Robotic Arc Welding Research Articles

Robotic MAG welding defects and quality assessment with a defect threshold decision model-driven method

The use of machine vision and deep learning methods for online monitoring and evaluation of welding quality during the welding process and timely detection and correction of welding defects is essential for intelligent welding manufacturing. In this paper, with the goal of online monitoring of the welding quality of the robotic arc welding for the excavator boom, we established a multi-source sensing system for robotic arc welding processes. We have developed various feature extraction algorithms for extracting information about welding pools, arc sounds, currents, voltages, and other features. We proposed to extract welding pool image features using Res2-MobileNetV3 to obtain a 12-dimensional welding pool feature, which enabled good inter-class discrimination among different welding defects. We designed a welding defect threshold decision (DD) strategy model using the responsive machine learning and lightweight network Res2-MobileNetV3 method, achieving a welding defect recognition accuracy of 96.59%. Finally, we tested the accuracy and reliability of the welding defect recognition model on an actual welding scene for the robotic welding of the excavator boom. It provided valuable scientific methods and technological approaches for intelligent welding in complex scenes.

EfficientNet-ECA: A lightweight network based on efficient channel attention for class-imbalanced welding defects classification

Welding defects recognition is crucial for ensuring weldment quality in robot arc welding. Nevertheless, due to unbalanced data distributions, limited computing resources in factory, as well as intraclass variability and interclass similarity among different welding defects, it is difficult to extract the most discriminative defect features from welding molten pool images on site, resulting in weak class-imbalanced defect recognition performance. To address the above issue, a novel lightweight network named EfficientNet-ECA is proposed for recognizing and classifying welding defects according to molten pool images of robot arc welding. Firstly, the EfficientNet-ECA network based on Efficient Channel Attention (ECA) is designed to extract the most discriminative features of different defects from molten pool images, where ECA is employed to enhance cross-channel information interactions. Secondly, a dynamic equalized focal loss function is proposed to both rebalance the loss contribution of class-imbalanced data of different defects and improve defects recognition accuracy. Subsequently, a class-imbalanced dataset containing eight typical welding defects is constructed to evaluate the EfficientNet-ECA model. Finally, the proposed model is comprehensively analyzed through ablation studies and compared with the existing state-of-the-art lightweight models, and results show that the proposed method exhibits better effectiveness and generalizability in classifying class-imbalanced defects across different welding scenarios, achieving the highest accuracy of 95.84% on a self-constructed AL5083 dataset and 96.50% on a publicly available SS304 dataset. Moreover, with fewer parameters and less complexity, the proposed method is more suitable for online welding quality monitoring in factory.

Robotizing double-electrode GMAW process through learning from human welders

Gas metal arc welding (GMAW) is the most robotized arc welding process and most widely used process for wire arc additive manufacturing (WAAM). Double-electrode GMAW (DE-GMAW) is its novel modification achieved by adding a second/bypass electrode. It provides the capability to freely adjust the base metal current (heat input) without changing the wire current (mass input). However, it must be robotized in order to be adaptive to manufacturing variations and constraints. Due to the complexity of the process, we propose learning from human welders through their attempts to adjust the bypass electrode in collaboration with the robotized GMAW. To generalize human success using a follower robot/surrogate, the distance between the wire and bypass electrode is proposed as the process state to quantify human observation and operation. As such, Inertial Measurement Unit (IMU) sensors are integrated to track the wire and bypass electrode operated by both the lead robot and the human welder. To interpret human adjustments per arc observation, a Convolutional Neural Network (CNN) is employed to process arc images and calculate the distance. The automatically obtained distance labels from IMU signals are used to train the CNN; however, they are noisy and inaccurate. Thus, the CNN is finely tuned through transfer learning using manually labeled distances. With accurately automatically calculated distances from the finely tuned CNN for all the data, the demonstrations from human welders are analyzed. The generalized knowledge is implemented by a robotic surrogate, substituting for the human welder to fully automate the DE-GMAW. Experiments demonstrated the superior performance of the fully robotized, and adaptively controlled, DE-GMAW process.

