• International Journal of Korean Welding Society
• /
• v.1 no.1
• /
• pp.23-30
• /
• 2001

Among the various welding conditions, the welding current that is inversely proportional to the tip-to-work-piece distance is an essential parameter as to monitor the GMAW process and to implement the welding automation. Considering the weld pool surface geometry including weld defects, it should modify the signal processing method for automatic seam tracking in horizontal fillet welding. To meet the above necessities, a mathematical model related with the weld pool geometry was proposed as in a conjunction with the two-dimensional heat flow analysis of the horizontal fillet welding. The signal processing method based on the artificial neural network (Adaptive Resonance Theory) was proposed for discriminating the sound weld pool surface from that with the weld defects. The reliability of the numerical model and the signal processing method proposed were evaluated through the experiments of which showed that they are effective for predicting the weld bead shape with or without the weld defects in a horizontal fillet welding.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.16 no.3
• /
• pp.81-87
• /
• 2007

It is important to investigate the relationship between weld process parameters and weld bead geometry for adaptive arc robot welding. However, it is difficult to predict an exact back-bead owing to gap in process of butt welding. In this paper, the quantitative prediction system to specify the relationship external weld conditions and weld bead geometry was developed to get suitable back-bead in butt welding which is widely applied on industrial field. Multiple regression analysis and artificial neural network were used as the research methods. And, the results of two prediction methods were compared and analyzed.

• Proceedings of the KWS Conference
• /
• 1996.10a
• /
• pp.139-141
• /
• 1996

Among various welding parameters, the welding current which is inversely proportional to the tip-to-workpiece distance in GMAW is an essential parameter to monitor the GMAW process of horizontal fillet joints. For the case of weld defect such as overlap in horizontal fillet welding, therefore, the signal processing for process monitoring or automatic seam tracking should be modified by considering the weld pool surface geometry including the corresponding weld defect. In other words, the adequate signal processing algorithm is indispensible to improve the performance of the arc sensor. However, arc sensor algorithm already developed usually focus on weld seam tracing but do not considering the weld qualities. In this paper, various experiments were carried out to investigate the tendencies of the weld defects when weaving motion is added, and the experimental method based on 2$^n$ factorial design was proposed for deriving the mathematical model between the leg length and the various welding conditions. Moreover, a signal processing method based on the artificial neural network(Adaptive Resonance Theory) was proposed far discriminating the current signal of sound weld beads from that of weld beads with overlap. Finally, the algorithm for weld seam tracking combined with the mathematical modeling and the signal processing method was carried out to track the weld line in conjunction with the improvement of the weld qualities. The reliability of the proposed algorithms were evaluated through various experiments, which showed that the proposed algorithms could be effectively used for arc welding automation.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.10a
• /
• pp.558-561
• /
• 2002

A CCD camera with a laser strip was applied to realize the automation of welding Process in GMAW. It takes relatively long time to process image on-line control using the basic Hough transformation, but it has a tendency of robustness over the noise such spatter and arc light. The adaptive Hough transformation was used to extract the laser stripe and to obtain specific weld points In this study, a neural network based on the generalized delta rule algorithm was adapted for the process control of GMA, such as welding speed, arc voltage and wire feeding speed.

• Journal of Welding and Joining
• /
• v.14 no.4
• /
• pp.79-88
• /
• 1996

For improvement in productivity and weld quality, weld seam tracking and welding parameter control are very essential in the welding of a structure which can not be cxactly fit-up due to mismatch, discontinous gap, deflection, etc.. In this study, an automatic weld seam tracking system is developed for I-butt joint structure, and the system consists of XYZ working table, vision sensor and user interface program. In the developed vision sensor system, an image projection algorithm for weld-line detection and an adaptive current control algorithm for gap variation were implemented. The user interface program developed in this study by basing on the objct oriented concept could provide very convenient way to utilize the tracking system with the pull-down menu driven structure. The developed system showed a good seam tracking and weld quality control capability corresponding to deflected weld lines and gap variations.

