Robot welding technologyWith advanced manufacturing techniques, welding product manufacturing automation, flexibility and intelligence has become a trend. Currently, welded by robot welding automation skills has become the primary mark of modernization. Welding robot task because of its versatility, strong strength, was more and more attention. Robot in the welding production skills can improve productivity, improve labor provided, unchanging and ensure welding quality, finish and small-batch products of welding automation.
was born from 60 months and until now, discussion has experienced three stages of welding robots, playback phases, the off-line programming and viable programming phase. Follow accounting machine control skills, improving from time to time, welding robots from a single stand-alone playback mode to multi-sensing, intelligent flexible manufacturing unit (System) bias, completed the transition from the second generation to the third generation will be welding robots pursue objective.
current, country exterior many using arc welding mechanical people system from all Shang see fundamental are belongs to first generation or associate II generation of welding because welding way and welding parameter is according to practice lessons premise advance set of, in welding Shi short external information sensing and timely adjustment control function, this class arc welding mechanical people on welding lessons premise of not degeneration requirements severe, welding Shi lack "flexible", displayed out clear of defects. In practice in arc welding process, welding is often change the premise of, such as machining and Assembly error on form the weld position and size changes, artifacts in the welding process heating and cooling if changes can create uneven weld bead penetration and deformation. Robot welding to restraint effect of various uncertain factors on welding quality in the process, improve my mechanics who work intelligent and reliable task, requires space to be arc-welding robot system not only proactive and timely tracking of weld, welding parameters and can adjust and welding quality control in a timely manner.
2. the primary composition of the Intelligent robot welding skills
blend has a typical multidisciplinary interspersed with modern welding skills feature, robot welding is the expression of related disciplines and skill. Welding robot for the introduction of intelligent skill reached by primary skill composition as shown in Figure 1. This include:
1) welding mechanical people on welding obligations of independence planning skills;
2) welding mechanical people of activities track control skills;
3) welding dynamic process of information sensing, and built die and intelligent control skills;
4) mechanical people welding system of integrated and control, will above welding obligations planning, and track track control, and sensing system, and process die, and intelligent control, subsystem of hardware and software integrated design, and consistent optimization transfers and control, Touch welding flexible manufacturing system of governance and control of material flow and information flow, robots and sensors, control of intelligent units and control of complex systems, and so on.
under the above primary skills related to research and identify the main narrative.
3. robot welding obligation independent planning skills
as already mentioned, arc welding robots are abroad-teaching-the playback, it cannot be enough welding to produce increasingly complex needs, there are still many problems to be deliberate. About arc welding mechanical people of research gradually to independence of transition, rendering has arc welding mechanical people of offline programming skills, a more intact of arc welding mechanical people offline programming system should include welding lessons obligations depicting (speech programming or graphics simulation), and operation hand level way planning, and activities learn and dynamics algorithm and the optimization, and for welding lessons obligations of joint level planning, and planning consequences animation simulation, and planning consequences offline correcting, and and mechanical people of communications interface (downloading), and Applying sensor self programming approaches and ways of online grading, and other major parts. Among its key skills include Visual design of sensor information access issues, planning and weld controller design.
National School of researchers on welding robot collision-free path planning, with redundant degrees of arc welding robot independent planning and discussion on the welding process parameter programming tasks, design opened up the off-line programming and simulation system structure as shown in Figure 2.
arc welding robot planning system including a CAD system, welding expert system, self-reliant planning system and mimic simulation systems. From a more general sense, a more perfect arc welding robot system should also include the reaction control systems, sensing systems prior to welding and weld inspection system.
4. robot welding seam tracking and lifted him to approach skills
robot welding work welding robot trajectory control the Prime refers to the initial welding position held and seam tracking control skills. Arc welding robots in all kinds of categories, and adaptability are the welding quality and efficiency of the most important elements. Arc welding robot of adaptability from the workpiece using the sensor input signal is detected in time control and the correct operation of machinery in order to adapt to the changes of the weld if and.
