Optimal design of the welding procedure gives the desired welding results under nominal welding conditions. During manufacturing, where the actual welding manufacturing conditions often deviate from the nominal ones used in the design, applying the designed procedure will produce welding results that are different from the desired ones. Adaption is needed to make corrections and adjust some of the welding parameters from those specified in the design. This is adaptive welding. While human welders can be adaptive to make corrections and adjustments, their performance is limited by their physical constraints and skill level. To be adaptive, automated and robotic welding systems require abilities in sensing the welding process, extracting the needed information from signals from the sensors, predicting the responses of the welding process to the adjustments on welding parameters, and optimizing the adjustments. This results in the application of classical sensing, modeling of process dynamics, and control system design. In many cases, the needed information for the weld quality and process variables of our concern is not easy to extract from the sensor’s data. Studies are needed to propose the phenomena to sense and establish the scientific foundation to correlate them to the weld quality or process variables of our concern. Such studies can be labor intensive, and a more automated approach is needed. Analysis suggests that artificial intelligence and machine learning, especially deep learning, can help automate the learning such that the needed intelligence for robotic welding adaptation can be directly and automatically learned from experimental data after the physical phenomena being represented by the experimental data has been appropriately selected to make sure they are fundamentally correlated to that with which we are concerned. Some adaptation abilities may also be learned from skilled human welders. In addition, human-robot collaborative welding may incorporate adaptations from humans with the welding robots. This paper analyzes and identifies the challenges in adaptive robotic welding, reviews efforts devoted to solve these challenges, analyzes the principles and nature of the methods behind these efforts, and introduces modern approaches, including machine learning/deep learning, learning from humans, and human-robot collaboration, to solve these challenges.
{"title":"Adaptive Intelligent Welding Manufacturing","authors":"Yuming Zhang, Qiyue Wang, Yukang Liu","doi":"10.29391/2021.100.006","DOIUrl":"https://doi.org/10.29391/2021.100.006","url":null,"abstract":"Optimal design of the welding procedure gives the desired welding results under nominal welding conditions. During manufacturing, where the actual welding manufacturing conditions often deviate from the nominal ones used in the design, applying the designed procedure will produce welding results that are different from the desired ones. Adaption is needed to make corrections and adjust some of the welding parameters from those specified in the design. This is adaptive welding. While human welders can be adaptive to make corrections and adjustments, their performance is limited by their physical constraints and skill level. To be adaptive, automated and robotic welding systems require abilities in sensing the welding process, extracting the needed information from signals from the sensors, predicting the responses of the welding process to the adjustments on welding parameters, and optimizing the adjustments. This results in the application of classical sensing, modeling of process dynamics, and control system design. In many cases, the needed information for the weld quality and process variables of our concern is not easy to extract from the sensor’s data. Studies are needed to propose the phenomena to sense and establish the scientific foundation to correlate them to the weld quality or process variables of our concern. Such studies can be labor intensive, and a more automated approach is needed. Analysis suggests that artificial intelligence and machine learning, especially deep learning, can help automate the learning such that the needed intelligence for robotic welding adaptation can be directly and automatically learned from experimental data after the physical phenomena being represented by the experimental data has been appropriately selected to make sure they are fundamentally correlated to that with which we are concerned. Some adaptation abilities may also be learned from skilled human welders. In addition, human-robot collaborative welding may incorporate adaptations from humans with the welding robots. This paper analyzes and identifies the challenges in adaptive robotic welding, reviews efforts devoted to solve these challenges, analyzes the principles and nature of the methods behind these efforts, and introduces modern approaches, including machine learning/deep learning, learning from humans, and human-robot collaboration, to solve these challenges.","PeriodicalId":23681,"journal":{"name":"Welding Journal","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"70003940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yu Jiang, Hongtao Zhang, P. He, X. Yang, Teng Yao, Qichen Wang, Liqin Wei, Z. Wenjie
Low-carbon steel Q235B was successfully joined by plasma-pulsed gas metal arc welding (plasma-GMAW-P) with an external magnetic field. The arc profile, temperature field, electrical signal, microstructure, and mechanical properties of this method were analyzed. The results indicated that the coupling degree of the two arcs increased with the strengthening of the magnetic field current. However, when the magnetic field current was greater than 1 A, the arc pro-file changed slightly with the increase of the magnetic field current. Fixed on the magnetic field current, the coupling degree first increased and then decreased with the increase of the plasma welding current, GMAW-P welding current, plasma gas flow rate, and nozzle height, respectively. The maximum temperature had no obvious influence on joint penetration at different magnetic field cur-rents. However, the average temperature had an inverse effect on joint penetration at different magnetic field currents. The weld fusion zone joint tensile test results showed that the ratio of depth to width increased with the application of magnetic field currents. Moreover, tensile strength on the upper and lower part of the tensile samples were 521 and 488 MPa, respectively, which were 4.6% and 3.2% higher than those without the magnetic field. The microhardness of the weld joints was higher than that without the magnetic field.