A lightweight efficient semantic segmentation with encoder-decoder for arc interference in robotic arc welding

Recognizing laser stripes under conditions of arc interference is a persistent challenge faced by linear structured light sensors due to their similarity with noise in morphological characteristics. In this paper, a symmetrical encoder-decoder semantic segmentation model named LSRNet (lightweight stripe recognition network) is proposed that enables real-time detection of laser stripes in environments with strong arc interference. It analyzes the difference between arc and laser stripe from the semantic point of view. Firstly, to facilitate quick and accurate recognition of laser stripes, we designed a Depthwise-separable Factorized Module that is both efficient and scalable with a larger receptive field. Moreover, due to the difficulty in distinguishing laser stripes from arcs in a local context, a Multi-scale Depthwise-separable Factorized Module is designed to capture more context information. Secondly, the Unified Attention Fusion Module is used to enhance the feature representation of the laser stripe. Additionally, The combination of DiceLoss(dice coefficient loss) and CELoss(cross-entropy loss) is introduced to optimize the loss, which addresses the pixel imbalance challenge that the pixels of the laser stripe are much less than the background pixels. Finally, based on the images collected in the industrial field, our ArcDataset is created for model training. Experimental results show that the proposed method achieves the highest intersection over union of 88.0% in laser stripe segmentation and a recognition speed of 101.3 FPS(frames per second).

A Novel Robotic-Vision-Based Defect Inspection System for Bracket Weldments in a Cloud–Edge Coordination Environment

Arc-welding robots are widely used in the production of automotive bracket parts. The large amounts of fumes and toxic gases generated during arc welding can affect the inspection results, as well as causing health problems, and the product needs to be sent to an additional checkpoint for manual inspection. In this work, the framework of a robotic-vision-based defect inspection system was proposed and developed in a cloud–edge computing environment, which can drastically reduce the manual labor required for visual inspection, minimizing the risks associated with human error and accidents. Firstly, a passive vision sensor was installed on the end joint of the arc-welding robot, the imaging module was designed to capture bracket weldments images after the arc-welding process, and datasets with qualified images were created in the production line for deep-learning-based research on steel surface defects. To enhance the detection precision, a redesigned lightweight inspection network was then employed, while a fast computation speed was ensured through the utilization of a cloud–edge-computing computational framework. Finally, virtual simulation and Internet of Things technologies were adopted to develop the inspection and control software in order to monitor the whole process remotely. The experimental results demonstrate that the proposed approach can realize the faster identification of quality issues, achieving higher steel production efficiency and economic profits.

Real-time seam tracking during narrow gap GMAW process based on the wide dynamic vision sensing method

Welding torch horizontal seam tracking is a crucial condition to realize automatic welding of narrow gap gas metal arc welding (NG-GMAW) robots. In the present study, a real-time welding seam tracking technology based on the passive vision sensor is proposed. To this end, a wide dynamic range (WDR) vision sensor is used to capture real-time welding images. In order to realize real-time identification of the welding torch center deviation in the process of swinging arc in a welding bead, a welding seam tracking algorithm based on the window information fusion (WIF) is proposed. In this regard, the local feature moving window recognition (LMWR) algorithm is applied to extract the groove center. Based on the improved welding image processing program, the weld pool center and welding wire center are prepared for seam tracking. In this regard, considering the trifling variation of groove and weld pool center, the welding horizontal deviation is obtained by combining them with welding wire center calculation. Aiming at establishing a correlation between consecutive images, the Kalman filter is used to estimate the deviation optimally. The tracking algorithm is successfully verified in the weld bead with a horizontal deviation, and the tracking accuracy of ±0.47 mm is achieved. The performed analyses demonstrate that the developed algorithm meets the requirements of real-time seam tracking in the GMAW process of the welding robot.

Optimization and simulation of arc welding robot parameters based on offline programming

Automatic welding is most frequently utilized in engineering and manufacturing to improve efficiency. The importance of a robot in a current welding system takes full advantage of manual welding and regulates the burden of operator duties. The virtual process affects the operator's responsibilities, bead geometry, production quality, and shop floor platform. The goal of this study is to identify the simulation process parameter that has the greatest influence on variables such arc length, speed, and duration in arc welding. The Robot Studio ABB software has been utilized to make offline programming operations and to synchronize the route configuration of virtual arc welding robot and gun on imported solid geometry. Based on parameter data gathered from the robot studio, the simufact welding simulation configured the melting efficiency and bead geometry. The two simulation numerical variables data were used as input for the optimal solution validation in multi-objective optimization. The resulting emphasis of the experiment pursued on 0.001 m thickness of stainless-steel welding material: the melting efficiency is 0.42% and 467 Nm/s arc voltages during the arc travel of 0.0207 m/s and bead sensitivity are 2.68 VM/Nm/s. The MATLAB Pareto genetic algorithm is used to optimize the arc length, arc speed, and arc duration parameters adjusted in module language edit text format, which shows the preferred parameter of selected thickness. The results show that arc length and arc speed are practical as best solution dominators, while arc time is the non-dominating solution.

