Chemometric Approach for Mechanical Properties Prediction during the Electromagnetic Casting Process

Cao

et al. 2020

IEEE Access

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 accuracy for mechanical property prediction of hot-rolled strips with R 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 accuracy of the deep neural network, and also provides a way for deep learning modeling when the data are insufficient.

Section: A Deep Neural Networkmentioning

confidence: 99%

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

Cao

et al. 2020

IEEE Access

“…The weights are iteratively adjusted to minimize the objective function to obtain a well-trained neural network. 19) The BRNN is a back propagation neural network with the training function of 'trainbr' to adjust the weights. 20) The 'trainbr' is a modification of the Levenberg-Marquardt algorithm and is effective to determine the optimal network parameters.…”

Section: Ys Prediction By Using Brnnmentioning

confidence: 99%

Composition-Processing-Property Correlation Mining of Nb–Ti Microalloyed Steel Based on Industrial Data

Liu

2020

Mater. Trans.

Modeling strength of hot rolled strip based on industrial data may cause misleading predictions because of the high dimension, low quality and unbalanced original data. Industrial data processing is essential to building a successful composition-process-property corresponding relationship model. In current work, the Pauta criterion was implemented to eliminate abnormal data, and uniform grid technique was applied to select out the represented data for obtaining the balanced training data set. Prior to modeling by Bayesian regularization neural network, principal component analysis was applied to alleviate the effect of correlation variables on the modeling. The yield strength prediction model of NbTi microalloyed steel was established with the relative error of «8%, indicating a good agreement between predicted value and measured value. Finally, the relationship of chemical composition, process parameters and yield strength of hot rolled strip under specific process parameters was analyzed for a further investigation.

A convolutional neural network-based model for predicting lime utilization ratio in the KR desulfurization process

2021

Metall. Res. Technol.

In the presented work, desulfurization process parameters and the lime utilization ratio were correlated by data-driven technique, and a convolutional neural network was applied to predict the lime utilization ratio in the Kambara Reactor (KR) desulfurization process. The results show that compared with the support vector regression model and random forest model, the convolutional neural network model achieves the best performance with correlation coefficient value of 0.9964, mean absolute relative error of 0.01229 and root mean squared error of 0.3392%. The sensitivity analysis was carried out to investigate the influence of process parameters on the lime utilization ratio, which shows that the lime weight and the initial sulfur content have the significant effect on the lime utilization ratio. By analyzing the influence of the lime weight on the lime utilization ratio under the current desulfurization process parameters, it can be concluded that decreasing the lime weight from 3256 kg to 2332 kg can increase the lime utilization ratio by about 5%.