Skip to main content

About this book

The primary aim of this volume is to provide researchers and engineers from both academia and industry with up-to-date coverage of recent advances in the fields of robotic welding, intelligent systems and automation. It gathers selected papers from the 2017 International Workshop on Intelligentized Welding Manufacturing (IWIWM’2017), held June 23-26, 2017 in Shanghai, China. The contributions reveal how intelligentized welding manufacturing (IWM) is becoming an inescapable trend, just as intelligentized robotic welding is becoming a key technology. The volume is divided into four main parts: Intelligent Techniques for Robotic Welding, Sensing in Arc Welding Processing, Modeling and Intelligent Control of Welding Processing, and Intelligent Control and its Applications in Engineering.

Table of Contents


Feature Articles


Intelligent Weld Manufacturing: Role of Integrated Computational Welding Engineering

A master welder uses his sensory perceptions to evaluate the process and connect them with his/her knowledge base to take the necessary corrective measures with his/her acquired skills to make a good weld. All these actions must take place in real time. Success depends on intuition and skills, and the procedure is labor-intensive and frequently unreliable. The solution is intelligent weld manufacturing. The ultimate goal of intelligent weld manufacturing would involve sensing and control of heat source position, weld temperature, weld penetration, defect formation and ultimately control of microstructure and properties. This involves a solution to a problem (welding) with many highly coupled and nonlinear variables. The trend is to use an emerging tool known as intelligent control. This approach enables the user to choose a desirable end factor such as properties, defect control, or productivity to derive the selection of process parameters such as current, voltage, or speed to provide for appropriate control of the process. Important elements of intelligent manufacturing are sensing and control theory and design, process modeling, and artificial intelligence. Significant progress has been made in all these areas. Integrated computational welding engineering (ICWE) is an emerging field that will aid in the realization of intelligent weld manufacturing. The paper will discuss the progress in process modeling, microstructure, properties, and process control and automation and the importance of ICWE. Also, control and automation strategies for friction stir welding will be discussed.
S. A. David, Jian Chen, Brian T. Gibson, Zhili Feng

Research Papers


A Reinforcement Learning Based Approach for Welding Sequence Optimization

We develop and implement a Q-learning based Reinforcement Learning (RL) algorithm for Welding Sequence Optimization (WSO) where structural deformation is used to compute reward function. We utilize a thermomechanical Finite Element Analysis (FEA) method to predict deformation. We run welding simulation experiment using well-known Simufact® software on a typical widely used mounting bracket which contains eight welding beads. RL based welding optimization technique allows the reduction of structural deformation up to ~66%. RL based approach substantially speeds up the computational time over exhaustive search.
Jesus Romero-Hdz, Baidya Saha, Gengis Toledo-Ramirez, Ismael Lopez-Juarez

Time-Optimal Path Planning for Dual-Welding Robots Based on Intelligent Optimization Strategy

Dual-welding robots are widely used with the industry development, and dual-welding robots usually have to deal with a large number of weld joints. In this condition, traditional manual teaching method is time-consuming and inefficient. In this paper, an intelligent optimization strategy is proposed to realize time-optimal path planning for dual-welding robots. First, the welding robot path optimization problem is presented. Then, good diversity and convergence velocity of discrete group competition particle swarm optimization (GC-PSO) algorithm are tested. Compared with particle swarm optimization (PSO), genetic particle swarm optimization (GPSO) and chaos particle swarm optimization (CPSO) algorithms, GC-PSO algorithm shows its better optimization effectiveness. In addition, a method of collision detection and obstacle avoidance is given. At last, an intelligent optimization strategy is applied to time-optimal path planning for dual-welding robots, and the global optimal result can be obtained quickly. Simulation results show that the intelligent path planning strategy is effective and can be used for welding robot path optimization.
Xuewu Wang, Bin Tang, Yixin Yan, Xingsheng Gu

Improving Stability of Welding Model with ME-ELM

Welding shape is important in evaluating welding quality, but accurate predictive model is hard to achieve, because welding is a complex nonlinear process, and the sampled data are inevitably contaminated. Extreme learning machine (ELM) is used to construct a single-hidden layer feedforward network (SLFN) in our study, for improving stability of welding model, M-estimation is combined with ELM and a new algorithm named ME-ELM is developed; researches indicate that it works more effective than BP and other variants of ELM in reducing influence, furthermore, it can improve the model’s anti-disturbance and robustness performance even if the data are seriously contaminated. Real TIG welding models are constructed with ME-ELM, by comparing with BP, multiple nonlinear regression (MNR), and linear regression (LR), conclusions can be gotten that ME-ELM can resist the interference effectively and has the highest accuracy in predicting the welding shape.
Jianxiong Ye, Han Ye, Zhigang Li, Xingling Peng, Jinlan Zhou, Bo Guo

