Mechanical Properties Of Hot Rolled Strip Research Articles

Prediction model of mechanical properties of hot-rolled strip based on improved feature selection method

Prediction model of mechanical properties of hot-rolled strip based on improved feature selection method

Deep learning for predictive mechanical properties of hot-rolled strip in complex manufacturing systems

Deep learning for predictive mechanical properties of hot-rolled strip in complex manufacturing systems

An intelligent online detection approach based on big data for mechanical properties of hot-rolled strip

An LightGBM prediction model based on big data is presented in order to online detect the mechanical properties of hot-rolled strip in this paper, which can achieve the greater accuracy than both the existing prediction approaches and hardware detection method for the local strips. A dataset of mechanical properties of hot-rolled strip is constructed firstly by collecting a steel plant's hot-rolled process control parameters, which includes 17,000 samples, and every sample contains 17 input characteristics and three output mechanical property parameters. Based on the dataset, an LightGBM intelligent prediction model is established and trained to predict the three mechanical properties of the hot-rolled strip steels. 17,000 data of hot rolling mill are used to verify the effectiveness of the model. Results show that the prediction accuracy for tensile strength, compressive strength and elongation are 0.99971, 0.99835, and 0.99631, respectively. Especially, the prediction accuracy for elongation is higher than the existing methods.

An intelligent online detection approach based on big data for mechanical properties of hot-rolled strip

An intelligent online detection approach based on big data for mechanical properties of hot-rolled strip

Prediction model for mechanical properties of hot-rolled strips by deep learning

The prediction of the mechanical properties of hot-rolled strips is a very complex, highly dimensional and nonlinear problem, and the published models might lack reliability, practicability and generalization. Thus, a new model was proposed for predicting the mechanical properties of hot-rolled strips by deep learning. First, the one-dimensional numerical data were transformed into two-dimensional data for expressing the complex interaction between the influencing factors. Subsequently, a new convolutional network was proposed to establish the prediction model of tensile strength of hot-rolled strips, and an improved inception module was introduced into this network to abstract features from different scales. Many comparative experiments were carried out to find the optimal network structure and its hyperparameters. Finally, the prediction experiments were carried out on different models to evaluate the performance of the new convolutional network, which includes the stepwise regression, ridge regression, support vector machine, random forest, shallow neural network, Bayesian neural network, deep feed-forward network and improved LeNet-5 convolutional neural network. The results show that the proposed convolutional network has better prediction accuracy of the mechanical properties of hot-rolled strips compared with other models.

Elevating Prediction Performance for Mechanical Properties of Hot-Rolled Strips by Using Semi-Supervised Regression and Deep Learning

In the present work, to solve the problem of the lacking enough labeled training data for deep learning, a safe semi-supervised regression supporting Bayesian optimization deep neural network (SAFER-BODNN) model was proposed to establish mechanical property prediction model of hot-rolled strips. The Pearson correlation coefficient was applied to reduce the data dimension. The safe semi-supervised regression was implemented to add the pseudo labels to the unlabeled data for training dataset expansion. The deep neural network was trained with Bayesian optimization to determine the optimal hyper-parameters of the network. The results show that the SAFER-BODNN model achieves good performance for mechanical property prediction of hot-rolled strips with correlation coefficient of 0.9610 for yield strength, 0.9682 for tensile strength, and 0.8619 for elongation, respectively. Compared with the deep neural network trained on the labeled dataset, the SAFER-BODNN model obtains stable smaller predicted errors. Among all the variables, C content and Mn content have large influence on the yield strength and tensile strength, coiling temperature has the largest influence on the elongation. The investigation makes full use of unlabeled data to elevate the prediction performance of the deep neural network, and also provides a way for deep learning modeling when the data are insufficient.

