase is directly proportional to the frequency and current and inversely proportional to the thickness. Meanwhile, the edge temperature drop can be rapidly compensated for using an edge heater, but it also leads to the appearance of a "cat‐ear" high‐temperature zone at the edge; thus, the transverse temperature uniformity cannot be guaranteed. To improve the temperature uniformity of the running strip, a magnetic shield plate is designed to weaken the effect of the cat ear, and the effects of the size and installation position of the shielding plate on the temperature distribution are analyzed. The standard deviation and uniformity of the temperature are used to determine the uniformity of the transverse temperature, and the optimal parameter combination mode of the magnetic shielding amounts is obtained, which can provide a reference for optimizing the heating process at the edge of the strip under moving conditions. [ABSTRACT FROM AUTHOR]

A dog‐bone profile is developed by inhomogeneous deformation of a slab, whose parameters can be used to describe the metal flow rule during the width reduction process by a sizing press. The accuracy of a dog‐bone profile is essential to ensure the high width precision of the roughing process. However, the existing equipment cannot satisfy the requirements of online measurement. To solve this problem, herein, a prediction model for dog‐bone parameters based on the finite‐element simulation and a neural network is presented. The finite‐element simulation can reproduce the field production process, revealing variations in the dog‐bone profile parameters. In addition, based on the simulation data, a neural network model is constructed to predict the dog‐profile parameters. By adjusting the hyperparameters, the prediction accuracy of the constructed deep neural network model with two hidden layers is improved. For the dog‐bone profile parameters, the mean squared error is less than 3 mm, the mean absolute percentage error is less than 2%, the maximum absolute percentage error is less than 9%, and the root mean squared error is less than 4 mm. Thus, an accurate prediction of the dog‐bone profile parameters is achieved, which can guide subsequent high‐precision width control. [ABSTRACT FROM AUTHOR]

Purpose: In the cold rolling process, friction coefficient, oil film thickness and other factors vary dramatically with the change in the rolling speed, which seriously affects the strip thickness deviation. This paper aims to improve the strip control precision with the forecast roll gap model based on CF-PSO-SVM approach in the rolling process. Design/methodology/approach: In this paper, a novel forecasting model of the roll gap based on support vector machine (SVM) optimized by particle swarm optimization with compression factor (CF-PSO) is proposed. Based on lots of online data, the roll gap models regressed by PSO-SVM, genetic algorithm (GA)-SVM and CF-PSO-SVM are obtained and verified through evaluating the performances with the decision coefficient (R2), mean absolute error and root mean square error. In addition, with the good forecasting performances of CF-PSO-SVM, a roll gap compensation model is studied. Findings: The results indicate that the proposed CF-PSO-SVM has excellent learning regression ability compared with other optimization algorithms. Meanwhile, a roll gap compensation model based on the rolling speed and plastic coefficient is obtained, which has been proved validated in product. Originality/value: In this paper, the SVM algorithm is combined with traditional rolling technology to solve the problems in actual production, which has great supporting significance for the improvement of production efficiency. [ABSTRACT FROM AUTHOR]

Strip shape prediction is one of the most important technical to improve the quality of products in hot strip rolling process. In this paper, three hybrid models, including GA-MLP, MEA-MLP and PCA-MEA-MLP, are proposed for profile and flatness predictions by combining genetic algorithm (GA), mind evolutionary algorithm (MEA), principal component analysis (PCA) and multi-layer perceptron (MLP) neural networks. Mean absolute error (MAE), mean absolute percentage error, root mean squared error are adapted to evaluate the performance of the models. The results show that the data-driven model based on intelligent algorithm optimization neural networks can achieve good prediction of profile and flatness. Comparing with the hybrid GA-MLP model, the training speed of the hybrid MEA-MLP model is faster and the training time is greatly reduced. The model establishing with the input data after dimensionality reduction by PCA can reduce training time and become simple. The innovation of this paper is to propose a data-driven fast response model based on intelligent algorithm optimization neural network to replace the traditional mechanism model based on mathematical formula analysis to study complex, non-linear strip shape control in hot rolling process. [ABSTRACT FROM AUTHOR]

Abstract In this paper, a prediction model for hot-rolled strip crown based on an artificial neural network (ANN) is presented. Considering the mean squared error (MSE) and correlation coefficient (R), production data of 10,133 coils collected from a hot rolling plant are used to establish models. A feed-forward neural network with one hidden layer, a non-dominated sorting genetic algorithm II (NSGA II) optimized ANN, and a deep neural network (DNN) are applied to evaluate the prediction performance. Parameter settings of the ANN and NSGA II including learning rate, hidden neurons, activation function and population size, and crossover probability, are investigated to acquire the optimal model. The structure of the DNN, including how many layers and units the network should contain, is also studied. Prediction performance comparisons of the ANN, NSGA II-ANN, and DNN are presented. Among the ANN, NSGA II-ANN and DNN, the results show that the DNN has the highest prediction accuracy. Root mean squared error (RMSE) of the proposed DNN is 2. 06 μ m , and 97.04% of the prediction data have an absolute error of less than 5 μ m. Through model response surfaces, effects of four key operating parameters are investigated. The results indicate that the proposed DNN with strong learning ability and generalization performance can be well applied to hot rolling production. Highlights • Three neural network based models are established based on a dataset from a hot-rolling line. • The proposed models used to predict strip crown can acquire high prediction precisions. • The best performed model precisely corresponds with the existing physical understandings. • Using the parameters obtained by the best performed model, strip crown is close to target values. [ABSTRACT FROM AUTHOR]