基于视觉感知的电弧增材制造尺寸控制方法研究  

Research on Dimensional Control Method for Wire Arc Additive Manufacturing Based on Visual Perception

作　　者：徐磊 Xu lei(Jilin Fusheng Automotive Lamp Co.,Ltd.,Jilin Jilin 132000)

机构地区：[1]吉林富晟车灯有限公司,吉林吉林132000

出　　处：《内燃机与配件》2025年第6期62-64,共3页Internal Combustion Engine & Parts

基　　金：一汽红旗H9轿车ADB前灯设计及制造工艺研究;国家新型显示与战略性电子材料专项;华为赛力斯问界M9 DLP投影大灯设计和开发;上汽智己DLP投影大灯制造工艺可行性研究。

摘　　要：铝合金已成为汽车零部件轻量化设计的重要材料,其结构的先验验证依赖于增材制造技术,本文根据铝合金电弧增材制造工艺成形特点,研制了一套适合该工艺的视觉传感在线形貌监测系统。基于连续光谱的带通复合滤光双视觉传感技术,在电弧增材制造平台上搭建了沉积层形貌视觉传感实验平台,获得了电弧下部局部成形区域的沉积层形貌动态信息。采用Delphi&Halcon联合编程方法及模块化设计,结合高斯线性平滑滤波加权因子及亚像素级边缘信息实时提取测量算法,提高了成形过程中强弧光下沉积层形貌辨识的准确性,实现了成形过程中沉积层形貌的精确获取,为后期零部件的小批量生产奠定研究基础。Based on the forming characteristics of the aluminum alloy arc additive manufacturing process,a set of visual sensing online topography monitoring system suitable for this process has been developed.Utilizing band-pass composite filtering dual-vision sensing technology with continuous spectrum,a visual sensing experimental platform for deposition layer topography has been established on the arc additive manufacturing platform,which obtains dynamic information on the deposition layer topography within the locally formed area beneath the arc.By adopting a joint programming method with Delphi and Halcon,along with modular design,and combining Gaussian linear smoothing filtering with weighted factors and real-time extraction and measurement algorithms for sub-pixel edge information,the accuracy of deposition layer topography identification under intense arc light during the forming process has been improved.This has enabled precise acquisition and real-time control of the deposition layer topography during the forming process.

关 键 词：铝合金 电弧增材制造 视觉感知 成形形貌 

分 类 号：TH164[机械工程—机械制造及自动化]