Weld Reinforcement Analysis Based on Long-Term Prediction of Molten Pool Image in Additive Manufacturing

Abstract

Wire Arc Additive Manufacturing (WAAM) has developed rapidly in recent years and has been widely used in industry. Cold Metal Transfer (CMT), a kind of Gas Metal Arc Welding (GMAW), is widely used in the modeling of thin parts. In the monitoring of CMT process, it is necessary to monitor the weld reinforcement of the deposited layer. It can provide effective alerting for welding and help to improve welding quality. In this paper, the weld reinforcement and the position of molten pool image on the deposited layer are determined by the molten pool visual acquisition system. According to the data of the same direction additive manufacturing experiment, the future molten pool image is predicted by the improved prediction network (PredNet). Meanwhile through a regression network (SERes), the predicted results are regressed to the accurate weld reinforcement information of the deposited layer in advance. The experimental results show that the monitoring system designed in this paper can predict the change of molten pool shape 140ms in advance and the average precision of the weld reinforcement regression is better than 0.3mm. This paper provides the necessary basis for the shape online monitoring and controlling in WAAM process.INDEX TERMS additive manufacturing, long-term prediction, weld reinforcement regression, molten pool image, deep learning.

Bacterial Memetic Algorithm Trained Fuzzy System-Based Model of Single Weld Bead Geometry

“…The bacterium b k+1 is given according to Eq. (15), where the Eval function is calculated as the f obj . The iteration stops if the stopping criteria τ k = ||g(b k )|| 2 < τ is fulfilled or maximum LM iter number is reached.…”

“…Now they are providing a base for applications for future intelligent welding manufacturing [11]. Recently, WBG models and the related process controls gained attention in wire and arc additive manufacturing (WAAM) [12]- [15] and multi-pass welding (MPW) [16]- [18].…”

“…during the supervised training. Equation(15) is interpreted as if the update vector modifies the bacterium towards the local minimum then we carry on with the new value, otherwise it is left unchanged. The iterations stops if maximum LM iter number is reached or the τ k < τ complex stopping criteria[54] is fulfilled.This article has been accepted for publication in a future issue of this journal, but has not been fully edited.…”

Quality Prediction of Fused Deposition Molding Parts Based on Improved Deep Belief Network

“…Additive manufacturing (AM) is a rapid prototyping technology born in the 1980s, which realizes the conversion from a 3D digital model to a physical model by continuously adding layers of materials. Compared with traditional subtractive manufacturing (cutting), additive manufacturing has the characteristics of energy-efficient, green, and recyclable [1][2][3], which meets the market demand for rapid product development and personalized customization. Fused deposition modeling (FDM) has become one of the most widely used additive manufacturing technologies at home and abroad because of its simple molding equipment, low production cost, and the ability to manufacture complex parts without extra tools [4,5].…”

Research on Automated Defect Classification Based on Visual Sensing and Convolutional Neural Network-Support Vector Machine for GTA-Assisted Droplet Deposition Manufacturing Process

“…The classification accuracy of the finally developed model on the UB-Moog dataset is 0.82 by optimizing hyper parameters. Wang et al [29], based on previous work [13], developed a prediction network (PredNet) to predict the change of molten pool shape 140ms in advance. Through regression network (SERes), the predicted results were regressed to the accurate weld reinforcement information of the deposited layer in advance.…”