{"title":"Arc Characteristics and Welding Process of Magnetic Field Assisting Plasma-GMAW-P","authors":"Yu Jiang, Hongtao Zhang, P. He, X. Yang, Teng Yao, Qichen Wang, Liqin Wei, Z. Wenjie","doi":"10.29391/2021.100.001","DOIUrl":"https://doi.org/10.29391/2021.100.001","url":null,"abstract":"Low-carbon steel Q235B was successfully joined by plasma-pulsed gas metal arc welding (plasma-GMAW-P) with an external magnetic field. The arc profile, temperature field, electrical signal, microstructure, and mechanical properties of this method were analyzed. The results indicated that the coupling degree of the two arcs increased with the strengthening of the magnetic field current. However, when the magnetic field current was greater than 1 A, the arc pro-file changed slightly with the increase of the magnetic field current. Fixed on the magnetic field current, the coupling degree first increased and then decreased with the increase of the plasma welding current, GMAW-P welding current, plasma gas flow rate, and nozzle height, respectively. The maximum temperature had no obvious influence on joint penetration at different magnetic field cur-rents. However, the average temperature had an inverse effect on joint penetration at different magnetic field currents. The weld fusion zone joint tensile test results showed that the ratio of depth to width increased with the application of magnetic field currents. Moreover, tensile strength on the upper and lower part of the tensile samples were 521 and 488 MPa, respectively, which were 4.6% and 3.2% higher than those without the magnetic field. The microhardness of the weld joints was higher than that without the magnetic field.","PeriodicalId":23681,"journal":{"name":"Welding Journal","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"70004042","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The use of Type 16-8-2 filler metal was examined for application in structural welds on 304H and 347H stainless steels for high-temperature service applications and compared to welds with matching filler metals 308H and 347, respectively. Microstructural stability during elevated temperature exposure, weld metal impact properties, and susceptibility to stress-relief cracking were examined. It was found that the lean composition and low ferrite (~ 2 Ferrite Number [FN]) in 16-8-2 weld metal provide high resistance to intermetallic phase formation. No hot cracking was observed despite the low ferrite level. The 16-8-2 weld metals displayed superior toughness as compared to the matching filler metal welds, especially after longer elevated-temperature exposure. Experimental evidence for some martensite transformation in aged 16-8-2 weld metal upon cooling to ambient temperature was presented and explained an increase in magnetic response (as FN) after postweld heat treatment at 1300 ̊F (705 ̊C). None of the tested weld metals failed by stress-relief cracking mechanisms under the applied test conditions. The 16-8-2 filler metal welds exhibited significantly lower levels of stress relief during high-temperature exposure and significantly higher tensile strength after high-temperature hold as compared to the matching filler metal welds.
