Entity

Time filter

Source Type


Gao X.,Guangdong University of Technology | Liu Y.,Guangdong University of Technology | Xiao Z.,Guangzhou Panyu Gofront Dyeing & Finishing Machinery Manufacturer Ltd | Chen X.,Guangzhou Panyu Gofront Dyeing & Finishing Machinery Manufacturer Ltd
Hanjie Xuebao/Transactions of the China Welding Institution | Year: 2015

A multi-sensor information fusion method based on support vector machine was studied to analyze the high-power disk laser welding status. During high-power disk laser welding, the metallic plume, spatters and molten pool are important phenomena which are related to the welding quality. An ultraviolet and visible sensitive video camera was used to capture the metallic plume and spatter dynamic images, and another infrared sensitive video camera was used to capture the molten pool images. The image processing and pattern recognition technologies were applied to extract the welding characteristics information and analyze the principal components. Weld bead width was used as a characteristic parameter that reflects the welding stability. After data normalization and characteristic analysis, the multi-sensor information was fused by the support vector machine, and the grid search method and particle swarm optimization were used to optimize the experimental parameters of support vector machine. Finally a fusion model based on support vector machine was established to estimate the weld bead width. Experimental results showed that the multi-sensor information fusion based on support vector machine could effectively predict the weld bead width, thus providing an experimental evidence for monitoring the high-power disk laser welding status. © 2015, Harbin Research Institute of Welding. All right reserved. Source


Chen Y.,Guangdong University of Technology | Gao X.,Guangdong University of Technology | Xiao Z.,Guangzhou Panyu Gofront Dyeing & Finishing Machinery Manufacturer Ltd | Chen X.,Guangzhou Panyu Gofront Dyeing & Finishing Machinery Manufacturer Ltd
China Welding (English Edition) | Year: 2015

Keyhole is one of the important phenomena in high-power laser welding process. By studying the keyhole characteristic and detecting the seam offset during high-power fiber laser welding, an infrared sensitive high-speed camera arranged off-axis orientation of laser beam was applied to capture the dynamic thermal images of a molten pool. The 304 austenitic stainless steel plate butt joint welding experiment with laser power 10 kW was carried out. Through analyzing the keyhole infrared image, the weld position was calculated. Least squares method was used to determine the actual weld position. Image filtering technique was used to process the keyhole image, and the keyhole centroid coordinate were calculated. Also, least squares method was used to fit the keyhole centroid curve equation and establish a nonlinear continuous model which described the deviation between keyhole centroid and weld seam. The heat accumulation effect affected by the infrared radiation was analyzed to determine whether the laser beam focus spot deviated from the desired welding seam. Experimental results showed that the keyhole centroid has related to the seam offset, and can reflect the welding quality. Copyright: © 2015 Editorial Board of CHINA WELDING. Source


Gao X.,Guangdong University of Technology | Mo L.,Guangdong University of Technology | Xiao Z.,Guangzhou Panyu Gofront Dyeing & Finishing Machinery Manufacturer Ltd | Chen X.,Guangzhou Panyu Gofront Dyeing & Finishing Machinery Manufacturer Ltd
Jixie Gongcheng Xuebao/Journal of Mechanical Engineering | Year: 2016

It is required that the laser beam focus spot be controlled to follow and track the weld joint accurately during butt joint laser welding. An approach for detecting and tracking the micro-gap (whose width is less than 0.1 mm) weld position based on Kalman filtering during butt joint laser welding is presented. A magneto-optical sensor is used to capture the micro-gap weld image sequence, and the weld position coordinate is extracted by analyzing the gray scale gradient feature of images. The state equation and measurement equation for weld position are established based on the eigenvector derived from the weld position and displacement variables. Considering that the system dynamic noise and measurement noise are Gauss white noises with zero mean random distribution, a Kalman filter algorithm is established, which could reduce the influence of system and process noises on the weld position measurement. The optimal state estimation of weld centre ordinates could be obtained through the optimal state estimation of weld position under the least squares condition. Laser welding experimental results show that the proposed method could accurately predict the micro-gap weld position during the butt joint laser welding. © 2016 Journal of Mechanical Engineering. Source


Mo L.,Guangdong University of Technology | Gao X.,Guangdong University of Technology | Xiao Z.,Guangzhou Panyu Gofront Dyeing & Finishing Machinery Manufacturer Ltd | Chen X.,Guangzhou Panyu Gofront Dyeing & Finishing Machinery Manufacturer Ltd
Hanjie Xuebao/Transactions of the China Welding Institution | Year: 2016

This paper proposes a novel method to realize accurate seam tracking of butt micro-gap (less than 0.1mm) joint during laser welding based on the magneto-optical color images. Images of weld seam zone are captured by a magneto-optical sensor, and grayscale distribution of the weld seam magneto-optical images in the RGB(Red, Green, Blue)and HSV(Hue, Saturation, Value) color space are analyzed. The grayscale map of component of the RGB figure is extracted, and weld seam edge is determined based on the threshold derived from the grayscale distribution curve of each component. Also, the outline of weld seam transitional zone is obtained by integrating weld seam edges of the three components. The histogram of each HSV component is analyzed to determine the threshold and then the synthetic seam transition zone segmentation is obtained. Experimental results show that the proposed method can effectively detect micro-gap weld seam which is generally hard to distinguish by human eyes. © 2016, Harbin Research Institute of Welding. All right reserved. Source


Gao X.,Guangdong University of Technology | Lin J.,Guangdong University of Technology | Xiao Z.,Guangzhou Panyu Gofront Dyeing & Finishing Machinery Manufacturer Ltd | Chen X.,Guangzhou Panyu Gofront Dyeing & Finishing Machinery Manufacturer Ltd
Hanjie Xuebao/Transactions of the China Welding Institution | Year: 2016

A BP neural network model based on ICA (Imperialist Competitive Algorithm) is proposed to recognize the arc welding penetration status. The weights and thresholds of the neural network are initialized using ICA which has the features of uneasy accessibility to local extremum and fast search speed. Then the BP algorithm is used to train the neural network. By capturing the images of the molten pool in welding process, three features of a molten pool image are processed. The features includes the weld pool area, weld pool width and the distance between the weld pool centroid and the bottom. These features are as the inputs of neural network to create the mapping relationship between the three features of molten pool and the weld penetration status, and eventually a predicted model of penetration status is established. Welding experimental results show that the welding penetration status can be accurately recognized using the ICA-BP neural network. © 2016, Editorial Board of Transactions of the China Welding Institution, Magazine Agency Welding. All right reserved. Source

Discover hidden collaborations