ProgramMaster Logo
Conference Tools for 2020 AWS Professional Program
Login
Register as a New User
Help
Submit An Abstract
Propose A Symposium
Presenter/Author Tools
Organizer/Editor Tools
About this Abstract
Meeting 2020 AWS Professional Program
Symposium 2020 AWS Professional Program
Presentation Title Analysis of Dynamic Evolution of Weld Pool for the Real-Time Monitoring of Penetration Using Deep Learning
Author(s) Yongchao Cheng, YuMing Zhang, Shujun Chen, Jun Xiao
On-Site Speaker (Planned) Yongchao Cheng
Abstract Scope Gas Tungsten Arc Welding (GTAW) as one of the major welding methods has been used widely in industrial manufacture, especially for aerospace and pressure vessels, etc. For these critical applications, the sensing and control of penetration state are required to complete the high-quality welding joint, in which the backside width is a key factor to present the penetration state. In the practical manufacture process, the backside of workpiece is not available due to space limitation. Therefore, how to sense welding process information plays a key role in the automatic and intelligent control of welding processes. There are so many sensing methods including voltage sensing, spectrum sensing, ultrasonic sensing, infrared sensing, weld pool oscillation sensing, and vision sensing, etc. In the traditional welding process, the welders observe the surface of weld pool and control the welding speed and current to complete the designed penetration state according to their professional experience, so the topside information of weld pool has a strong tie with backside width. Hence, we select the vision sensing method to simulate the welder’s observation, in which the vision sensing method can be divided into passive sensing and active sensing. For passive sensing, the intensive disturbance from the welding process like high-intensity arc brightness, spatter and smoke can reduce the prediction accuracy, so the active vision is used as the sensing method to monitor the penetration state. The simplified and clear information can help us complete the identification efficiently, and thus an active sensing method is developed at the University of Kentucky using a single stripe laser pattern to extract the topside information of weld pool. The feature in the reflected image is not easily extracted for us using the traditional approach. Luckily, deep learning has a great ability for non-linear fitting and generalization which is a perfect and strong tool to solve this problem. A convolutional neural network model is designed to training and predict the penetration state or backside width of weld pool. In this research, a series of GTAW experiments using random weld current and speed in a certain range to make the full penetration have adopted and conducted to acquire dataset. The welding experimental platform contains the welding system, sensing system and control system. For the sensing system, a laser generator, an imaging plane, and two cameras (topside and backside) are installed around the weld torch on the same plane. The laser pattern generated from the lase generator is projected onto the topside surface of weld pool, and the reflected laser will be received by the imaging plane and captured by the camera. The topside camera acquires the characteristics of the topside surface of weld pool as the feature, and the backside camera is used to sense the brightness of backside weld pool as the label in the CNN model. In this experimental platform, the welding current control, welding speed control and sensing of two-side images are conducted and completed using personal computer that is based on the Python programming environment. The dynamic penetration status of weld pool in GTAW is studied and divided into many stages such as part penetration, critical penetration, full penetration and excessive penetration. From the analysis of experiments and physical model, the results show that active vision images can establish an obvious relationship with the penetration/backside width of the weld pool and establish the theory foundation for real-time intelligent control. The application of deep learning (CNN) can calculate the all characteristics of vision sensing. The designed model of CNNs showed that the error of test data is 0.497 mm after processing the predicted width using the moving average method and the model can process 61.79 data pairs per second. Therefore, this research can realize and complete the innovation and application of information sensing in the field of real-time high-quality penetration control. Keywords:Intelligent Welding, Active Vision, Penetration Width, Prediction, Deep Learning