Study on the Cracks of NiTiNb/TC4 Lap Joints Welded by Micro Laser Welding

Dissimilar metal sheets of Ni47Ti44Nb9 (at.%, short for NiTiNb) shape memory alloy and Ti6Al4V (TC4) alloy with the same thickness of 0.2 mm are joined by using micro laser welding technology. The microstructure and crack in the weld are studied by optical microscope (OM) and scanning electron microscope (SEM). The results show that the cracks are easy to generate during the laser lap welding of NiTiNb alloy and TC4 alloy due to plenty of brittle intermetallic compound of Ti2Ni. However, the welding cracks can significantly reduce or disappear by adding filler metal of Ni foil. No-defect joint is obtained when NiTiNb alloy is located on the upper and 50 μm thick Ni foil is added during the lap welding process. The crack sensitivity can be decreased because of finer grains and elliptic boundaries. The shear load of the joint can reach 162 N.
Yuhua Chen, Zilin Zhan, Yuqing Mao, Yangyang Yu, Weiwei Lu

Research on the Ultrasonic Welding of Titanium Alloy After Embedding Fiber Bragg Grating Sensor

Fiber Bragg grating (FBG) sensor is a preferred carrier for information transmission and sensing of smart metal structures. The electroplated nickel FBG is embedded in a direct or indirect way and welded by ultrasonic welding to research the rapid prototyping and sensing properties of the titanium alloy intelligent structural parts. The experiment of embedding electroplated nickel FBG into titanium alloy in the direct way shows that titanium alloy is not suitable for embedding matrix. The experiment of embedding electroplated nickel FBG in the indirect way shows that the figure of FBG temperature sensitivity is 2.13 times larger than that of original bare fiber grating, and is 1.11 times larger than that of direct way. This fact means that embedding metallized FBG into the titanium alloy structure in the indirect way is an effective way.
Zhengqiang Zhu, Qiankun Xiao

Analysis of Vacuum Chamber Structure Based on Visual Finite Element Modeling

The feasibility of improving the vacuum chamber structure of vacuum packaging machine is discussed. The finite element software, Visual Environment, is used to simulate the structures of prime vacuum chamber and improved vacuum chamber which is built most through weld process. The results show that the maximum weld residual stress of the improved vacuum chamber changes and the stress concentration is greatly reduced with finite element analysis (FEA). At the same time, a path for a particular site is created and the range of stress changes is found from the original 270–340 to 240–310 MPa. The improved method not only saves resources but also improves product quality, which is verified by the actual product.
Yanhu Wang, Xizhang Chen

Single-Channel Blind Source Separation and Its Application on Arc Sound Signal Processing

Welding arc sound signal is an important signal in intelligent welding diagnosis, due to its informative, noncontact, easy collected. However, due to the interference of ambient noise, the arc sound signal is highly complex and noisy, which seriously limits the application of arc sound signals. In this paper, a single-channel blind source separation (BSS) algorithm based on the ensemble empirical mode decomposition (EEMD) is proposed to purify and denoise the arc sound signals. First, EEMD is used to decompose one channel signal to several intrinsic mode functions (IMFs). Second, principal component analysis (PCA) is used to reduce the multidimension IMFs to low-dimension IMFs, which are regarded as the virtual multichannels signals. Finally, independent component analysis (ICA) separates the virtual multichannels signals into target sources. The approach was tested by simulation and experiments. The simulated results show that signals separated from mixed signal using this approach highly match the source signals that make up the mixed signal. Moreover, experimental results indicated that the source signals of arc sound were effectively separated with the environmental noise signals. The statistical characteristics of the spectrum in 5–6.5 kHz band extracted from the arc sound source signals can accurately identify the two types of weld penetrations.
Wenjing Ren, Guangrui Wen, Riwei Luan, Zhe Yang, Zhifen Zhang

Investigation on Surface Quality in a Hybrid Manufacturing System Combining Wire and Arc Additive Manufacturing and Machining

Wire and arc additive manufacturing (WAAM) has gained popularity in recent years due to its unique efficiency and cost advantages. Nevertheless, due to the stair-stepping effect and the liquidity of molten metal, the achieved geometric accuracy and surface quality are still very limited. The combination of WAAM and machining, namely hybrid manufacturing, provides a fundamental solution to the above problem. Because machining is performed after depositing several layers, the deposition width, deposition height, and surface waviness have great effects on the machined surface quality, in addition to the machining parameters including spindle speed and feedrate. In this paper, the dependence of the machined surface quality (characterized by surface roughness) on the influencing factors mentioned above is investigated based on quadratic general rotary unitized design (QGRUD). To reduce the number of experiments, a comprehensive factor, namely material removal area (MRA), is introduced to characterize the deposition width, deposition height, and surface waviness. The analysis results show that spindle speed is the most influential factor, followed by MRA and feedrate. Furthermore, a high spindle speed and a moderate feedrate are preferred, which contribute to not only improving the surface quality and the efficiency but also reducing the demand of geometric accuracy for WAAM.
Fang Li, Shujun Chen, Junbiao Shi, Hongyu Tian

Preliminary Research on Intelligent Mobile Tool Cart for Industrial Manufacturing in a Factory Environment

