面向大型结构件爬行机器人智能焊接技术

Intelligent welding technology for large structures crawling robot

大型结构件的焊接在船舶制造、油气化工、核电工程、能源电力、建筑钢构和轨道交通等高端制造领域中应用广泛,传统手工焊以及轨道小车、龙门吊等半自动焊已无法满足高效、高质量的焊接要求。大型结构件智能化焊接成为发展趋势,以无轨导全位置爬行焊接机器人为代表的智能化装备正逐步推动解决大型结构件的自动化焊接问题。该文首先系统介绍了国内外面向大型结构件焊接的智能机器人研究现状,并对有轨和无轨机器人的应用场景及优缺点进行比较,分析认为无轨爬行机器人的灵活性好、适应性强,更适用于大型结构件的焊接。其次,针对大型结构件多层多道焊接时因机械加工、装配、热变形、焊道过多、焊缝堆叠等导致的焊接精度及焊接质量难以保证等问题进行分析,认为自动打底和多层多道自动排道技术还存在多项难点亟待突破。最后,从自动打底、焊缝跟踪、多层多道自动排道方面对相关技术的探索进行归纳,预测了多模态深度学习和多传感器融合技术将成为引领焊接机器人发展的关键技术,为大型结构件的智能化焊接提供方向性指导。

[Significance]Welding has reached a very important position in large structural workpieces.The welding of large structural members also has been covered many fields,including high-end manufacturing fields such as shipbuilding,oil and gas chemical industry,nuclear power engineering,energy and power,building steel structure,and rail transit.The problems of traditional manual welding are instability and low efficiency.Even semi-automatic welding such as gantry welding and rail robot welding,which has not yet to meet the requirements of more efficient and high-quality automation industry.The welding development trend of large structural parts is more intelligent and automatic welding.In this case,the welding robot that has no track and can crawl in all positions has become the representative of intelligent welding equipment,and gradually has promote and solve the automatic intelligent welding of large structural parts.[Progress]In this paper,the research status of intelligent robots for welding large structural parts at home and abroad was systematically introduced,and the welding robots were classified according to the welding robots with and without rails.The application scenarios,advantages and disadvantages of these two types of welding robots were compared.Through these comparisons,it was concluded that the welding robot without rail was more suitable for welding large structural parts because of its better flexibility,adaptability and convenience.There were many problems in the welding process of large structural parts,especially when multi-layer and multi pass thick plates were welded.At the same time,these problems caused low welding process accuracy and low welding quality.Problems included low machining accuracy,inaccurate assembly,metal thermal deformation,many weld passes,weld beads stacked together and many other reasons.In the welding of large structural parts,there were still many difficulties that need to be solved urgently.Automatic backing welding technology and multi-layer and multi-channel automatic routing technology were two very critical and difficult problems that hindered the progress of the welding industry.Until now,the backing welding of large structural parts has mainly depends on manual work,because backing welding was the first weld connecting two welding workpieces,which was very critical in the whole welding process.In the process of automatic backing welding,there were strict requirements on the assembly clearance,unfitness,blunt edge,curling and welding heat input.If these factors were not well controlled,a large number of welding defects such as missing welding and incomplete welding would occur in the process of backing welding.Multi layer and multi pass welding also depended on manual arrangement to cause unstable welding quality.Therefore,it was more and more important to automatically plan the number of welding passes and layers,arrange the welding sequence,determine the position coordinates of each weld pass and adjust the welding parameters.[Conclusions and Prospects]Based on this situation,this paper summarizes the exploration of related technologies from the aspects of automatic backing welding technology,seam tracking technology,multi-layer and multi-channel automatic lane arrangement technology,etc.Multi-modal deep learning and multi-sensor fusion technology are widely used in many fields,and have become a concern in military,industrial and high-tech development.Thus,these two technologies will be the key to developing welding robots and provide guidance for the intelligent welding of large structural parts.In the future,artificial intelligence technology will lead the welding robot to achieve better welding quality.