Automated Non Destructive Testing for the Aerospace Industry

该项目的目的是使航空航天工业中使用的无损检测(NDT)系统自动化，以简化检查程序，并提高系统的灵敏度、可靠性和重复性。航空零部件在生产过程中的质量认证和在服役中对这些零部件造成的缺陷的评估是经常依赖无损检测的关键任务。目前，这些任务是由人类检查员执行的，其多样化的经验可能导致结果解释的不同。此外，分配给这类任务的人力会产生大量的经常性费用，而且很耗时。参与该项目的工业合作伙伴L-3MAS和普华永道已经确定了两种将从自动化中受益的常见方法：荧光渗透检测和涡流检测。一个标准的机器人系统将被用来处理与每种技术相关的探测器。根据3D激光系统提供的数据并使用CAD文件(如果有)，将生成初始扫描路径。如果CAD文件不可用或不可靠(尤其是对于正在使用或修改的部件)，则将使用初始或最大信封体积。数据融合算法将被用来建立被检查表面的缺陷图，并在随后的扫描中改进扫描路径。首先，智能专家系统将基于对探头中信号的解释为用户提供关于是否存在缺陷的决定。该项目的最终结果将是一个功能齐全的小型机器人无损检测检测工作台，作为与人类检查员进行比较的基准。此外，还建议对红外热成像等新型无损检测方案进行评估。这一新的方案允许从感应涡流中收集热红外信号。

The aim of this project is to automate non-destructive testing (NDT) systems used in the aerospace industry so as to ease inspection procedures as well as to enhance systems sensitivity, reliability and repeatability. Quality certification of aeronautical parts in production and evaluation of defects created on these parts in service are critical tasks frequently relying on NDT. Currently, these tasks are, performed by human inspectors whose diversified experience may lead to variation in results interpretation. In addition, manpower assigned to such tasks results in significant recurrent costs and is time consuming. L-3 MAS and PWC, the industrial partners involved in this project,have identified two common methods that wouid benefit from automation: Fluorescent Penetrant Inspection (FPI) and Eddy Current Testing (ECT). A standard robotic system will be used to handle the probe associated with each technique. Based on data provided by a 3D laser system and using CAD files if available, an initial scanning'path will be generated. In case CAD files are not available or not reliable (particularly for in-service or modified components), an initial or largest envelope volume will be used instead. Data fusion algorithms will be used to build a defect map of the inspected sllrface and to refine the scanning path in subsequent passes. Firially, an intelligent expert system will provide the user with a decision on whether a defect is present or not based on the interpretation of the signals froin the probes., The final result of this project will be a functional small-scale robotic NDT inspection workcell serving as a benchmark for comparison to human inspectors. Additionally, it is proposed to evaluate novel NDT schemes such as Infrared Thermography. This novel scheme allows to collect thermal infrared signatures from induced Eddy currents.