{"title":"Filler Metal 16-8-2 for Structural Welds on 304H and 347H Stainless Steels for High-Temperature Service","authors":"C. Fink, Huimin Wang, B. Alexandrov, J. Penso","doi":"10.29391/2020.99.029","DOIUrl":"https://doi.org/10.29391/2020.99.029","url":null,"abstract":"The use of Type 16-8-2 filler metal was examined for application in structural welds on 304H and 347H stainless steels for high-temperature service applications and compared to welds with matching filler metals 308H and 347, respectively. Microstructural stability during elevated temperature exposure, weld metal impact properties, and susceptibility to stress-relief cracking were examined. It was found that the lean composition and low ferrite (~ 2 Ferrite Number [FN]) in 16-8-2 weld metal provide high resistance to intermetallic phase formation. No hot cracking was observed despite the low ferrite level. The 16-8-2 weld metals displayed superior toughness as compared to the matching filler metal welds, especially after longer elevated-temperature exposure. Experimental evidence for some martensite transformation in aged 16-8-2 weld metal upon cooling to ambient temperature was presented and explained an increase in magnetic response (as FN) after postweld heat treatment at 1300 ̊F (705 ̊C). None of the tested weld metals failed by stress-relief cracking mechanisms under the applied test conditions. The 16-8-2 filler metal welds exhibited significantly lower levels of stress relief during high-temperature exposure and significantly higher tensile strength after high-temperature hold as compared to the matching filler metal welds.","PeriodicalId":23681,"journal":{"name":"Welding Journal","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43624890","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Junfei Wu, Yanfei Han, C. Jia, Qingyuan Yang, Chuansong Wu
Underwater pulse-current wet welding was proposed in part 1 of this two-part report. The novel technology obtained improved metal transfer and welding process stability. The main reason for droplet oversizing and long transfer cycles was found to be the deviated large droplet stage. In this part, the waveform optimization for both bubble behaviors and metal transfer were investigated. Efforts were made for shortening the duration of the deviated large droplet stage. Pulse current influences on bubble evolution was studied. It was found that two different separation modes can be adjusted by appropriately changing the current values when the bubbles are necking. Quickly reducing the welding current can sharply lower the impact force on the droplets due to intense gas flow changes inside. Under the optimized pulse current, the range of the metal transfer cycle became narrower, and droplet diameters were smaller than that of the original condition. Stable and improved metal transfer processes were achieved with a frequency of 7.52 Hz and an average droplet diameter of 2.4 mm, which was about 1.5 times the wire diameter. The optimized pulse waveform greatly improved weld formation with less spatter and a more uniform appearance.
{"title":"Underwater Pulse-Current FCAW - Part 2: Bubble Behaviors and Waveform Optimization","authors":"Junfei Wu, Yanfei Han, C. Jia, Qingyuan Yang, Chuansong Wu","doi":"10.29391/2020.99.028","DOIUrl":"https://doi.org/10.29391/2020.99.028","url":null,"abstract":"Underwater pulse-current wet welding was proposed in part 1 of this two-part report. The novel technology obtained improved metal transfer and welding process stability. The main reason for droplet oversizing and long transfer cycles was found to be the deviated large droplet stage. In this part, the waveform optimization for both bubble behaviors and metal transfer were investigated. Efforts were made for shortening the duration of the deviated large droplet stage. Pulse current influences on bubble evolution was studied. It was found that two different separation modes can be adjusted by appropriately changing the current values when the bubbles are necking. Quickly reducing the welding current can sharply lower the impact force on the droplets due to intense gas flow changes inside. Under the optimized pulse current, the range of the metal transfer cycle became narrower, and droplet diameters were smaller than that of the original condition. Stable and improved metal transfer processes were achieved with a frequency of 7.52 Hz and an average droplet diameter of 2.4 mm, which was about 1.5 times the wire diameter. The optimized pulse waveform greatly improved weld formation with less spatter and a more uniform appearance.","PeriodicalId":23681,"journal":{"name":"Welding Journal","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44674792","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
P. P. G. Ribeiro, P. Assunção, E. Braga, R. A. Ribeiro, A. Gerlich
The hot-wire gas metal arc welding (HW-GMAW) process is widely used to increase the melting rate of a secondary wire through Joule heating without significantly increasing the total heat input to the substrate. Because there is limited knowledge regarding the associated arc dynamics and its influence on bead geometry, the present study considers how these are affected by the hot-wire polarity (negative or positive), hot-wire feed rate, and hot-wire orientation using a two-factor full factorial experiment with three replicates. During welding, high-speed imaging synchronized with current and voltage acquisition to study the arc dynamics. After this, each replicated weld was cut into three cross sections, which were examined by standard metallography. The preliminary results suggest that the arc was stable within the range of process parameters studied. The arc polarity played a role on arc position relative to the hot wire, with a decrease in penetration depth observed when the arc was attracted to the hot wire.
