Comparisons of Different Data-Driven Modeling Techniques for Predicting Tensile Strength of X70 Pipeline Steels

Wu, Shijing; Ren, Jia-kuan; Zhou, Xiaoguang; Cao, Guangming; Liu, Zhenyu; Yang, Jian

doi:10.1007/s12666-019-01624-0

Cited by 13 publications

(8 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since its particular importance, it is always referred to as the artery of the national economy. Many scholars have conducted numerous studies on the mechanical properties of uncorroded pipelines [1][2][3][4]. Unfortunately, due to their aging and features of the materials, pipelines are incredibly vulnerable to external environmental erosion which may result in defects in generation and then reduce their reliability together with a series of catastrophic chain reactions.…”

Section: Introductionmentioning

confidence: 99%

Tensile Fracture Behavior of Corroded Pipeline: Part 1—Experimental Characterization

Yang

Liu

Feng

2020

Advances in Materials Science and Engineering

View full text Add to dashboard Cite

The understanding of the axial tensile behavior of environmentally corroded pipelines is of great significance for the design, maintenance, and evaluation of such structures. This article presents some experimental data recorded from 210 tensile tests on pipe, which were corroded from grade of 10% to 70% by electrochemical accelerated corrosion method. The fracture modes show that, for the uncorroded pipe, the fracture frequently occurs in the middle of the specimen and then propagates perpendicular to the loading direction. However, for the corroded pipe, the crack’s position, evolution angle, and path have strong randomness. The comparative analysis based on the macroscopic stress-strain relationship shows that the rapid decrease of the yield stress, ultimate strength, and strain at the fracture for corroded pipe are correlated with the fracture patterns; i.e., the fracture patterns of pipe are changed from uniform to scattered with the continuous increase of the corrosion rate. The reduction factor based on experimental data is recommended for the consideration of the corrosion effect on the tensile strength of the steel pipe. Discussion on the tensile capacity during the service time is also presented.

show abstract

Section: Introductionmentioning

confidence: 99%

Tensile Fracture Behavior of Corroded Pipeline: Part 1—Experimental Characterization

Yang

Liu

Feng

2020

Advances in Materials Science and Engineering

View full text Add to dashboard Cite

show abstract

“…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

Self Cite

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

“…By using the RF algorithm, the better prediction performance was produced than ANN, SVM, Regression Decision Tree, Ridge Regression (RR), or Stepwise Regression (SR) in some issues, such as aqueous solubility prediction in medicine development, 18) mineral distribution prediction in mineral exploration, 19) soil organic carbon prediction in environmental science 20,21) or material properties prediction in metallurgical engineering. 22) For the GBDT algorithm, there are two novel efficient implementations proposed in recent years, which are eXtreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM). The better performance of XGBoost or LightGBM has been shown in assessing the potential toxicities of pharmaceuticals and industrial chemicals, 23) price prediction, 24,25) risk prediction in the financial industry, 26) global solar radiation prediction for the use of renewable energy, 27) prediction of the bioactive molecule and protein-protein interactions in the chemical and biological fields, 28,29) compared to ANN, DNN, SVM, RF, k-nearest neighbor (KNN), autoregressive integrated moving average model (ARIMA) or Naïve Bayes (NB).…”

Section: Ensemble Learning Based Methods For Crown Prediction Of Hot-rolled Stripmentioning

confidence: 99%

Ensemble Learning Based Methods for Crown Prediction of Hot-Rolled Strip

Li¹,

Gong²,

Xing³

et al. 2021

ISIJ Int.

View full text Add to dashboard Cite

Comparisons of Different Data-Driven Modeling Techniques for Predicting Tensile Strength of X70 Pipeline Steels

Cited by 13 publications

References 23 publications

Tensile Fracture Behavior of Corroded Pipeline: Part 1—Experimental Characterization

Tensile Fracture Behavior of Corroded Pipeline: Part 1—Experimental Characterization

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

Ensemble Learning Based Methods for Crown Prediction of Hot-Rolled Strip

Contact Info

Product

Resources

About