Sweden and the United States the company has developed laser scanning and structural light Visual sensors for welding seam tracking system. Domestic hit Lv Weixin, Dr Zhang jiong was developed based on laser scanning and high functioning Visual system for linear array CCD sensors.
in the initial welding robot Visual help skills seminar, Dr Lv Weixin vision system based on laser scanning, designing the next lookup algorithm, finished workpiece weld feature in the inevitable limitations of self help . Guo Zhenmin Li Jinquan and identify by Visual servoing and image recognition skills discussed the initial welding robot welding Rotary and weld to identify and track issues in a timely manner.
5. weld pool dynamic process and Visual sensing, modeling and intelligent moderation skills
robot welding quality key is useful for accurate control of the complete welding dynamic process. Because the complexity of the welding process, theory indicates classical restraint means larger usefulness was limited. Enlightenment of a skilled welder operation skill, in recent years, imitating the welder operations welding dynamic process of intelligent control method has been introduced, the first touch of weld pool dynamic process Visual sensing, modeling and intelligent control.
sensing skills of welding process the welding process of the sensor, is a key link in quality control of the welding process. Future trend of welding is Welding automation, robots and intelligent, and sensory skills is the most important part. Welding sensor can be classified by their target measurement and inspection operations, inspection and monitoring processes two types of welding. In terms of sensing makes sense, first into acoustic, mechanical, electric, optical sensors, and so on.
acoustic sensor is mainly used for GMAW process of droplet detection perforated, plasma welding. Mechanical sensing primary means the molten pool oscillation technique developed in recent years. Characteristics of the arc sensor for direct detection of arc itself (current, voltage), do not need additional sensors and maintenance required for external sensor and noise removal installation, use brief. Current use of seam tracking and weld control is an area first. Compared with the other sensing method, optical sensors without contact with the welding circuit, detection of signals does not affect the normal welding process is one of most future prospects of sensing skills. Welding application of infrared thermal radiation sensor optical sensing of weld process information is a systematic approach, more research results in this field .
welding weld pool vision sensing process, direct vision is one of the best non-contact sensors. Vision sensing skills's primary strengths is not directly contacts the workpiece, do not disturb the normal welding process, access to information, universal. And welding process because it can get dynamic pools in two or three dimensions of information, compared with the other welding process information detection method, which detects welding process information of weld pool is a direct reflection of the dynamic behavior of molten metal, better quality control of the welding process.
direct vision sensor used in welding include the offline work status determination; online compensation as fixed precision, mechanical local tolerance, welding of welding deformation in the process of welding path deviation; in welding process control of welded joints and the timely sensing of weld pool geometry; monitoring of droplet transfer modes.
in recent years, followed the development of machine vision for the accounting skills, applying machine vision front directly see the weld pool, after image disposal get the weld pool geometry information on welding quality of the close links, has become a major discussion of bias. According to the imaging light is in the Visual inspection system to help light the light so the weld zone itself, direct Visual automatic mode and passive detection system can be divided into two broad categories.
automatic direct vision sensing in order to reduce arcing effect on image quality, automatic visual inspection methods directly using laser-assisted source on welded areas, such as artificial lighting, to improve image quality. Because the laser with a single wavelength in favour of good, good correlation characteristics, so using laser assisted light source can achieve a clear image. Automatic vision because of its more expensive equipment required, the system is complex, often limit the use of its implementation.
passive direct vision sensing current visual image sensor is used as a welding process skills, study more without extra help light lighting zone uses the arc itself, namely passive direct Visual inspection methods.
on passive directly Visual sensing more deepening of research started Yu 80 years months medium-term, country exterior Research task who had directly application arc light shines melt pool ahead of artifacts clearance gets welding district weld information, according to melt pool ahead differences near at ARC light strength of Flash burning to completed welding process in the of weld track; application with CCD camera of micro accounting machine control system on pulse MIG/MAG welding melt pool behavior for has see and control of research ; United States Ohio University opened up a place in the welding torch and coaxial electrodes observation of weld pool vision sensing system of integration, in molten pool during TIG welding check and MIG welding seam tracking for a preliminary discussion.