{"title":"Metal Transfer Mechanisms in Hot-Wire Gas Metal Arc Welding","authors":"P. P. G. Ribeiro, P. Assunção, E. Braga, R. A. Ribeiro, A. Gerlich","doi":"10.29391/2020.99.026","DOIUrl":"https://doi.org/10.29391/2020.99.026","url":null,"abstract":"The hot-wire gas metal arc welding (HW-GMAW) process is widely used to increase the melting rate of a secondary wire through Joule heating without significantly increasing the total heat input to the substrate. Because there is limited knowledge regarding the associated arc dynamics and its influence on bead geometry, the present study considers how these are affected by the hot-wire polarity (negative or positive), hot-wire feed rate, and hot-wire orientation using a two-factor full factorial experiment with three replicates. During welding, high-speed imaging synchronized with current and voltage acquisition to study the arc dynamics. After this, each replicated weld was cut into three cross sections, which were examined by standard metallography. The preliminary results suggest that the arc was stable within the range of process parameters studied. The arc polarity played a role on arc position relative to the hot wire, with a decrease in penetration depth observed when the arc was attracted to the hot wire.","PeriodicalId":23681,"journal":{"name":"Welding Journal","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48656944","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Welding has been automated/robotized greatly. However, in typical automated/robotic welding applications, the welding parameters are preset and not adjusted adaptively to overcome the effect from unpredicted disturbances. This imperfection cannot meet the increasing requirements from the welding/manufacturing industry on quality, efficiency, and flexibility. Combining information sensing/processing with traditional welding manufacturing techniques has been a major directive to revolutionize the welding industry (Ref. 1). In practical welding, the weld penetration, as measured by the back-side weld bead width, is a critical factor determining the integrity of the weld produced. However, the back-side bead width is difficult to monitor directly during manufacturing because it occurs underneath the surface of the workpiece being processed. Therefore, predicting the back-side bead width using conveniently sensible information from the welding process becomes a fundamental issue in intelligent welding. Many studies have been done to predict the weld penetration using different characteristic information from the welding process. They typically 1) sense observable phenomena from the welding process using, or based on, different sensors/phenomena such as infrared, pool oscillation, laser ultrasonic, and active vision methods (Refs. 2–5); 2) define and extract characteristic features from sensed phenomena; and 3) build a model to correlate the extracted characteristic features to the penetration state (Refs. 6, 7). However, the characteristic features are proposed subjectively based on the individual’s understanding of the physics, thus lacking a systematic way to ensure success in leading to a good model. Iteration is often needed such that the development efficiency is low. To address this general challenge, researchers recently started to apply deep-learning-based methods to extract the information automatically. Therefore, the major remaining challenge is reduced to acquiring adequate information from the welding process. Skilled welders can judge the weld penetration per their observed welding phenomena during the process. The welding community believes that images from the observable welding scene, including the 3D weld pool surface, contain sufficient information to predict the weld penetration (Ref. 8). While earlier efforts followed the aforementioned procedure to first propose characteristic features, the deep learning method has recently been applied, with a concentration on using convolutional neural networks (CNNs), to directly map images to the penetration (Refs. 9–14). The training for the parameters, including the convolutional kernels and Prediction of Weld Penetration Using Dynamic Weld Pool Arc Images
{"title":"Prediction of Weld Penetration Using Dynamic Weld Pool Arc Images","authors":"Wenhua Jiao, Qiyue Wang, Yongchao Cheng, Rui Yu, Yuming Zhang","doi":"10.29391/2020.99.027","DOIUrl":"https://doi.org/10.29391/2020.99.027","url":null,"abstract":"Welding has been automated/robotized greatly. However, in typical automated/robotic welding applications, the welding parameters are preset and not adjusted adaptively to overcome the effect from unpredicted disturbances. This imperfection cannot meet the increasing requirements from the welding/manufacturing industry on quality, efficiency, and flexibility. Combining information sensing/processing with traditional welding manufacturing techniques has been a major directive to revolutionize the welding industry (Ref. 1). In practical welding, the weld penetration, as measured by the back-side weld bead width, is a critical factor determining the integrity of the weld produced. However, the back-side bead width is difficult to monitor directly during manufacturing because it occurs underneath the surface of the workpiece being processed. Therefore, predicting the back-side bead width using