Research on Repairing Edge of Trimming Die with Arc Welding Robot

Research on Repairing Edge of Trimming Die with Arc Welding Robot

Photobiological hazards in shielded metal arc welding

PurposePhotobiological hazards caused by artificial optical radiation are assessed for the most commonly used arc welding technique, the SMAW (Shielded Metal Arc Welding), which operates with a hand-held system and is widely used both at occupational and domestic environments, expanding our previous investigation of a robotic arc welding process. MethodsThe complex exposure limits of the emitted blue and visible light, ultraviolet and infrared, are assessed through the European Directive 2006/25/EC, using three dedicated sensors set to measure irradiance from various typical welding procedures in the controlled environment (currents, electrodes, etc.) of a welding laboratory. In this sense, field measurements are employed, applying existing policies. Occupational limits are also applicable to the domestic welding. ResultsApart from sub-minute overexposures, which were detected in every spectral band, overexposures in the order of one second were also observed at very close distances, which are common at complex working environments. Investigation of the initial welding procedure, which is often performed without the use of Personal Protective Equipment, revealed exposure of the order of the corresponding limit. ConclusionsSimulation of a ‘bad’ welding procedure revealed increased exposure, indicating the importance of training in the occupational environment. Concern for the exposure of near-by workers (working a few meters away from the welding point) is also crucial. Future work needs to incorporate more welding techniques and measurements from original workplaces, in order to set the basis for an integrated risk assessment and provide valuable information concerning occupational diseases.

Process Simulation and Optimization of Arc Welding Robot Workstation Based on Digital Twin

For the welding cell in the manufacturing process of large excavation motor arm workpieces, a system framework, based on a digital twin welding robot cell, is proposed and constructed in order to optimize the robotic collaboration process of the welding workstation with digital twin technology. For the automated welding cell, combined with the actual robotic welding process, the physical entity was digitally modeled in 3D, and the twin welding robot operating posture process beats and other data were updated in real time, through real-time interactive data drive, to achieve real-time synchronization and faithful mapping of the virtual twin as well as 3D visualization and monitoring of the system. For the robot welding process in the arc welding operation process, a mathematical model of the kinematics of the welding robot was established, and an optimization method for the placement planning of the initial welding position of the robot base was proposed, with the goal of smooth operation of the robot arm joints, which assist in the process simulation verification of the welding process through the virtual twin scenario. The implementation and validation process of welding process optimization, based on this digital twin framework, is introduced with a moving arm robot welding example.

Neutrosophic MCDM Methodology to Select Best Industrial Arc Welding Robot

Industrial robots have made it possible for industrial companies to make goods of a good quality at lower costs. As a result, industrial robots are an integral component of sophisticated manufacturing systems. Industrial robots may be programmed to do a wide variety of tasks, including welding, painting, construction, and debugging. All of the elements are completed with an exceptional level of endurance, swiftness, and accuracy. The efficiency of industrial robots is governed by a number of different factors, some of which are in direct opposition to one another; for a strong choice approach, all of these criteria must be examined concurrently. For the purpose of selecting an industrial robot for the arc soldering process, a straightforward multi-criteria decision-making (MCDM) approach that is VIKOR method will be described in this research. The VIKOR method used to rank the robots. The results of the VIKOR methodology are provided here in the form of a priority of rating. The findings demonstrated that the MCDM strategies are highly helpful when selecting robots to utilize.

