Real-time monitoring of high-power disk laser welding statuses based on deep learning framework

2019 ◽  
Vol 31 (4) ◽  
pp. 799-814 ◽  
Author(s):  
Yanxi Zhang ◽  
Deyong You ◽  
Xiangdong Gao ◽  
Congyi Wang ◽  
Yangjin Li ◽  
...  
Keyword(s):  
Deep Learning ◽  
Laser Welding ◽  
Real Time ◽  
High Power ◽  
Real Time Monitoring ◽  
Disk Laser ◽  
Learning Framework

Real-time monitoring of high-power disk laser welding based on support vector machine

2018 ◽  
Vol 94 ◽  
pp. 75-81 ◽  
Author(s):  
Juequan Chen ◽  
Teng Wang ◽  
Xiangdong Gao ◽  
Li Wei
Keyword(s):  
Support Vector Machine ◽  
Laser Welding ◽  
Real Time ◽  
High Power ◽  
Support Vector ◽  
Real Time Monitoring ◽  
Disk Laser

scholarly journals Real-time Monitoring for Disk Laser Welding Based on Feature Selection and SVM

2017 ◽  
Vol 7 (9) ◽  
pp. 884 ◽  
Author(s):  
Teng Wang ◽  
Juequan Chen ◽  
Xiangdong Gao ◽  
Yuxin Qin
Keyword(s):  
Feature Selection ◽  
Laser Welding ◽  
Real Time ◽  
Real Time Monitoring ◽  
Disk Laser

scholarly journals Comparison of Forecasting Models for Real-Time Monitoring of Water Quality Parameters Based on Hybrid Deep Learning Neural Networks

Water ◽  
2021 ◽  
Vol 13 (11) ◽  
pp. 1547
Author(s):  
Jian Sha ◽  
Xue Li ◽  
Man Zhang ◽  
Zhong-Liang Wang
Keyword(s):  
Neural Network ◽  
Water Quality ◽  
Deep Learning ◽  
Real Time ◽  
Best Management Practices ◽  
Input Data ◽  
Management Practices ◽  
Quality Parameters ◽  
Water Quality Parameters ◽  
Real Time Monitoring

Accurate real-time water quality prediction is of great significance for local environmental managers to deal with upcoming events and emergencies to develop best management practices. In this study, the performances in real-time water quality forecasting based on different deep learning (DL) models with different input data pre-processing methods were compared. There were three popular DL models concerned, including the convolutional neural network (CNN), long short-term memory neural network (LSTM), and hybrid CNN–LSTM. Two types of input data were applied, including the original one-dimensional time series and the two-dimensional grey image based on the complete ensemble empirical mode decomposition algorithm with adaptive noise (CEEMDAN) decomposition. Each type of input data was used in each DL model to forecast the real-time monitoring water quality parameters of dissolved oxygen (DO) and total nitrogen (TN). The results showed that (1) the performances of CNN–LSTM were superior to the standalone model CNN and LSTM; (2) the models used CEEMDAN-based input data performed much better than the models used the original input data, while the improvements for non-periodic parameter TN were much greater than that for periodic parameter DO; and (3) the model accuracies gradually decreased with the increase of prediction steps, while the original input data decayed faster than the CEEMDAN-based input data and the non-periodic parameter TN decayed faster than the periodic parameter DO. Overall, the input data preprocessed by the CEEMDAN method could effectively improve the forecasting performances of deep learning models, and this improvement was especially significant for non-periodic parameters of TN.

Segmentation of Molten Pool Infrared Images Using Morphology during High Power Disk Laser Welding

2012 ◽  
Vol 201-202 ◽  
pp. 91-94
Author(s):  
Yan Xi Zhang ◽  
Xiang Dong Gao
Keyword(s):  
Laser Welding ◽  
High Power ◽  
Molten Pool ◽  
Welding Process ◽  
Histogram Equalization ◽  
Welding Quality ◽  
Infrared Images ◽  
Traditional Methods ◽  
Disk Laser ◽  
Laser Welding Process

Configuration of a molten pool is related to the laser welding quality. Analyzing the configuration of a molten pool is important to monitor the laser welding process. This paper proposes a method of segmentation of a molten pool and its shadow during high power disk laser welding, consequently provides the groundwork for reconstruction of the molten pool and analysis of welding quality. Subsection linear stretching histogram equalization was applied to enhance the contrast of the original images firstly, and then edge detection was used to highlight the edges. After that we used the morphology filtering method to produce the segmentation mask, and then combined the mask with the original images to get the final segmentation results. Also, the proposed method was compared with other traditional methods. The experimental results showed that our method not only could give better segmentation results and process large quantities images automatically, but also overcame the less-segmentation problems of traditional methods.

Analysis of Image Characteristics of Plume and Spatter of High Power Disk Laser Welding Based on K-L Transform

2012 ◽  
Vol 532-533 ◽  
pp. 330-334
Author(s):  
Qian Wen ◽  
Xiang Dong Gao
Keyword(s):  
Laser Welding ◽  
High Power ◽  
High Speed ◽  
Metal Vapor ◽  
Mean Value ◽  
Gray Level ◽  
Characteristic Parameters ◽  
Transform Method ◽  
Disk Laser ◽  
Vapor Plume

Metal vapor plume and spatters are the important phenomena in the process of high power disk laser welding, and there exists a close relationship with the welding stability. The images of metal vapor plume and spatters which captured by a high speed camera during high power disk laser welding were analyzed in this experiment. Image processing techniques such as median filtering, Wiener filtering, gray level threshold and lightness transform were used to process the images so that the image characteristic parameters such as the area and number of spatters in an image, the average gray, mean value, variance and entropy of a spatter gray level image and the coordinate ratio of the centriod of plume and the welding point can be extracted. To reflect the actual welding results obviously by those characteristic parameters, K-L transform method was used to get a new set of characteristic parameters. Experimental results showed that this new set of characteristic parameters could reflect the actual welding effectively.

Real-Time Monitoring of Coaxial Protection Fiber Laser Welding of Austenitic Stainless Steels

2017 ◽  
Vol 44 (5) ◽  
pp. 0502003 ◽  
Author(s):  
任勇 Ren Yong ◽  
武强 Wu Qiang ◽  
邹江林 Zou Jianglin ◽  
陈乐 Chen Le ◽  
肖荣诗 Xiao Rongshi
Keyword(s):  
Laser Welding ◽  
Fiber Laser ◽  
Real Time ◽  
Stainless Steels ◽  
Austenitic Stainless Steels ◽  
Real Time Monitoring ◽  
Fiber Laser Welding

Weld width prediction of weldment bottom surface in high-power disk laser welding

2017 ◽  
Vol 25 (9) ◽  
pp. 2524-2531
Author(s):  
陈子琴 CHEN Zi-qin ◽  
高向东 GAO Xiang-dong ◽  
王 琳 WANG Lin
Keyword(s):  
Laser Welding ◽  
High Power ◽  
Bottom Surface ◽  
Disk Laser ◽  
Weld Width
Sign in / Sign up

Export Citation Format

Share Document