conveniently sensible information from the welding process becomes a fundamental issue in intelligent welding. Many studies have been done to predict the weld penetration using different characteristic information from the welding process. They typically 1) sense observable phenomena from the welding process using, or based on, different sensors/phenomena such as infrared, pool oscillation, laser ultrasonic, and active vision methods (Refs. 2–5); 2) define and extract characteristic features from sensed phenomena; and 3) build a model to correlate the extracted characteristic features to the penetration state (Refs. 6, 7). However, the characteristic features are proposed subjectively based on the individual’s understanding of the physics, thus lacking a systematic way to ensure success in leading to a good model. Iteration is often needed such that the development efficiency is low. To address this general challenge, researchers recently started to apply deep-learning-based methods to extract the information automatically. Therefore, the major remaining challenge is reduced to acquiring adequate information from the welding process. Skilled welders can judge the weld penetration per their observed welding phenomena during the process. The welding community believes that images from the observable welding scene, including the 3D weld pool surface, contain sufficient information to predict the weld penetration (Ref. 8). While earlier efforts followed the aforementioned procedure to first propose characteristic features, the deep learning method has recently been applied, with a concentration on using convolutional neural networks (CNNs), to directly map images to the penetration (Refs. 9–14). The training for the parameters, including the convolutional kernels and Prediction of Weld Penetration Using Dynamic Weld Pool Arc Images","PeriodicalId":23681,"journal":{"name":"Welding Journal","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47242066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
S. Zou, Zhijiang Wang, Sheng-Sun Hu, G. Zhao, Wandong Wang, You-Quan Chen
In gas tungsten arc welding (GTAW), the filler wire increases the deposition efficiency and influences the welding stability. Its interactions with the gas tungsten arc (GTA) are significant to better understand the welding process and to monitor and control weld quality. In view of this, the first part of the work, Effects of Filler Wire Intervention on Gas Tungsten Arc: Part I — Mechanism, explained the interaction mechanisms between the filler wire and the gas tungsten arc based on the proposed arc-sensing method of detecting probe voltage (i.e., the voltage signal between the filler wire and the tungsten electrode/workpiece). In this second part of the work, experiments were designed to make the filler wire melt in different areas of the arc to study the dynamic behaviors of the droplet and its effect on the arc. Typical metal transfer modes are discussed, and droplet oscillation is geometrically characterized through image processing and then analyzed in the time domain and time-frequency domain. The results show that the liquid droplet affects the arc through its transfer to the weld pool, its oscillation, and occupying the arc space. Information about these dynamic behaviors can be easily reflected in the probe voltage, which would be a valuable signal to monitor the process stability in GTAW with filler wire. This work shows the potential of the proposed sensing method for monitoring and controlling weld quality in all welding positions, GTA-based additive manufacturing, etc.
{"title":"Effects of Filler Wire Intervention on Gas Tungsten Arc: Part II - Dynamic Behaviors of Liquid Droplets","authors":"S. Zou, Zhijiang Wang, Sheng-Sun Hu, G. Zhao, Wandong Wang, You-Quan Chen","doi":"10.29391/2020.99.025","DOIUrl":"https://doi.org/10.29391/2020.99.025","url":null,"abstract":"In gas tungsten arc welding (GTAW), the filler wire increases the deposition efficiency and influences the welding stability. Its interactions with the gas tungsten arc (GTA) are significant to better understand the welding process and to monitor and control weld quality. In view of this, the first part of the work, Effects of Filler Wire Intervention on Gas Tungsten Arc: Part I — Mechanism, explained the interaction mechanisms between the filler wire and the gas tungsten arc based on the proposed arc-sensing method of detecting probe voltage (i.e., the voltage signal between the filler wire and the tungsten electrode/workpiece). In this second part of the work, experiments were designed to make the filler wire melt in different areas of the arc to study the dynamic behaviors of the droplet and its effect on the arc. Typical metal transfer modes are discussed, and droplet oscillation is geometrically characterized through image processing and then analyzed in the time domain and time-frequency domain. The results show that the liquid droplet affects the arc through its transfer to the weld pool, its oscillation, and occupying the arc space. Information about these dynamic behaviors can be easily reflected in the probe voltage, which would be a valuable signal to monitor the process stability in GTAW with filler wire. This work shows the potential of the proposed sensing method for monitoring and controlling weld quality in all welding positions, GTA-based additive manufacturing, etc.","PeriodicalId":23681,"journal":{"name":"Welding Journal","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2020-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42428006","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The susceptibility of austenitic, ferritic, and duplex stain-less steels to solidification cracking was evaluated by the new Transverse Motion Weldability (TMW) test. The focus was on austenitic stainless steels. 