Composition–Processing–Property Correlation of Cold-Rolled IF Steel Sheets Using Neural Network

Mohanty, Itishree; Datta, Someswar; Bhattacharjee, D.

doi:10.1080/10426910802543947

Cited by 26 publications

(11 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on these data, the machine learning algorithm can be used for data mining and relating chemical compositions, process parameters and mechanical properties, so as to realize the mechanical property prediction of steels [10][11][12]. Some researchers have applied the commonly used algorithms, such as artificial neural network [13][14][15][16][17][18], support vector machine [19], nerofuzzy inference system [20], semi-parametric single index model [21] etc., to investigate the prediction model of mechanical properties and have made some achievements. However, the artificial neural network is prone to over fitting with the increase of the hidden layer neuron number [22].…”

Section: Introductionmentioning

confidence: 99%

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

Yang

Cao

et al. 2020

IEEE Access

View full text Add to dashboard Cite

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.

show abstract

Section: Introductionmentioning

confidence: 99%

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

Yang

Cao

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…This is because that higher CT favors coarse and widely spaced precipitates of carbides and nitrides during hot rolling. This type of precipitation fixes carbon without employing any strengthening effect . However, at the point of FDT = 790 °C and CT = 620 °C, the YS is especially high, which is contradict with physical metallurgy principle.…”

Section: Mechanical Property Prediction Modelingmentioning

confidence: 96%

High‐Dimensional Data‐Driven Optimal Design for Hot Strip Rolling of Microalloyed Steel

Zhou

Cao

et al. 2018

steel research int.

View full text Add to dashboard Cite

To solve the contradiction between consumer's personal demand and mass production of steel enterprise, the optimal design for hot strip rolling of steel attracts researcher's attention. The core of the ensemble system is the artificial neural network (ANN) model, which is based on the industrial data. However, industrial data are difficult to be used for modeling because of their high dimension, low quality, and unbalanced. In current work, the method for industrial data processing is proposed for microalloyed steel. Based on the treated data, the mechanical property prediction models are established with the relative error of ±8% for yield strength (YS), ±6% for tensile strength (TS), and the absolute error of ±4% for elongation (EL), respectively. Combined with the multi‐objective optimization algorithm, the ANN models are implemented for designing the hot rolling process for S360 grade steel to control the stability of product quality.

show abstract

“…The complex correlation between the property, structure, and composition are challenging to be modeled using physical models, and it is replaced by an established artificial intelligence-based data-driven modeling technique, artificial neural network (ANN), 24 which is successfully used in the field of materials engineering. [25][26][27][28] To have better understanding on the influence of nanoparticles individually as well as in tandem as reinforcement for UHMWPE composites on the mechanical properties, ANN modeling is used. For the modeling, a database consisting individual additions of multi-walled carbon nanotube (MWCNT) and graphene is generated from published work and merged.…”

Section: Introductionmentioning

confidence: 99%

Design of the ultrahigh molecular weight polyethylene composites with multiple nanoparticles: An artificial intelligence approach

Vinoth

Datta

2019

Journal of Composite Materials

View full text Add to dashboard Cite

This study proposes a suitable composite material for acetabular cup replacements in hip joint that involves ultrahigh molecular weight polyethylene, a clinically proven material, as the matrix. To design new ultrahigh molecular weight polyethylene composites with multiple reinforcements for the improvement in mechanical and tribological performance, artificial neural network and genetic algorithm, the two artificial intelligence techniques, are employed. Published reports on the use of ultrahigh molecular weight polyethylene reinforced with multi-walled carbon nanotube and graphene are used as database to develop two artificial neural network models for Young's modulus and tensile strength. The optimum solutions are obtained using genetic algorithm, where the artificial neural network models are used as the objective functions. Two different composites, derived from the optimum solutions, are made reinforcing both multi-walled carbon nanotube and graphene. Tensile and wear tests show significant enhancement in the properties. The structures of the composites are also characterized, and wear mechanisms are discussed.

show abstract

Composition–Processing–Property Correlation of Cold-Rolled IF Steel Sheets Using Neural Network

Cited by 26 publications

References 6 publications

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

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

High‐Dimensional Data‐Driven Optimal Design for Hot Strip Rolling of Microalloyed Steel

Design of the ultrahigh molecular weight polyethylene composites with multiple nanoparticles: An artificial intelligence approach

Contact Info

Product

Resources

About