Explore the vast range of titles available.

Due to the emergence of mass personalized production, intelligent welding systems must achieve high levels of productivity and flexibility. Therefore, a self-learning welding-task sequencing method that is driven by data and knowledge was developed during this study. First, a minimized dataset of welding sequences, which is required to predict the welding deformation, was designed according to the number and directions of the welds included in the welding tasks. The dataset consisted of a finite number of welding sequences and their corresponding welding deformation data. Then, an algorithm to predict the welding deformation was developed. To improve the interpretability of the results, domain knowledge was integrated into the construction and training processes of a self-learning model. Finally, a case study regarding bracket welding was investigated. With FEA as the benchmark, the maximum relative error of the welding deformation predicted by the algorithm designed to predict the welding deformation was 8%. The maximum deformation of the optimal welding-task sequence output by the self-learning welding-task sequencing method driven by data and knowledge was 32.31% less than that produced by the rule-based reasoning method. The study results demonstrate that the proposed welding-task sequencing method is effective for welding sequence planning of laser welding bracket structures.

In the modern era, welding systems have been made smart with the inclusion of contemporary information technologies such as intelligent manufacturing and machine learning (ML). The ML has been integrated with a wide application area of metal joining to achieve the status of intelligent welding systems (IWS). The IWS, using ML, has drawn massive consideration from researchers and industrialists to obtain high product quality and cost-effective solutions. Intelligent welding uses modern computers for sensing, learning, decision-making, monitoring, and control, thus replacing/minimizing human interference. ML-integrated welding is primarily for modeling, identification, optimization, prediction, and controlling multiple variables. Citing the necessity and importance of ML models in weld quality and process optimization, the current study is aimed on describing basics of ML techniques, their types, models, and adaptability scenarios in numerous industrially sought IWS.

In recent years, welding feedback control systems and weld quality estimation systems have been developed with the use of artificial intelligence to increase the quality consistency of robotic welding solutions. This paper introduces the utilization of an intelligent welding system (IWS) for feedback controlling the welding process. In this study, the GMAW process is controlled by a backpropagation neural network (NN). The feedback control of the welding process is controlled by the input parameters; root face and root gap, measured by a laser triangulation sensor. The NN is trained to adapt NN output parameters; wire feed and arc voltage override of the weld power source, in order to achieve consistent weld quality. The NN is trained offline with the specific parameter window in varying weld conditions, and the testing of the system is performed on separate specimens to evaluate the performance of the system. The butt-weld case is explained starting from the experimental setup to the training process of the IWS, optimization and operating principle. Furthermore, the method to create IWS for the welding process is explained. The results show that the developed IWS can adapt to the welding conditions of the seam and feedback control the welding process to achieve consistent weld quality outcomes. The method of using NN as a welding process parameter optimization tool was successful. The results of this paper indicate that an increased number of sensors could be applied to measure and control the welding process with the developed IWS.

With the rapid development of artificial intelligence and intelligent manufacturing, the traditional teaching-playback mode and the off-line programming (OLP) mode cannot meet the flexible and fast modern manufacturing mode. Therefore, the intelligent welding robots have been widely developed and applied into the industrial production lines to improve the manufacturing efficiency. The sensing system of welding robots is one of the key technologies to realize the intelligent robot welding. Due to its unique characteristics of good robustness and high precision, the structured light sensor has been widely developed in the intelligent welding robots. To adapt to different measurement tasks of the welding robots, many researchers have designed different structured light sensors and integrated them into the intelligent welding robots. Therefore, the latest research and application work about the structured light sensors in the intelligent welding robots is analyzed and summarized, such as initial weld position identification, parameter extraction, seam tracking, monitoring of welding pool, and welding quality detection, to provide a comprehensive reference for researchers engaged in these related research work as much as possible.

Intelligent robotic welding involves replicating the role of a manual professional welder to adaptively control the welding process. This is necessary to achieve accurate, fast and high-quality welding process in addition to the challenging factors for humans to operate in the welding environment. Therefore, robotic welding exists since the early days of robotics and it is still an active research area. This is why there have been numerous researches in this area for a very long time. Among various techniques proposed by researchers for the adaptive control of the robotic welding process, vision-based control is the most popular due to its non-invasiveness. Therefore, in this paper, we review, analyse and categorise the proposed vision-based techniques with the aim of covering the different image processing and feature extraction aspect of the techniques. The focus is mainly on the active vision system where various image processing techniques have been utilised in extracting the welding seam features. The challenges and difficulties to extract seam features in active vision system have been highlighted. The trends and new approaches have been indicated in order to provide a comprehensive source for researchers who are planning to carry out research related to the intelligent robot vision techniques for welding automation.