Proceedings Inclusion? Definite: Other (describe below)

OTHER PAPERS PLANNED FOR THIS SYMPOSIUM

101st Year of the American Welding Society: Changes in Welding & Outlook
Additive Manufacture of the First Concentric Reduction Sleeve Used in the Brazilian Oil and Gas Industry
Additive Manufacturing Using Semi-Solids
Advances in Robotic Arc DED AM for Shipbuilding
An Efficient Part-Scale Model to Predict Distortion on Overhang Structure in Laser Powder Bed Fusion Additive Manufacturing
An Overview of LME Cracking in Advanced High Strength Automotive Steels
Analysis and Control of Humping Defect in Laser Wire Direct Deposition Process
Analysis of Dynamic Evolution of Weld Pool for the Real-Time Monitoring of Penetration Using Deep Learning
Avoiding Tool Failure and Defects in Friction Stir Welding Using Machine Learning
Bondline Properties in High Frequency Welding of Steels
Building a Digital Twin for Welding Process Monitoring, Visualization and Control Based on Deep Learning
Characterization of the Fourth Generation of Nickel-Chromium Welding Products for Nuclear Service
Chipping Study of Friction Element Welding
Computational Modeling on Defect Formation during Self-Reacting Friction Stir Welding
Copper Foil Welding for Battery Manufacturing
Correlating Laser Power and Microstructural Properties of Additively Manufactured Ti-6Al-4V Fabricated by Directed Energy Deposition
Creep Strength Variation in Simulated ICHAZs of Grade 91 Steel
Design of an Austenitic Steel Weldment System Using ICME
Development of Industry 4.0 Software for Arc-Welding Energy Monitoring and Traceability
Development of Interlayer Technology to Join Advanced Materials for the Transportation Industry
Development, Optimization, and Mechanical Testing of ER80S-G Girth Welds on Internally Clad API 5L Grade X65 Pipes
Dissimilar Joining of Carbon Fiber Reinforced Composite to Magnesium Alloy by Friction Self-Piercing Riveting Process
Effect of Aluminum on Primary Carbides in Chromium Carbide Overlays
Effect of Chemical Composition of Welding Consumable on Slag Formation and Weld Corrosion Resistance
Effect of Heating Rate on Austenite Formation in a Low Carbon Microalloyed Steel
Effect of Phase Transformations on Mechanical Properties of Ti Free Grade 300 Maraging Steel Manufactured by Laser Powder Bed Fusion (LPBF)
Electron Beam Brazing of Stainless Steel using NIORO Filler Metal
Evaluation of Braze Joints for Hydrogen Purification Diffuser
Experimental Studies of Wire Feeding Electron Beam Welding-Brazing of Aluminum to Stainless Steel and Titanium
Failure Analysis: Evaluating a Low Strength Solder Joint
Fundamentals and Techniques for High-Speed Imaging of Welding
GMA Weldability Between AISI 316L Pipes Manufactured by Conventional and Additive Manufacturing Processes
High Deposition Gas Metal Arc Variants for Directed Energy Deposition Additive Manufacturing
High Efficient Modification of Arc Welding Processes via Modeling and Sensing Weld Pool Behaviors
High Penetration Buried Arc Gas Metal Arc Welding for Shipyard Steel Plate Fabrication
Hybrid Laser Welding on API 5L Pipeline Steels
Hyper-Duplex Stainless Steel Intermetallic Precipitation Behavior during Cladding
Improper Weld Field Fabrication Caused Premature Failure
In-situ Measurement and Numerical Simulation for Linear Friction Welding of Complex Titanium Structures
In-situ Synchrotron Diffraction Measurement of Solidification Behavior during Laser Welding of Multi-Principal-Component Alloy
Induction Heating-Assisted Friction Stir Welding in Low Carbon Steel Plates
Influence of Composition on the Solidification and Mechanical Properties of HP-Modified Heat-Resistant Austenitic Stainless Steels
Influence of Cooling Rate on Microstructure Formation in Rapid Solidification of Ni2MnGa Alloy
Influence of Hydrogen on the Softened HAZ in Various Carbon Steel Welds
Integrated Computational Materials Engineering (ICME) Techniques to Enable a Material-Informed Digital Twin Prototype for Marine Structures
Integrated Modeling of Multi-Process Automotive Lap Joining – Part 3: Microstructure and Mechanical Properties
Investigation of a Curly Toolpath in Friction Stir Welding
Investigation of Laves Phase Formation in Inconel 718 Fabricated by Laser Powder Bed Fusion
Investigation of Thermally Assisted Friction Element Welding
Joining Dissimilar Metals of DP590 Steel and AZ31B Magnesium Sheets by Ultrasonic Spot Welding
Low-cycle Fatigue Evaluations of Ni-Steel Dissimilar Joints for Coke Drum Welding Repairs
Machine Learning for Automatic Recognition of Microstructures
Material Characteristics of Wire Arc Additive Manufactured Inconel 718
