Torch Position Research Articles

Digital control of torch position and weld pool in MIG welding using image processing device

A digital computer control system for improving the stability of the torch position and weld pool shape in metal inert gas (MIG) welding is proposed. It is shown how the characteristic of the digital controller may be selected for yielding a desired transient behavior. >

Weld-line tracking control of arc welding robot using fuzzy logic controller

This paper deals with the development of a weld-line tracking control system for an arc welding robot using a fuzzy logic controller. The horizontal and vertical displacements of the welding point, which are outputs of the arc sensor developed at Mitsubishi Heavy Industries, Ltd., arc identified from the relation between the oscillation position of the welding torch and the wire tip-to-work distance. Using this information, it is possible to know how much and in what direction to adjust the torch position of the arc welding robot. However, these values contain much noise caused by the hazardous environment wherein extremely bright welding sparks, high temperature and toxic fumes and gasses. So it is reasonable to consider these values as fuzzy information. Hence, a fuzzy filter and a fuzzy logic controller based on linguistic control rules have been designed in the development of the weld-line tracking control system. Finally, through simulation and experimental results, the effectiveness of the fuzzy logic controller with fuzzy filter is discussed in comparison with other types of controller.

画像処理による溶接制御システムの開発

Welding control system for Narrow gap GTAW was developed. It is capable of simultaneous control of both welding torch position and welding heat input by image processing the thermal pattern of melting zone.This system is characterized in that an isothermal pattern of weld zone is detected with infrared camera, and the thermal pattern of melting zone is selected by image processing. The position of welding torch is controlled according to the judgement of asymmetric extend of thermal pattern.Welding heat input control is conducted by comparing the area of thermal pattern with the area of standard pattern, too. According to this system, it is possible to maintain an optimum welding condition nn real time.

ＧＭＡ溶接におけるチップ・母材間距離の推定法

The weld line tracking sensor is indispensable to improve the flexibility of the arc welding robot applications. Recently, a sensing method which utilizes the electric arc phenomenon, or more correctly the welding current, has been developed and prevalently used. The sensing accuracy, however, is directly affected by the estimate of the contact tip-workpiece distance.The authors investigated the algorism of estimating the contact tip-workpiece distance quantitatively on the assumption that the welding arc is in an equilibrium state, with the following results.(1) The electrode extension (LE) in each of the pulsed-current and non-pulsed-current MAG processes can be calculated using average value (Ia) and effective value (Ie) of current and wire feed rate (v).LE=(v-K3Ia)/(K4I2e) where K3, K4=constant(2) The voltage drop (VE) in the extended electrode can be calculated with Ia, Ie, v and LE.VA=(4η0Ia/πd2)[(1-β/α)LE+(βv/αbI2e){exp(bI2eLE/v)-1}](3) Then, the arc length (LA) can be calculated as follows, from Ia and from Va which is given by subtracting VE from welding voltage (V).LA=(VA-K6-K8Ia)/(K5+k7Ia) where K5-K8=constant(4) The contact tip-workpiece distance (LE+LA) can be correctly estimated by the above equations.It is possible, by judging a torch position displacement in the groove on basis of the contact tip-work-piece distance trend during the oscillation, to establish the arc sensing system capable of tracking the weld line with the higher degree of accuracy than hitherto.

Sampled-data control of torch position and arc length in GMA welding using image processing device.

Sampled-data control of torch position and arc length in GMA welding using image processing device.

Adaptive Control of Industrial Robots for Arc Welding

The paper examines the problem of adaptation to inexact prepared or inaccurate positioned parts when welding by industrial robots, an idea for adaptive control is presented, based on the parts deviation from their preliminary taught model, an example of the model is described for the case of strightforward joint, taking into account the welding torch position and the sensor parameters, results from experiments made with the arc welding robot re 251 are presented, some noice and accuracy problems are discussed, and suggestions for further developments are given.

配管の自動溶接に関する研究(第1報)

In a nuclear power plant, a large number of pipes must be set in very complex and narrow places. A remote control pipe welding equipment is developed, in order to solve difficulty of welding in such a narrow place. Characteristics of this equipment are summarized as follows.(1) All welding sequences are controlled by a mini-computer.(2) From the information displayed on the screen of control panel, which inculdes a TV monitor, the operator can evaluate results of welding, torch positioning and the groove shape.(3) The operating panel includes a keyboard which is used for parameter input and for communication with the welding equipment to make necessary adjustment.

Application of Microcomputers to Automatic Weld Quality Control

The demand for increased productivity of welding operations has led to the expanded use of computer control to allow higher production rates while maintaining weld quality. The charateristics of the gas metal arc welding process and the relationship between welding parameters, the desired output of the welding process, and the automation of the process are discussed. A strategy for two-axis welding torch positioning and velocity control is developed based on preview control techniques. To evaluate the applicability of the proposed control method the motion of heat source along different welding paths is simulated on an analog computer with on-line control of process time constants by an LSI-11 microcomputer. The simulation results show that high quality seam tracking can be accomplished by controlling the torch motion using the proposed method. The method appears to be suitable for on-line control of welding processes.