Intelligent welding parameter control is fast becoming a key instrument for attaining quality consistency in automated welding. Recent scientific breakthroughs in intelligent systems have turned the focus of adaptive welding control to artificial intelligence-based welding parameter control. The aim of this study is to combine artificial neural network (ANN) decision-making software and a machine vision system to develop an adaptive artificial intelligence (AI)-based gas metal arc welding (GMAW) parameter control system. The machine vision system uses a laser sensor to scan the upcoming seam and gather seam profile data. Based on further processing of the seam profile data, welding parameters are optimized by the decision-making system. In this work, the developed system is tested in a multivariable welding condition environment and its performance is evaluated. The quality of the welds was consistent and surpassed the required quality level. Additionally, the heat-affected zone (HAZ) was evaluated by microscopy, X-ray, and scanning electron microscope (SEM) imaging. It is concluded that the developed ANN system is suitable for implementation in automated applications, can improve quality consistency and cost efficiency, and reduce required workpiece preparation and handling.

The H-beam riveting and welding work cell is an automated unit used for processing H-beams. By coordinating the gripping and welding robots, the work cell achieves processes such as riveting and welding stiffener plates, transforming the H-beam into a stiffened H-beam. In the context of intelligent manufacturing, there is still significant potential for improving the productivity of riveting and welding tasks in existing H-beam riveting and welding work cells. In response to the multi-agent system of the H-beam riveting and welding work cell, a recurrent multi-agent proximal policy optimization algorithm (rMAPPO) is proposed to address the multi-agent scheduling problem in the H-beam processing. The algorithm employs recurrent neural networks to capture and process historical information. Action masking is used to filter out invalid states and actions, while a shared reward mechanism is adopted to balance cooperation efficiency among agents. Additionally, value function normalization and adaptive learning rate strategies are applied to accelerate convergence. This paper first analyzes the H-beam processing flow and appropriately simplifies it, develops a reinforcement learning environment for multi-agent scheduling, and applies the rMAPPO algorithm to make scheduling decisions. The effectiveness of the proposed method is then verified on both the physical work cell for riveting and welding and its digital twin platform, and it is compared with other baseline multi-agent reinforcement learning methods (MAPPO, MADDPG, and MASAC). Experimental results show that, compared with other baseline methods, the rMAPPO-based agent scheduling method can reduce robot waiting times more effectively, demonstrate greater adaptability in handling different riveting and welding tasks, and significantly enhance the manufacturing efficiency of stiffened H-beam.

A vibration-resistant detection method of weld position and gap based on laser vision sensing is proposed in this paper due to the failure problem of automatic weave weld tracking of V-butt welds with gaps due to arc light, molten metal splash, seam gap variations, and inertial vibration of the weave motion in the manufacture of weave gas metal arc welding for pipelines vessels and ships. An improved random sampling consistency algorithm and an adaptive grayscale centroid algorithm are proposed to overcome the interference of arc light and molten metal splash, achieving the simultaneous image detection of weld position and gap. Moreover, a moving polynomial fitting algorithm is proposed to overcome vibration interference in the direction of weave motion and correct the weld position. Finally, the experimental results of Z-weave welding seam tracking of S-curve welds show that the proposed method can significantly reduce the weld tracking error, meeting the practical welding requirements. This study provides a new solution for eliminating the vibration interference of system devices in practical weave welding manufacturing.

Intelligent robotic welding requires automatic finding of the seam geometrical features in order for an efficient intelligent control. Performance of the system, therefore, heavily depends on the success of the seam finding stage. Among various seam finding techniques, active laser vision is the most effective approach. It typically requires high-quality lasers, camera and optical filters. The success of the algorithm is highly sensitive to the image processing and feature extraction algorithms. In this work, sequential image processing and feature extraction algorithms are proposed to effectively extract the seam geometrical properties from a low-quality laser image captured without the conventional narrow band filter. A novel method of laser segmentation and detection is proposed. The segmentation method involves averaging, colour processing and blob analysis. The detection method is based on a novel median filtering technique that involves enhancing of the image object based on its underlying structure and orientation in the image. The method when applied enhances the vertically oriented laser stripe in the image which improves the laser peak detection. The image processing steps are performed to make sure that the laser profile is accurately extracted within the region of interest (ROI). Feature extraction algorithm based on pixels’ intensity distribution and neighbourhood search is also proposed that can effectively extract the seam feature points. The proposed algorithms have been implemented and evaluated on various background complexities, seam sizes, material type and laser types before and during the welding operation.

At present, the majority of welding robots belong to the teach-and-playback category in welding manufacturing engineering applications; trajectory teaching in advance of welding is time-consuming and lack of efficiency. The currently published welding seam tracking methods are also based on existing trajectories. Intelligent Welding Manufacturing (IWM) is the current research focus in welding manufacturing. The core technology of IWM is Intelligent Robot Welding Technology (IRWT). A sensing-technology-based system was the key to realize IRWT. In this paper, a teaching-free welding method based on laser visual sensing system (LVSS) for robotic gas metal arc welding (GMAW) is studied. First, a LVSS was established. A fast-unified calibration method for LVSS is proposed to improve the calibration efficiency. Then, using an image processing method based on prior knowledge, feature point with sub-pixel accuracy can be obtained in real-time. Finally, an online welding trajectory planning method is proposed to implement teaching-free welding. In order to verify the accuracy and robustness of the proposed method, experiments on V-grooves and fillet welds were performed. The results showed that the control accuracy on the V-groove and the fillet welds is suitable for most robot welding applications.