Mechanical Properties Assessment of Carbon and Low-Alloy Steel Parts Built by GMA-DED
Mechanics of the Solid-State Bonding and Prediction of Thermomechanical Responses in Friction Stir Welding
Mechanistic Models and Machine Learning to Mitigate Common Defects in Metal Printing
Metallic Powder Core Tubular Wire Development for Additive Manufacturing
Microstructural and Mechanical Property Characterization of Reaction Synthesis Aluminum Metal Matrix Composites Produced by Additive Manufacturing
Microstructural Characterization and Mechanical Property Evaluation of Fusion Zones in a 10 wt% Ni Steel Weld Metal
Microstructure and Mechanical Properties of 410 Stainless Steel Using Different Shielding Gases in the Wire-Arc Additive Process
Microstructure and Mechanical Properties of Intercritical-Treated Grade 91 Steel
Microstructures and Microhardness of LENS DED Deposits of Ti/TiC MMCs
Mitigation of Liquid Metal Embrittlement in Galvannealed AHSS Welds Using High Entropy Alloy Filler
Modal Analysis of Ultrasonic Welding to Enable Multi-Spot Dissimilar Material Joining
Modeling and Comparing Human Welder’s Operation Between Stereo Camera and Virtual Reality
Modeling of In-Situ Tempering of a Creep Resistant Ferritic Martensitic Steel during Multi-Layer Additive Manufacturing
Modeling of Mash Seam Welding Using Improved Electro-Thermo-Mechanical Simulation
Modelling of Laser Cladding Using Gaussian Heat Source Profile and Verification with Experimental Results
Multi-Scale and Multi-Physics Modeling of Process-Microstructure-Property Relationship in Metal Additive Manufacturing
New Approaches in Friction Welding Advanced PM Nickel Base Superalloys
Numerical Study on Heat Transfer Mechanisms in Electric Arc Columns
Optimization of Tool Path and Microstructure in Large Scale Metal Additive Manufacturing with Multi-Heat Sources
Phase Transformation Analysis and Microstructural Characterization of the Heat Affected Zone in Grade 92 Steel Welds
Procedure Qualification Schemes and Build Results for Directed Energy Deposition Additive Manufacturing
Process Control Using Predictive Equations for Zero Programming Laser Cladding Facility
Process, Microstructure and Fracture Mode of Magnesium to Steel Dissimilar Metal Spot Joints
Pulsed-Arc Welding of Battery Tabs for Vehicle Electrification
PWHT of 347SS Weldments for Thermal Energy Storage Concentrating Solar Power Applications
Real-Time Prediction of Weld Penetration from Dynamic Weld Pool-Arc Images Deep Learning Based
Recent Developments in Metal Additive Manufacturing
Research and Development of a Novel TIG Welding Technology for Joining Thin Sheets Applying in Metal Forming Field
Residual Stress Measurement on High-Strength Steel Panels for Shipbuilding Application
Resistance Spot Welding Challenges, Root Causes, and Suggested Welding Practices When Welding 3rd Generation AHSS
Role of Retained Austenite in Mechanical Response of Additively Manufactured 17-4 PH Stainless Steel
SMART Camera in the Welding Torch for Manual Welding Quality and Productivity
State-of-the-Art of Underwater Wet Welding Practice
Strain Ageing of Inconel 740H and 347H
Suitability of Using Thin Wire in Wire Arc Additive Manufacturing (GMA-DED)
Sulfur Implantation in Alloy 690 for the Study of Ductility-Dip Cracking
Surface Modification via Metal Deposition and Selective Alteration
Surface Treatment of 3-D Printing Parts
Temper Bead Welding
The Correlation of Hardness to Toughness and the Superior Low Temperature Impact Properties of Martensite in RPV Steels Applied to Temper Bead Qualification
The Role of Inclusions in Duplex Stainless Steels Produced Using Laser-Based Directed Energy Deposition Additive Manufacturing
Ultrasonic Vibration Assisted Electron Beam Additive Manufacturing
Unit Block-Based Process Planning Strategy of WAAM for Complex Shell-Shaped Component
Universal Representation of Arc Shape and Arc Column Characteristics for DC Electric Arcs Burning in Argon and Helium
Variation in Microstructure and Properties in the Heat Affected and Fusion Zones of Low-Density Fe-Mn-Al-C Steel Welds
Variations in Microstructure and Mechanical Properties with Solidification Mode in L-PBF 316L
Weld Metal Cooling Rate-Mechanical Property Response of Fe-10Ni Steel Gas Metal Arc Weld Deposits and Other Experimental Observations
Weldability and Microstructural Evolution of Fusion Welds in a γ’ Strengthened TRIP Steel
Welding and Joining Challenges in the Astrospace Industry
Welding of FeMnAl High-Manganese Lightweight Steel
Wetting and Low Temperature Bonding of Zirconia Metallized with Ti-Containing Active Solders

Questions about ProgramMaster? Contact programming@programmaster.org