Online obstacle avoidance path planning and application for arc welding robot

• The system of online collision-free path planning for the arc welding robot is established. • An improved lazy-PRM algorithm with repulsion field is proposed to solve the path planning. • The collision-free path of the 7-DOF arc welding robot can be planned online. • The path quadratic optimization strategy is proposed to fine-tuning the pose and position of the welding gun. • Fast trajectory planning for each workpiece at any position during the welding process. The offline path planning problem for arc welding robots becomes increasingly difficult to achieve with the increasing product complexity and the open work environment, such as the welding tasks in ship formation. Most research considers the path planning problem without time limitations or other constraints that conform to the actual production process. During the welding process, the placement of workpieces seriously affects the efficiency of finding feasible paths for the welding robots for weld tasks that cannot be repeated accurately. Thus, in the flow welding workshop, the improved lazy Probabilistic Roadmap (PRM) algorithm based online collision-free path planning method is proposed to finish the welding tasks of general stiffened plates at an arbitrary position on the workbench within the robot's reach. The work environment is built with the grid method to reduce the computational dimension of collision detection. The local path planner is implemented by the lazy-PRM algorithm with the repulsive force field to rapidly search the collision-free path for the welding robot with specific safe distances. The robot posture adjustment strategy is also applied to the collision detection process to adjust some poses of the robot at candidate paths to be feasible. The improved path planning method is tested on both the simulated environment and actual application scenario and proved to be suitable for the path planning problem of the arc welding robot. The results of this study have been integrated into a computer approach, which can accelerate the motion planning process and reduce human efforts.

A sensor enabled robotic strategy for automated Defect-Free Multi-Pass High-Integrity welding

High-integrity welds found in safety–critical industries require flaw-free joints, but automation is challenging due to low-volume, often-unique nature of the work, alongside high-uncertainty part-localisation. As such, robotic welding still requires tedious manually taught paths oroffline approaches based on nominal Computer-Aided-Design (CAD). Optical and laser sensors are commonly deployed to provide online adjustment of pre-defined paths within controlled environments. This paper presents a sensor-driven approach for defect-free welding, based on the as-built joint geometry alongside the requirement for no-accurate part localisation or CAD knowledge. The approacha) autonomously localises the specimen in the scene without CAD requirement, b) adapts and generates accurate welding paths unique to the as-built workpiece and c) generates robot kinematics based on an external-control strategy. The proposed approach is validated through experiments of unconstrained placed joints, where the increased accuracy of the generated welding paths, with no common seam tracking, is validated with an average error of 0.12 mm, 0.4°. Coupling with a multi-pass welding framework, the deployment of fully automated robotic arc welding takes place for different configurations. Non-Destructive-Testing (NDT) in the form of Ultrasound-Testing (UT) inspection validates the repeatable and flaw-free nature of the sensory-driven approach, exploiting direct benefits in quality alongside reduced re-work.

A hierarchical visual model for robot automatic arc welding guidance

PurposeThe traditional vision system cannot automatically adjust the feature point extraction method according to the type of welding seam. In addition, the robot cannot self-correct the laying position error or machining error. To solve this problem, this paper aims to propose a hierarchical visual model to achieve automatic arc welding guidance.Design/methodology/approachThe hierarchical visual model proposed in this paper is divided into two layers: welding seam classification layer and feature point extraction layer. In the welding seam classification layer, the SegNet network model is trained to identify the welding seam type, and the prediction mask is obtained to segment the corresponding point clouds. In the feature point extraction layer, the scanning path is determined by the point cloud obtained from the upper layer to correct laying position error. The feature points extraction method is automatically determined to correct machining error based on the type of welding seam. Furthermore, the corresponding specific method to extract the feature points for each type of welding seam is proposed. The proposed visual model is experimentally validated, and the feature points extraction results as well as seam tracking error are finally analyzed.FindingsThe experimental results show that the algorithm can well accomplish welding seam classification, feature points extraction and seam tracking with high precision. The prediction mask accuracy is above 90% for three types of welding seam. The proposed feature points extraction method for each type of welding seam can achieve sub-pixel feature extraction. For the three types of welding seam, the maximum seam tracking error is 0.33–0.41 mm, and the average seam tracking error is 0.11–0.22 mm.Originality/valueThe main innovation of this paper is that a hierarchical visual model for robotic arc welding is proposed, which is suitable for various types of welding seam. The proposed visual model well achieves welding seam classification, feature point extraction and error correction, which improves the automation level of robot welding.

