Forming process prediction of a self-piercing riveted joint in carbon fibre reinforced composites and aluminium alloy based on deep learning

Liu, Yang; Wu, Qingjun; Wang, Pengyue; Zhuang, Weimin

doi:10.1016/j.jmapro.2023.10.015

Cited by 9 publications

(2 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Daigo et al proposed the use of PSPNet to estimate the thickness of steel in heavy melting scrap [ 30 ]. The CNN and conditional generation antagonism model were utilized by Liu et al to predict the cross-sectional shape and damage morphology of self-piercing riveted joints in carbon fiber-reinforced composites and aluminum alloy [ 31 ]. Recently, Kato et al evaluated the internal cracks of timbers using CNNs [ 32 , 33 ].…”

Section: Introductionmentioning

confidence: 99%

In-Situ Classification of Highly Deformed Corrugated Board Using Convolution Neural Networks

Rogalka,

Grabski,

Garbowski

2024

Sensors

View full text Add to dashboard Cite

The extensive use of corrugated board in the packaging industry is attributed to its excellent cushioning, mechanical properties, and environmental benefits like recyclability and biodegradability. The integrity of corrugated board depends on various factors, including its geometric design, paper quality, the number of layers, and environmental conditions such as humidity and temperature. This study introduces an innovative application of convolutional neural networks (CNNs) for analyzing and classifying images of corrugated boards, particularly those with deformations. For this purpose, a special device with advanced imaging capabilities, including a high-resolution camera and image sensor, was developed and used to acquire detailed cross-section images of the corrugated boards. The samples of seven types of corrugated board were studied. The proposed approach involves optimizing CNNs to enhance their classification performance. Despite challenges posed by deformed samples, the methodology demonstrates high accuracy in most cases, though a few samples posed recognition difficulties. The findings of this research are significant for the packaging industry, offering a sophisticated method for quality control and defect detection in corrugated board production. The best classification accuracy obtained achieved more than 99%. This could lead to improved product quality and reduced waste. Additionally, this study paves the way for future research on applying machine learning for material quality assessment, which could have broader implications beyond the packaging sector.

show abstract

Section: Introductionmentioning

confidence: 99%

In-Situ Classification of Highly Deformed Corrugated Board Using Convolution Neural Networks

Rogalka,

Grabski,

Garbowski

2024

Sensors

View full text Add to dashboard Cite

show abstract

“…Liu Yahui [8] et al investigated the metal deformation behavior during self-pierce riveting of thin aluminum alloy plates and the mechanical behavior of self-pierce riveting joint head in destructive testing, with a view to establishing a general strength model for self-pierce riveting joint head and realizing the data processing and strength prediction of riveted joints. Liu Yang [9] et al proposed a deep learning-based prediction method for riveting process, obtaining the original images of the model dataset according to the simulation results, and using image segmentation technology to classify the damage patterns. The adopted deep learning model can accurately predict the deformation state and damage evolution of riveted materials in different connection stages.…”

Section: Introductionmentioning

confidence: 99%

Multi-indicator optimization of riveting joint forming quality of aluminum alloy sheets based on response surface test

Liu,

Zhao,

Wang

et al. 2023

Advanced Manufacturing Research

View full text Add to dashboard Cite

In this paper, the effects of the interaction between punch diameter, die depth and punch speed on the quality of riveted joints are investigated using the BBD response surface test method. The results show that the mold depth has the greatest influence on the key dimensional parameters of riveted joints, followed by the punching speed, and then the punch diameter, while the punch diameter and the mold depth are the two factors with the most obvious interaction. The optimum riveted joint process parameters determined are punch diameter of 5.24 mm, die depth of 1.44 mm, and stamping speed of 5.00 mm/s. The corresponding relative errors predicted by numerical simulation and response surface optimization objective are 5.96 % for neck thickness, 3.29 % for interlocking value, and 1.37 % for bottom thickness; and the relative errors predicted by experimental results and optimization objective are 13.42 % for neck thickness, 13.42 % for interlocking value, and 1.37 % for bottom thickness. 13.42 %, interlock value is 4.23 %, and bottom thickness is 2.23 %, the model accuracy is high, and the optimization method of response surface test can effectively improve the quality of riveted joints. Through numerical simulation, the metal flow law and stress distribution during the riveting and forming process of aluminum alloy plate were analyzed, and the strength test of riveted joint was carried out, and the maximum destructive shear load of the joint was 1.8 KN, and the strength of the joint was improved, which verified the validity of the response surface optimization method.

show abstract

MA-SPRNet: A multiple attention mechanisms-based network for self-piercing riveting joint defect detection

Zhang,

Zhao,

Ren

et al. 2024

Computers and Electrical Engineering

View full text Add to dashboard Cite

Forming process prediction of a self-piercing riveted joint in carbon fibre reinforced composites and aluminium alloy based on deep learning

Cited by 9 publications

References 40 publications

In-Situ Classification of Highly Deformed Corrugated Board Using Convolution Neural Networks

In-Situ Classification of Highly Deformed Corrugated Board Using Convolution Neural Networks

Multi-indicator optimization of riveting joint forming quality of aluminum alloy sheets based on response surface test

MA-SPRNet: A multiple attention mechanisms-based network for self-piercing riveting joint defect detection

Contact Info

Product

Resources

About