304L and 316L were least susceptible, 321 was significantly more susceptible, and 310 was much more susceptible. However, some 321 welds were even less susceptible than 304L welds. These 321 welds were found to have much finer grains to better resist solidification cracking. Quenching 321 during welding revealed spontaneous grain refining could occur by heterogeneous nucleation. For 304L, 316L, and 310, a new explanation for the susceptibility was proposed based on the continuity of the liquid between columnar dendrites; a discontinuous, isolated liquid allows bonding between dendrites to occur early to better resist cracking. In 304L and 316L, the dendrite-boundary liquid was discontinuous and isolated, as revealed by quenching. The liquid was likely depleted by both fast back diffusion into -dendrites (body-centered cubic) and the L + + reaction, which consumed L while forming . In 310, however, the dendrites were separated by a continuous liquid that prevented early bonding between them. Back diffusion into -dendrites (face-centered cubic) was much slower, and the L + + reaction formed little . Quenching also revealed skeletal/lacy formed in 304L and 316L well after solidification ended; thus, skeletal/lacy did not resist solidification cracking, as had been widely believed for decades. The TMW test further demonstrated that both more sulfur and slower welding can increase susceptibility.
{"title":"Solidification Cracking Susceptibility of Stainless Steels: New Test and Explanation","authors":"Kun Liu, P. Yu, S. Kou","doi":"10.29391/2020.99.024","DOIUrl":"https://doi.org/10.29391/2020.99.024","url":null,"abstract":"The susceptibility of austenitic, ferritic, and duplex stain-less steels to solidification cracking was evaluated by the new Transverse Motion Weldability (TMW) test. The focus was on austenitic stainless steels. 304L and 316L were least susceptible, 321 was significantly more susceptible, and 310 was much more susceptible. However, some 321 welds were even less susceptible than 304L welds. These 321 welds were found to have much finer grains to better resist solidification cracking. Quenching 321 during welding revealed spontaneous grain refining could occur by heterogeneous nucleation. For 304L, 316L, and 310, a new explanation for the susceptibility was proposed based on the continuity of the liquid between columnar dendrites; a discontinuous, isolated liquid allows bonding between dendrites to occur early to better resist cracking. In 304L and 316L, the dendrite-boundary liquid was discontinuous and isolated, as revealed by quenching. The liquid was likely depleted by both fast back diffusion into -dendrites (body-centered cubic) and the L + + reaction, which consumed L while forming . In 310, however, the dendrites were separated by a continuous liquid that prevented early bonding between them. Back diffusion into -dendrites (face-centered cubic) was much slower, and the L + + reaction formed little . Quenching also revealed skeletal/lacy formed in 304L and 316L well after solidification ended; thus, skeletal/lacy did not resist solidification cracking, as had been widely believed for decades. The TMW test further demonstrated that both more sulfur and slower welding can increase susceptibility.","PeriodicalId":23681,"journal":{"name":"Welding Journal","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2020-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48358886","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
S. Zou, Zhijiang Wang, Sheng-Sun Hu, G. Zhao, Wandong Wang, You-Quan Chen
For gas tungsten arc welding (GTAW), the effects of filler wire on the GTA are worth being clarified, which will help deepen the understanding of arc characteristics and in-spire new ideas for the real-time monitoring of weld quality. To this end, this work proposed a novel sensing method of detecting probe voltages (i.e., the voltage signals between a filler wire and tungsten electrode/workpiece). Based on this method, in this first part of the work, a tungsten probe was used to replace the filler wire and to interact with the arc in the specific experiments to elucidate the static and dynamic interaction mechanisms between the GTA and filler wire. The results showed that the filler wire intervention deflects the arc to various degrees and will change the volt-age signals. As a metal conductor, the filler wire will in-crease the arc voltage by increasing the average electric field strength. However, its effects on the different areas of the arc are not always consistent, which makes the change trend of the probe voltages not always the same. Moreover, due to thermal inertia, the probe voltage does not strictly change synchronously with the arc voltage under the dynamic disturbance. This work lays a theoretical foundation for monitoring the stability of the GTAW process.
