scholarly journals Mechanical Properties Prediction for Hot Rolled Alloy Steel Using Convolutional Neural Network

IEEE Access ◽  
2019 ◽  
Vol 7 ◽  
pp. 47068-47078 ◽  
Author(s):  
Zhi-Wei Xu ◽  
Xiao-Ming Liu ◽  
Kai Zhang
Keyword(s):  
Neural Network ◽  
Mechanical Properties ◽  
Convolutional Neural Network ◽  
Alloy Steel ◽  
Mechanical Properties Prediction ◽  
Hot Rolled

Mechanical Properties Prediction for Cold-formed Steel Angle Connection with Various Flange Cleat Thickness

2015 ◽  
Vol 74 (4) ◽  
Author(s):  
Yeong Huei Lee ◽  
Cher Siang Tan ◽  
Shahrin Mohammad ◽  
Yee Ling Lee
Keyword(s):  
Mechanical Properties ◽  
Numerical Models ◽  
Steel Structures ◽  
Rotational Stiffness ◽  
Steel Angle ◽  
Cold Formed Steel ◽  
Mechanical Properties Prediction ◽  
Thin Walled ◽  
Angle Connection ◽  
Hot Rolled

Connection is an important element in structural steelwork construction. Eurocode does not provide adequate design information for mechanical properties prediction of top-seat flange cleat connection, especially for thin-walled cold-formed steel structures. Adopting hot-rolled design with neglecting thin-walled behaviour could lead to unsafe or uneconomic design. This research aims to provide accurate mechanical properties prediction for bolted top-seat flange cleat connection in cold-formed steel structures. The scope of work focuses on the effect of various thickness of the flange cleat to the rotational stiffness and strength behaviour of a beam-to-column connection. Experimentally verified and validated finite element modelling technique is applied in the parametric investigation. Two categories of flange cleat thickness, ranged from 2 mm to 40 mm are studied. From the developed numerical models, it is observed that Eurocode has overestimated the initial rotational stiffness prediction, calculated with component method. The over-estimation would influence the overall stiffness of structures and force distribution within the components. As a conclusion, a set of newly proposed accurate predictions for initial rotational stiffness and strength of cold-formed steel top-seat flange cleat connection, with the influence of the thickness of flange cleat is presented.

scholarly journals Modeling and Composition Design of Low-Alloy Steel’s Mechanical Properties Based on Neural Networks and Genetic Algorithms

Materials ◽  
2020 ◽  
Vol 13 (23) ◽  
pp. 5316
Author(s):  
Zhenlong Zhu ◽  
Yilong Liang ◽  
Jianghe Zou
Keyword(s):  
Neural Network ◽  
Mechanical Properties ◽  
Heat Treatment ◽  
Genetic Algorithm ◽  
Alloy Steel ◽  
Deep Neural Network ◽  
Treatment Process ◽  
Mechanical Performance ◽  
Low Alloy Steel ◽  
Forward Selection

Accurately improving the mechanical properties of low-alloy steel by changing the alloying elements and heat treatment processes is of interest. There is a mutual relationship between the mechanical properties and process components, and the mechanism for this relationship is complicated. The forward selection-deep neural network and genetic algorithm (FS-DNN&GA) composition design model constructed in this paper is a combination of a neural network and genetic algorithm, where the model trained by the neural network is transferred to the genetic algorithm. The FS-DNN&GA model is trained with the American Society of Metals (ASM) Alloy Center Database to design the composition and heat treatment process of alloy steel. First, with the forward selection (FS) method, influencing factors—C, Si, Mn, Cr, quenching temperature, and tempering temperature—are screened and recombined to be the input of different mechanical performance prediction models. Second, the forward selection-deep neural network (FS-DNN) mechanical prediction model is constructed to analyze the FS-DNN model through experimental data to best predict the mechanical performance. Finally, the FS-DNN trained model is brought into the genetic algorithm to construct the FS-DNN&GA model, and the FS-DNN&GA model outputs the corresponding chemical composition and process when the mechanical performance increases or decreases. The experimental results show that the FS-DNN model has high accuracy in predicting the mechanical properties of 50 furnaces of low-alloy steel. The tensile strength mean absolute error (MAE) is 11.7 MPa, and the yield strength MAE is 13.46 MPa. According to the chemical composition and heat treatment process designed by the FS-DNN&GA model, five furnaces of Alloy1–Alloy5 low-alloy steel were smelted, and tensile tests were performed on these five low-alloy steels. The results show that the mechanical properties of the designed alloy steel are completely within the design range, providing useful guidance for the future development of new alloy steel.

scholarly journals Comparative evaluation of performance and scalability of convolutional neural network implementations on a multisystem HPC architecture

