弧焊机器人增材制造成形信息检测及控制研究

发布时间：2017-12-28 01:37

本文关键词：弧焊机器人增材制造成形信息检测及控制研究　出处：《哈尔滨工业大学》2016年硕士论文　论文类型：学位论文

【摘要】：熔化极气体保护电弧增材制造(Gas Metal Arc-Additive Manufacturing,GMA-AM)是一种金属材料增材制造工艺,具有生产成本低、熔敷效率高、灵活性强等优点,然而,其熔敷过程易受干扰,熔敷道尺寸容易偏离规划值,随着尺寸偏差不断累积,后续熔敷过程有可能无法按照规划进行,得到的成形件尺寸也会与规划结果相差甚远。因此提高GMA-AM成形精度是一个亟待解决的问题,本文建立了一套智能GMA-AM系统,对多层多道结构GMA-AM成形信息检测与闭环控制展开了研究。首先结合智能GMA-AM的需求,本文基于弧焊机器人技术以及主动视觉传感技术,设计并建立了一套智能GMA-AM系统,该系统能够在逐道熔敷过程中实时检测熔敷道尺寸信息,在人机交互界面显示并存储,同时具有操作者在线调控熔敷规范参数的接口,支持人机协作,操作者和机器都可以对熔敷过程进行决策,为实现熔敷道尺寸闭环控制打下了良好基础。在熔敷过程中采用META线结构光传感器提取熔敷道截面轮廓信息,开发了一套基于点云数据的多层多道GMA-AM熔敷道尺寸实时检测算法,根据结构光条纹形态学特征,从点云数据中提取熔敷道顶点以及焊趾特征来计算熔敷道高度、熔敷道宽度以及熔敷道中心距,并结合熔敷道截面积信息对其加以修正,使检测结果更接近实际值。通过阶跃试验辨识了熔敷道尺寸与熔敷规范参数之间的传递函数,制定了以“控制熔敷道宽度为主要目标,控制熔敷道高度与熔敷层表面平整度为次要目标”的控制策略,结合人机协作控制思想,设计了由熔敷速度控制熔敷道宽度的单神经元PI控制器以及由熔敷电压调整熔敷道高度的经验规则控制器,同时由操作者调整熔敷道中心距,保证各个熔敷道平整搭接。最后,进行了多层两道结构的闭环控制熔敷试验,成形件熔敷道宽度偏差小于0.5mm,高度偏差无明显积累,同时熔敷层上表面平整,表明采用本文所设计的多层多道GMA-AM熔敷道尺寸控制器能够保证成形件尺寸与规划结果相吻合,达到了提高GMA-AM成形精度的目标。
[Abstract]:MIG arc increasing material manufacturing (Gas Metal Arc-Additive Manufacturing, GMA-AM) is a kind of metal material increase in material manufacturing process, has low production cost, high deposition efficiency, flexibility and other advantages, however, the melting process is vulnerable to interference, deposited Chi Cunrong easy to deviate from the value of planning, with the continuous size deviation the accumulation of subsequent deposition process may not be carried out according to the plan, the forming size will vary greatly with the planning results. Therefore, improving the forming accuracy of GMA-AM is a problem to be solved urgently. In this paper, a set of intelligent GMA-AM system is established, and the information detection and closed loop control of multi-layer and multi-channel GMA-AM forming is studied. Firstly, combined with the GMA-AM requirements, the arc welding robot technology and active vision sensing technology based on the establishment of a set of intelligent GMA-AM system design and the system can gradually deposited in the process of real-time detection of weld size information in the man-machine interface display and store at the same time, the operator has online regulation of cladding parameters interface, support man-machine collaboration, the operator and the machine can make decision on the deposition process, in order to achieve closed-loop control of cladding size and lay a good foundation. In the deposition process of weld cross section contour information extraction using META line structure light sensor was developed based on the real-time multi pass GMA-AM cladding size detection algorithm of point cloud data, according to the morphological characteristics of structured light stripe extraction, cladding road vertex and weld toe feature from point cloud data to calculate the cladding height, width and cladding cladding road center distance, and combined with the weld cross section area of the corrected information, making the results more close to the actual value. Through the step test identification of the transfer function between the cladding channel size and cladding parameters, to set up control weld width as the main target, the control strategy of cladding height and smooth surface cladding layer as a secondary target, combined with collaborative control thought, design of a single neuron PI the controller consists of melting speed control cladding width and cladding by voltage adjusting cladding height controller experience road rules, at the same time by the operator to adjust the weld road center distance, to ensure the smooth lap weld. Finally, the two closed-loop control structure of multilayer cladding forming test, weld width deviation is less than 0.5mm, the height deviation without obvious accumulation, and the cladding layer on the surface, show that the designed multi-layer cladding GMA-AM controller can guarantee the size of parts size and planning results GMA-AM, to improve the forming accuracy of target.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TG409;TP242

【相似文献】