Tensile, Hardness, XRD and Surface Vonmises Stress of 316 L Stainless Steel Built by Wire Arc Additive Manufacturing (WAAM)

Wire arc additive manufacturing (WAAM) is a popular wire feed additive manufacturing technology that creates components through the deposition of material layer-by-layer. WAAM has become a promising alternative to conventional machining due to its high deposition rate, environmental friendliness, and cost-competitiveness. It is used to Fabricate complex shaped parts. The variable parameters are current, welding speed, shielding gas, and gas flow rate. This research fabricates 316 L stainless steel (WAAM plate) using a wire arc welding robot machine. Substrate and Side edges are removed using Microwire cut EDM, and the vertical milling machine finishes the surface. The tensile, hardness and X-ray Diffraction are compared with the standard 316 L stainless steel. The modelling and analysis of 316L stainless steel are carried out using COMSOL Multiphysics 5.3 software. It is concluded that the additive manufacturing of 316L stainless steel by wire and arc process is feasible.

A vision-based method for narrow weld trajectory recognition of arc welding robots

A vision-based method for narrow weld trajectory recognition of arc welding robots

A study on tensile properties and microstructural characteristics of wire arc additive manufactured low carbon steel cylindrical components

Wire arc additive manufacturing (WAAM) is an additive manufacturing (AM) technology that uses a modified robotic arc welding machine to manufacture parts in a layer-by-layer pattern. Wire-based AM has many major advantages, including good fabrication flexibility and control, as well as easier automated manufacturing. This article compares the tensile proprieties of WAAM low carbon steel cylinders made by pulsed current gas metal arc welding (PGMAW) and pulsed current cold metal transfer (PCMT) arc welding techniques. The microhardness and tensile properties of the WAAM cylindrical parts were evaluated in the bottom and top zones. The PCMT-WAAM cylinder showed higher hardness, yield strength (YS), and ultimate tensile strength (UTS) than the PGMAW-WAAM cylinder. The microstructural characteristics of the produced low carbon steel cylinders varied from one process to another process. The inhomogeneity in tensile properties and hardness is caused by variations in microstructural characteristics.

Adaptive path planning for the gantry welding robot system

Group welding is one of the crucial components with difficulties in shipbuilding with irregular distribution and a great variety of welding seams. The traditional manual teaching method has drawbacks of the long development period, low efficiency, and poor welding accuracy in dealing with such weld seams in the welding process. In order to realize automation planning and welding for arc welding robot, a gantry welding robot system with adaptive path planning ability is proposed to solve the problem with various welding workpieces and position. The 3D laser scanning sensor is first adopted to identify the type and position of the workpiece. Then, the discrete genetic algorithm based on inverted pyramid structure (GA-IPS) is applied for welding sequence planning, the discrete two-stage adaptive reference direction guided evolutionary algorithm with modified dominance relation for many-objective optimization (AREA-APD) is proposed for dual-objective welding sequence planning. Furthermore, the lazy probabilistic roadmap method with transition points (Lazy PRM-TP) is applied to local trajectory planning. And the laser vision system is used to realize the precise location and tracking of the welding seam. Besides, the computer system is designed to realize human-computer interaction and process monitoring. The simulation and experiments show the superior performance of the proposed strategy.

로봇 아크용접에서 용접선 추적시스템에 대한 연구

During the robotic arc welding process, not only are the images taken often unclear, but the features are also not obvious due to the interference of arc, splashing, and smoke. Therefore, it is necessary to take these visual interferences into account when designing the imaging processes and algorithms for welding inspections. The objective of this study was to develop intelligent and cost-effective algorithms for image processing techniques in robotic arc welding procedures using an infrared sensor. A set of image processing algorithms were designed to detect the geometry of bead images by analyzing experimental images obtained from an infrared camera. Pre-processing procedures were required to improve the contrast of these images. After verifying their effectiveness, the proposed algorithms were compared to determine the optimal algorithm for each step in the image-processing procedure. The experimental results indicate a small error and suggest that these algorithms can be employed in the robotic welding process.

Kinematic Modeling of 3P2R Welding Robot Based on D-H Parameters

To meet the demand of different vehicle models welding task, as well as lower the cost of specialized welding facility, a kind of 3P2R arc welding robot has been developed and SolidWorks is used to set up the Three Dimensional (3D) solid model. The developed robot includes 3P2R body, controller, and a welding power supply. The direct kinematical model and inverse solution have been worked out basing on Denavit-Hartenberg (D-H) method, and the direct kinematical model and inverse solution are verified by experiment. This robot could also be used to burring or spray field with the actuator changed.