Prediction model based on XGBoost for mechanical properties of steel materials

In order to predict the mechanical properties of steel materials, an intelligent prediction approach based on XGBoost model and big data is presented in this paper. The effectiveness of the model was verified by using it to predict the mechanical properties of hot-rolled strips. Firstly, a dataset with 17,710 samples was established by using the practical hot-rolled process data, where every sample has 17 characteristics (C, Si, Mn, P, S, Alt, V, Ti, Nb, Ni, Cr, Cu, Mo, B, N, final rolling temperature and curling temperature) and three output variables (tensile strength, compressive strength and elongation). 90% of 17,710 samples were used as training samples and others were used as testing samples. The simulation results showed that the accuracy of the model for tensile strength, compressive strength and elongation of the hot-rolled strip was 0.99895, 0.99576, 0.96260, respectively, which were superior to the results by using BP neural network model.

Algorithmic design and application of feedback control for coiling temperature in hot strip mill

Microstructure and mechanical properties of hot-rolled strips are directly influenced by coiling temperature, and feedback control is one of the most significant approaches to improve coiling temperature control precision. Laminar cooling control system was introduced, and the novel feedback control algorithm based on proportional–integral controller and Smith predictor was designed, respectively, and the online adaptive algorithm was used to optimize parameters of proportional–integral controller. This control system has been successfully applied in a hot strip mill, and the record curves indicate that the novel feedback control system can get better dynamic characteristic. The coiling temperature can be precisely controlled with a high accuracy of ±15°C over 98.3%.

A Knowledge-Based System for Temperature Prediction in Hot Strip Mills

Rolling temperature is an important factor affecting mechanical properties of hot rolled strip significantly. Traditional techniques cannot meet higher precision control imperatives. In the present work, a novel knowledge-based system has been developed for the temperature prediction in hot strip mills. Neural network has been used for this purpose, which is an intelligent technique that can solve nonlinear problem of temperature control by learning from the samples. Furthermore, an annealing robust learning algorithm was presented to adjust the hidden node parameters as well as the weights of the adaptive neural networks. Simulations in a multi-object mode have been carried out to verify the effectivity of new neural optimization system. Calculation results confirm the feasibility of this approach and show a good agreement with experimental values obtained from a steel plant.

Modeling of structure of double-phase low-carbon chromium steels

A physical model for determining the relative amount of phase components and the size of ferrite grains after decomposition of austenite in the process of cooling of double-phase steels is suggested. The main products of austenite transformation, i.e., polygonal ferrite, pearlite, bainite, and martensite, are considered. The driving forces of the transformation and the concentration of carbon on the phase surface are determined with the use of methods of computational thermodynamics. The model is based on equations of the classical theory of nucleation and growth. It allows for the structural features of the occurrence of γ → α transformation and contain some empirical parameters. The latter are determined using data of dilatometric measurements of the kinetics of austenite transformation and metallographic measurements of the size of ferrite grains. The model is used for predicting the kinetics of the transformation under the complex cooling conditions implemented by the VOEST-ALPINE STAHL LINZ GmbH rolling mill within the computer system for control of mechanical properties of hot-rolled strip.

A significant step forward : the full metallurgical control of the mechanical properties of hot-rolled strip with VAI-Q Strip

VAI-Q Strip is a quality control system, based on physical-metallurgical models, for hot-rolled strip. This joint development of VOEST-ALPINE Industrieanlagenbau and VOEST-ALPINE Stahl Linz allows the automatic on-line control of the mechanical properties using very accurate prediction methods. Since the beginning of 2000, an extended version of the system is in operation at VOEST-ALPINE Stahl Linz and very good quality results, as well as cost savings, have been achieved.

Cold and hot rolling of iron powder

A BPMF atomized iron powder heated for 20 min at a temperature of 800, 900, 1000, or 1100°C is still characterized by good flowability, readily descends into the region of deformation, and can be rolled into strip. At a temperature above 1100°C this powder sinters into a bar, and can no longer be rolled. The mechanical properties of hot-rolled strips are an order higher than those of cold-rolled ones. Even so, the mechanical properties of strips produced by the hot rolling of the powder under the conditions of the experiments described are inferior to those of P/M parts shaped at room temperature and sintered under optimum conditions. In experiments on the hot rolling of powders between rolls with plain barrels it has proved impossible, because of axial splitting and edge cracking, to obtain nonporous strip of width up to 60 mm. The hot rolling of powders into narrow strips should be performed in closed passes; to reduce the cost of rolls, wide strips should be rolled hot between plain roll barrels, and their edges should then be trimmed.