Automated manufacturability analysis in smart manufacturing systems: a signature mapping method for product-centered digital twins
Automated manufacturability analysis in smart manufacturing systems: a signature mapping method for product-centered digital twins
复制标题
DOI：
10.1007/s10845-022-01991-4
复制
发表时间：
2022-08
影响因子：
8.3
通讯作者：
K. Xia;Thorsten Wuest;R. Harik
中科院分区：
文献类型：
--
作者：
K. Xia;Thorsten Wuest;R. Harik
开发负担得起的和可定制的网络物理生产系统和数字孪生（DT）实施为当前的工业4.0和智能制造计划注入了新的活力。精确处理材料处理过程以进行可制造性分析的能力进一步连接了当今智能制造系统的物理和网络组件。在这项工作中，我们提出了一个以产品为中心的签名映射方法，自动数字孪生具有智能传感，基于签名的特征提取器和知识分类的混合实现。首先，我们集成了3D扫描和表面重建，以实现从虚拟环境（计算机辅助工程数据）和现实世界的生产环境（扫描点云帧）中检索形状。其次，Shape Terra是一种用于内部曲率的算法，它模拟持久热量值，以便从检索到的形状文件中快速提取签名。最后，一个系统的集成建议的形状分析的基础上知识分类的原型实现。该测试平台的目的是说明一个概念验证DT辅助工艺自主性快速3D表面签名喂养。因此，通过混合智能传感和模拟方法，我们利用形状签名作为制造知识，通过集成领域知识和数据驱动的决策。此外，人机互操作性使系统级智能控制在复杂的材料处理、成形、测量和检测过程中变得可行。
Developing affordable and customizable cyber-physical production system and Digital Twin (DT) implementations infuses new vitality for current Industry 4.0 and Smart Manufacturing initiatives. The ability to precisely address material handling processes for manufacturability analysis further connects the physical and cyber components of today’s smart manufacturing systems. In this work, we propose a product-centered signature mapping approach to automated digital twinning featuring a hybrid implementation of smart sensing, signature-based feature extractor, and knowledge taxonomy. First, we integrate 3D scanning and surface reconstruction at to implement shape retrieval from both the virtual environment (from Computer-Aided Engineering data) and the real-world production environment (from scanned point cloud frames). Second,Shape Terra, an algorithm for intrinsic curvatures, simulates Persistent Heat Values for fast signature extraction from retrieved shape files. Finally, a systematic integration of the proposed shape analysis based on knowledge taxonomy is prototypically implemented. The objective of this testbed is to illustrate a proof-of-concept DT-aided process autonomy fed by rapid 3D surface signatures. As a result, by hybridizing smart sensing and simulative approaches, we exploit shape signatures as manufacturing knowledge by integrating domain knowledge and data-driven decision-makings. Moreover, human–machine interoperability enabling system-level intelligent controls becomes feasible in complex material handling, shape forming, measuring, and inspection processes.