• Proceedings of the KWS Conference
• /
• 2006.05a
• /
• pp.160-162
• /
• 2006

• Journal of Welding and Joining
• /
• v.17 no.5
• /
• pp.40-46
• /
• 1999

The sensors employed in the robotic are welding system must detect the changes in weld characteristics and produce the output that is in some way related to the change being detected. Such adaptive systems, which synchronise the robot arm and eyes using a primitive brain will form the basis for the development of robotic GMA(Gas Metal Arc) welding which increasingly higher levels of artificial intelligence. The objective of this paper is to realize the mapping characteristics of bead height through learning. After learning, the neural estimation can estimate the bead height desired from the learning mapping characteristic. The design parameters of the neural network estimator(the number of hidden layers and the number of nodes in a layer) are chosen from an estimation error analysis. A series of bead of bead-on-plate GMA welding experiments was carried out in order to verify the performance of the neural network estimator. The experimental results show that the proposed neural network estimator can predict the bead height with reasonable accuracy and guarantee the uniform weld quality.

• Journal of Welding and Joining
• /
• v.34 no.1
• /
• pp.21-25
• /
• 2016

The automotive vehicle is made through the following processes such as press shop, welding shop, paint shop, and general assembly. Among them, the most important process to determine the quality of the car body is the welding process. Generally, more than 400 pressed panels are welded to make BIW (Body In White) by using the RSW (Resistance Spot Welding) and GMAW (Gas Metal Arc Welding). Recently, as the needs of light-weight material due to the $CO_2$ emission issue and fuel efficiency, new joining technologies for aluminum, CFRP (Carbon Fiber Reinforced Plastic) and etc. are needed. Aluminum parts are assembled by the spot welding, clinching, and SPR (Self Piercing Rivet) and friction stir welding process. Structural adhesive boning is another main joining method for light-weight materials. For example, one piece aluminum shock absorber housing part is made by die casting process and is assembled with conventional steel part by SPR and adhesive bond. Another way to reduce the amount of the car body weight is to use AHSS (Advanced High Strength Steel) panel including hot stamping boron alloyed steel. As the new materials are introduced to car body joining, productivity and quality have become more critical. Productivity improvement technology and adaptive welding control are essential technology for the future manufacturing environment.

• Proceedings of the KWS Conference
• /
• 2009.11a
• /
• pp.10-10
• /
• 2009

파이프 용접은 중력의 영향으로 인하여 위치에 따라 같은 용접변수라도 비드 형상이 매우 달라 지게 된다. 또한 지금까지 많은 용접 기술자들이 위험하고 까다로운 환경에서 수작업으로 용접을 실행하였다. 따라서 이러한 이유로 용접 자동화 공정이 반드시 필요하게 된다. 본 연구에서는 FCAW를 사용하여 파이프 모재 대신 필릿 평판을 아래보기, 위보기 자세를 포함하여 9개 자세에서 실행하였다. 용접 자세를 비롯한 용접 변수와 비드 형상 변수간의 관계를 비선형 회귀 분석과 구간적 3차 에르미트 보간법을 이용하여 주어진 용접 변수에서의 비드 단면의 형상을 예측하고, 비드의 결함 유무를 파악하였다. 이러한 방법을 통하여 자세에 따라서 용접 결함이 없는 용접 변수를 구할 수 있었다. 시각센서를 이용하여 용접 후 비드 형상에 대해 모니터링을 실시하였다. 모니터링의 알고리즘은 영상획득, 이진화, 세선화, 적응형 미디언 필터링, 적응형 허프 변환, 용접 결함 검출의 순서로 구성되어 있으며, 본 연구에서는 보다 빠른 영상처리를 위하여 적응형 미디언 필터링을 제시하였다. 모니터링을 통하여 2차원 비드 단면뿐만 아니라, 디루니 삼각법을 적용하여 3차원으로 비드 표면을 표현할 수 있다. 보간법을 사용하여 얻은 비드 형상과 시각 센서를 통하여 얻은 비드 형상간의 비교를 통하여 본 연구의 적합성 여부를 확인하였다.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.7 no.6
• /
• pp.57-63
• /
• 1998

An adaptive control in the robotic GMA welding is employed to monitor information about weld characteristics and process parameters as well as to modify those parameters to hold weld quality within acceptable limits. Typical characteristics are the bead geometry, composition, microstructure, appearance, and process parameters which govern the quality of the final weld. The main objectives of this thesis are to realize the mapping characteristics of bead width through learning. After learning, the neural estimation can estimate the bead width desired form the learning mapping characteristic. The design parameters of the neural network estimator(the number of hidden layers and the number of nodes in a layer) are chosen from an estimation error analysis. A series of bead of bead-on-plate GMA welding experiments was carried out in order to verify the performance of the neural network estimator. The experimental results show that the proposed neural network estimator can predict the bead width with reasonable accuracy and guarantee the uniform weld quality.