2021 ◽  
Vol 2062 (1) ◽  
pp. 012008
Author(s):  
Sunil Pandey ◽  
Naresh Kumar Nagwani ◽  
Shrish Verma
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
High Performance ◽  
Processing Unit ◽  
Neural Network Training ◽  
Central Processing ◽  
Network Training ◽  
Hot Rolled ◽  
Hot Rolled Steel ◽  
Performance Computing

Abstract The convolutional neural network training algorithm has been implemented for a central processing unit based high performance multisystem architecture machine. The multisystem or the multicomputer is a parallel machine model which is essentially an abstraction of distributed memory parallel machines. In actual practice, this model corresponds to high performance computing clusters. The proposed implementation of the convolutional neural network training algorithm is based on modeling the convolutional neural network as a computational pipeline. The various functions or tasks of the convolutional neural network pipeline have been mapped onto the multiple nodes of a central processing unit based high performance computing cluster for task parallelism. The pipeline implementation provides a first level performance gain through pipeline parallelism. Further performance gains are obtained by distributing the convolutional neural network training onto the different nodes of the compute cluster. The two gains are multiplicative. In this work, the authors have carried out a comparative evaluation of the computational performance and scalability of this pipeline implementation of the convolutional neural network training with a distributed neural network software program which is based on conventional multi-model training and makes use of a centralized server. The dataset considered for this work is the North Eastern University’s hot rolled steel strip surface defects imaging dataset. In both the cases, the convolutional neural networks have been trained to classify the different defects on hot rolled steel strips on the basis of the input image. One hundred images corresponding to each class of defects have been used for the training in order to keep the training times manageable. The hyperparameters of both the convolutional neural networks were kept identical and the programs were run on the same computational cluster to enable fair comparison. Both the convolutional neural network implementations have been observed to train to nearly 80% training accuracy in 200 epochs. In effect, therefore, the comparison is on the time taken to complete the training epochs.

scholarly journals An Unconventional Approach for Analyzing the Mechanical Properties of Natural Fiber Composite Using Convolutional Neural Network

2021 ◽  
Vol 2021 ◽  
pp. 1-15
Author(s):  
Govindaraj Ramkumar ◽  
Satyajeet Sahoo ◽  
G. Anitha ◽  
S. Ramesh ◽  
P. Nirmala ◽  
...  
Keyword(s):  
Neural Network ◽  
Mechanical Properties ◽  
Convolutional Neural Network ◽  
Natural Fiber ◽  
Fiber Composite ◽  
Ground Truth ◽  
Fiber Composites ◽  
Natural Fiber Composites ◽  
Training Process ◽  
Natural Fiber Composite

Over the past few years, natural fiber composites have been a strategy of rapid growth. The computational methods have become a significant tool for many researchers to design and analyze the mechanical properties of these composites. The mechanical properties such as rigidity, effects, bending, and tensile testing are carried out on natural fiber composites. The natural fiber composites were modeled by using some of the computation techniques. The developed convolutional neural network (CNN) is used to accurately predict the mechanical properties of these composites. The ground-truth information is used for the training process attained from the finite element analyses below the plane stress statement. After completion of the training process, the developed design is authorized using the invisible data through the training. The optimum microstructural model is identified by a developed model embedded with a genetic algorithm (GA) optimizer. The optimizer converges to conformations with highly enhanced properties. The GA optimizer is used to improve the mechanical properties to have the soft elements in the area adjacent to the tip of the crack.

scholarly journals Steel Defects Analysis Using CNN (Convolutional Neural Networks)

2020 ◽  
Vol 23 (6) ◽  
pp. 1155-1171
Author(s):  
Rodion Dmitrievich Gaskarov ◽  
Alexey Mikhailovich Biryukov ◽  
Alexey Fedorovich Nikonov ◽  
Daniil Vladislavovich Agniashvili ◽  
Danil Aydarovich Khayrislamov
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Image Data ◽  
Training Image ◽  
Data Set ◽  
Image Encoding ◽  
Almost Everywhere ◽  
Defects Analysis ◽  
Hot Rolled ◽  
Hot Rolled Steel

Steel is one of the most important bulk materials these days. It is used almost everywhere - from medicine to industry. Detecting this material's defects is one of the most challenging problems for industries worldwide. This process is also manual and time-consuming. Through this study we tried to automate this process. A convolutional neural network model UNet was used for this task for more accurate segmentation with less training image data set for our model. The essence of this NN (neural network) is in step-by-step convolution of every image (encoding) and then stretching them to initial resolution, consequently getting a mask of an image with various classes on it. The foremost modification is changing an input image's size to 128x800 px resolution (original images in dataset are 256x1600 px) because of GPU memory size's limitation. Secondly, we used ResNet34 CNN (convolutional neural network) as encoder, which was pre-trained on ImageNet1000 dataset with modified output layer - it shows 4 layers instead of 34. After running tests of this model, we obtained 92.7% accuracy using images of hot-rolled steel sheets.

Sign in / Sign up

Export Citation Format

Share Document