{"title":"Effects of Filler Wire Intervention on Gas Tungsten Arc: Part I - Mechanism","authors":"S. Zou, Zhijiang Wang, Sheng-Sun Hu, G. Zhao, Wandong Wang, You-Quan Chen","doi":"10.29391/2020.99.023","DOIUrl":"https://doi.org/10.29391/2020.99.023","url":null,"abstract":"For gas tungsten arc welding (GTAW), the effects of filler wire on the GTA are worth being clarified, which will help deepen the understanding of arc characteristics and in-spire new ideas for the real-time monitoring of weld quality. To this end, this work proposed a novel sensing method of detecting probe voltages (i.e., the voltage signals between a filler wire and tungsten electrode/workpiece). Based on this method, in this first part of the work, a tungsten probe was used to replace the filler wire and to interact with the arc in the specific experiments to elucidate the static and dynamic interaction mechanisms between the GTA and filler wire. The results showed that the filler wire intervention deflects the arc to various degrees and will change the volt-age signals. As a metal conductor, the filler wire will in-crease the arc voltage by increasing the average electric field strength. However, its effects on the different areas of the arc are not always consistent, which makes the change trend of the probe voltages not always the same. Moreover, due to thermal inertia, the probe voltage does not strictly change synchronously with the arc voltage under the dynamic disturbance. This work lays a theoretical foundation for monitoring the stability of the GTAW process.","PeriodicalId":23681,"journal":{"name":"Welding Journal","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2020-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48300540","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Chaoxian Li, Qiyue Wang, Wenhua Jiao, Michael T. Johnson, Yuming Zhang
An innovative method was proposed to determine weld joint penetration using machine learning techniques. In our approach, the dot-structured laser images reflected from an oscillating weld pool surface were captured. Experienced welders typically evaluate the weld penetration status based on this reflected laser pattern. To overcome the challenges in identifying features and accurately processing the images using conventional machine vision algorithms, we proposed the use the raw images without any processing as the input to a convolutional neural network (CNN). The labels needed to train the CNN were the measured weld penetration states, obtained from the images on the backside of the workpiece as a set of discrete weld penetration categories. The raw data, images, and penetration state were generated from extensive experiments using an automated robotic gas tungsten arc welding process. Data augmentation was performed to enhance the robustness of the trained network, which led to 270,000 training examples, 45,000 validation examples, and 45,000 test examples. A six-layer convolutional neural network trained with a modified mini-batch gradient descent method led to a final testing accuracy of 90.7%. A voting mechanism based on three continuous images increased the classification accuracy to 97.6%.
{"title":"Deep Learning-Based Detection of Penetration from Weld Pool Reflection Images","authors":"Chaoxian Li, Qiyue Wang, Wenhua Jiao, Michael T. Johnson, Yuming Zhang","doi":"10.29391/2020.99.022","DOIUrl":"https://doi.org/10.29391/2020.99.022","url":null,"abstract":"An innovative method was proposed to determine weld joint penetration using machine learning techniques. In our approach, the dot-structured laser images reflected from an oscillating weld pool surface were captured. Experienced welders typically evaluate the weld penetration status based on this reflected laser pattern. To overcome the challenges in identifying features and accurately processing the images using conventional machine vision algorithms, we proposed the use the raw images without any processing as the input to a convolutional neural network (CNN). The labels needed to train the CNN were the measured weld penetration states, obtained from the images on the backside of the workpiece as a set of discrete weld penetration categories. The raw data, images, and penetration state were generated from extensive experiments using an automated robotic gas tungsten arc welding process. Data augmentation was performed to enhance the robustness of the trained network, which led to 270,000 training examples, 45,000 validation examples, and 45,000 test examples. A six-layer convolutional neural network trained with a modified mini-batch gradient descent method led to a final testing accuracy of 90.7%. A voting mechanism based on three continuous images increased the classification accuracy to 97.6%.","PeriodicalId":23681,"journal":{"name":"Welding Journal","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2020-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47154096","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：