The geometry of the weld pool contains accurate, instantaneous information about welding quality. Thus, weld pool sensing and control plays a significant role in automated arc welding. Previous studies have focused on inferring penetration through models and controlling penetration by various methods, such as adaptive control, model based fuzzy logic, etc. In the present work, a weld pool imaging system employing a LaserStrobe (tradename) high shutter speed camera is used to obtain contrasting images and eliminate arcing interference. Two image processing tools based on edge detection and connectivity analysis extract online information about the weld pool length and width. A neurofuzzy control system elicited from both human experience and experimental results has been developed to control the welding current and welding speed in real time based on changes in weld pool dimensions. Closed loop control of welding speed is used to achieve desirable weld pool geometry.
[Show abstract][Hide abstract] ABSTRACT: In manual welding process, skilled welders can ensure the weld quality through compensating for deviation observed from the weld pool surface. In this paper a three dimensional vision sensing system was used to mimic the human vision system to observe the three-dimensional weld pool surface in pipe GTAW process. Novel characteristic parameters containing information about the penetration state specified by its back-side weld pool width and height were proposed based on the reconstructed three dimensional weld pool surfaces. Then, variation in characteristic parameters and their relationships with the back-side parameters were studied through experiments under different welding conditions. Direct measurement of penetration is not preferred in a manufacturing site, soft-sensing method was thus proposed as an alternative to obtain it in real time due to established soft-sensing model and auxiliary variables which can be sensed in real time. In order to obtain the penetration status in real time conveniently, back-propagation neural network, principle component analysis based back-propagation neural network and global best adaptive mutation particle swarm optimization based back-propagation neural network models were established to estimate the penetration based on the proposed characteristic parameters. It was found that the top-side characteristic parameters proposed can reflect the back-side weld pool parameters accurately and the models are capable of predicting the penetration status in real time by observing the three-dimensional weld pool surface.
[Show abstract][Hide abstract] ABSTRACT: Template matching or correlation is an image-processing technique that searches for specific features or characteristics within an image. This paper describes the use of a configurable feature correlation method for the reliable processing of high quality digitized welding images in order to supply measurements for use with a closed loop control system. A generic approach has been adopted to produce a versatile and robust image-processing technique capable of analysing and providing measurements from images with different properties or characteristics. Low cost, commercially available image acquisition hardware was used, and the image-processing software development was undertaken using a commonly available PC-based visual programming environment. The general application of this work has been concerned with the demonstration of vision-based closed loop process control strategies for a series of different welding processes. Closing the process control loop in this manner is believed to represent a significant advancement of the vision-based remote viewing or ‘monitoring’ systems that are presently available.
International Journal of Production Research 03/2004; 42(5):975-995. DOI:10.1080/00207540310001619632 · 1.48 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Purpose – Top-face control of weld penetration in TIG welding is required for fully automated systems to overcome variations in the welding process and fixturing systems. Design/methodology/approach – This paper presents a system based upon based on the real-time vision measurement and control of the upper surface or “topface” weld pool size. The primary objective has been to demonstrate the feasibility of using vision-based image processing to provide measurements and the subsequent control of upper bead weld geometrical properties during the weld formation or molten phase and correlate this to the backface weld bead size. Findings – Vision based measurement of the upper surface of the weld pool can be used, in real-time, to control the weld pool size. This allows more uniform weld penetration to be achieved in the presence of disturbances. Research limitations/implications – The system requires that the pool edges can be accurately identified using a correlation method. This requires images with good contrast between the weld pool and the workpiece. Practical implications – The system is applicable to both continuous and pulsed TIG welding. Originality/value – A novel reference feature correlation-based image analysis algorithm has been developed that may be configured to operate with a number of different welding processes. The issues of system integration, i.e. interfacing the system with legacy welding equipment are also discussed.
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.