Intellectualization of manufacturing is a general trend due to the development of technology and science. This paper presents a preliminary research on intelligent mobile tool cart (IMTC) for working alongside workers in manufacturing process. Tool cart problem is considered as an exemplary intelligent mobile tool cart service system. A type of differential-driven mobile cart with a tool basket which is mounted on the top of the mobile cart compose the intelligent mobile tool cart. The IMTC is designed to provide workers with remote motion control and tools positioning services. Daily used instructions are exercised as commands to control the IMTC in the factory environment by using Leap Motion sensor for finger/hand position detection. The IMTC provides its real-time position information to the worker. To fulfill estimating the location of IMTC, QR localization method is proposed. We verified the features of the IMTC and feasibility of the proposed localization method through experimental trials.
Shengnan Gai, Qiang Luo, Shujun Chen

Microstructure and Mechanical Properties of Friction Stir Weld of Dissimilar Ti6Al4V Titanium Alloy to AA2024 Aluminum Alloy

Dissimilar Ti6Al4V titanium alloy and AA2024 aluminum alloy sheets with a thickness of 3 mm were friction stir welded successfully, and the microstructure and mechanical properties of the butt joints were investigated. The results show that: the stirred zone (SZ) exhibits a mixture structure, which is characterized by fine recrystallized grains of aluminum alloy and titanium particles. Unfilled defects are observed among titanium particles in SZ. Moreover, at the aluminum side the thermal-mechanically affected zone (TMAZ) and the heat affected zone (HAZ) appeared like in the traditional FSW-joints. But, at the titanium side, a recrystallization band with a width of 6–10 μm and a layer with fibrous structures are found at the joint interface. Also, the Ti–Al compounds layer with some micro-cracks is presented in the fibrous structure. The hardness distribution of the joint along the cross-section centerline varies significantly due to the existence of different broken titanium particles. The ultimate tensile strength (UTS) of the joint reached 83% of aluminum base metal and the joint failed with a ductile–brittle fracture mode.
Yuhua Chen, Wenming Cao, Shuhan Li, Chao Chen, Jilin Xie

Control of Current Waveform for Pulsed MIG Welding of Aluminum Alloy Sheets

In this chapter, a simplified model of pulsed MIG welding is established and simulated using MATLAB. In addition, the anti-interference of a current waveform is simulated using the adaptive neural network feedforward control. Depending on the “one droplet per pulse” relationship in pulse frequency modulation, faster adjustment of the arc length, compared with the adjustment of the constant current characteristic, is achieved by increasing or decreasing the base time of the average current.
Min Xu, Jiaxiang Xue

Short Papers and Technical Notes


Study of Ultrasonic Phased Array in Underwater Welding

The popular sensors used in water have a lot of difficulties when compared with those used in land welding automation. It is meaningful to find a new kind of sensor which is suitable to work in underwater welding. Ultrasonic phased array (PA) can work in water conveniently and send out required sound beam. Compared with single-ultrasonic sensor, PA works faster and effectively. In this chapter, interference principle of acoustic beam is analyzed first. Then, the relationship of focusing precision, PA shape, gap distance between adjacent units, sensor element number, and time resolution is revealed with simulation. Proper physical parameters of PA are determined. At last, high time resolution circuit based on complex programmable logic device (CPLD) is given out. It works together with sound-emitting and -receiving circuits to realize fast scan on welding workpiece, so that the seam line can be deduced with underwater distance detection.
Jianxiong Ye, Zhigang Li, Xingling Peng, Jinlan Zhou, Bo Guo

Type Identification and Feature Extraction of Weld Joint for Adaptive Robotic Welding

In recent years, intelligent robotic welding has been an active research area. Vision sensors have been widely used in robotic welding systems for information collection and processing. For better welding quality and efficiency, it is necessary to achieve accurate and fast information processing and intelligent decision-making for welding robot. For weld joint information processing, most of the reported works focus on the feature extraction of weld joint concerning a specific type or a regular shape. In this chapter, an algorithm is proposed to identify joint type and extract relevant feature values by extracting three feature lines and two key turning points. Three types of weld joints are inspected and the results indicate that the algorithm is of high efficiency and robustness.
Ran Li, Manshu Dong, Xiaochao Zhang, Hongming Gao

Kinematic Model Analysis of an 8-DOF Photographic Robot

The photographic robot studied in this chapter is an 8-DOF PRRPR-S type. In order to obtain a stable and repeatable lens shooting trajectory, it is necessary to balance the robot’s own weight and structural rigidity. First, based on the structural analysis of the photographic robot, the main parameters of the mechanism design are extracted. Then, with the help of the photographic robot calibration, the kinematics model of the robot is established. The DH model is applied on the first seven axes of the photographic robot. The 6-parameter model is used on the end actuator attitude adjustment rotation axis. Finally, the motion of each axis is simulated in MATLAB, which verifies the kinematic model.
Xiaowei Xie, Xingang Miao, Su Wang, Feng Zhang


Additional information