NIST 先进制造数字孪生项目：为制造业数字孪生制定标准与验证方法
Digital Twins for Advanced Manufacturing
NIST 的“先进制造数字孪生”项目开发技术指南并推动标准制定，帮助制造商明确数字孪生需求、有效管理数据，并构建可量化不确定性的模型。项目包含三部分：研发数字孪生实施与测试方法、推进标准开发与测试、利用 NIST 数字孪生测试平台支持参考实现。其工作涵盖 ISO 23247 第 7 部分 VVUQ 框架新标准提案，以及第 5 部分数字线程和 MTConnect 接口的参考实现。
研究资料选读：用于理解制造AI的数据、验证与标准条件。原文为研究计划或项目说明，不代表生产效果已验证。
阅读全文（需登录） →登录一次，后续文章自动展开正文
Manufacturing is undergoing a digital transformation. At the center of this transformation are digital twins: synchronized virtual models that help manufacturers represent, diagnose, predict, and optimize their operations. Digital twins can help increase production, improve product quality, and reduce manufacturing costs. But building digital twins correctly is challenging, especially for small and medium-sized enterprises. A major barrier is the lack of standards, including common rules for vocabulary, design, interoperability, trustworthiness, and methods and procedures for verifying and validating digital twins.
The Digital Twins for Advanced Manufacturing project develops technical guidance and contributes to standards that help manufacturers identify digital twin requirements, manage and use data effectively, and create models that can be validated with quantified uncertainties. The research plan has three main parts: (1) research and develop methodologies to help companies implement and test digital twins, (2) advance digital twin standards development and testing, and (3) use a NIST Digital Twin Testbed to support research and reference implementations. The goal is to make digital twins easier to develop and validate, more reliable, and more widely adopted thereby helping U.S. manufacturers stay competitive.
Objective
To provide measurement science and open standards to help U.S. manufacturers better define, measure, analyze, and control advanced manufacturing systems using trustworthy digital twins and enable a marketplace for digital twin users and technology providers.
Technical Idea
This project will advance the use of digital twins in manufacturing by developing the measurement science and standards needed to make them reliable, interoperable, and trustworthy. By building on recent advances in smart sensors, the Industrial Internet of Things (IIoT), artificial intelligence, and modeling and simulation, the project will focus on ensuring that digital twins can accurately represent and support complex manufacturing systems.
Project products will include:
- Reference architectures and standards that define how digital twins should integrate data across machines, processes, and lifecycle stages, and how Verification, Validation, and Uncertainty Quantification (VVUQ) should be performed for data, models, and digital twin results.
- Testbeds and reference implementations to demonstrate the developed/proposed standards for evaluating the accuracy, trustworthiness, and interoperability of digital twins, providing valuable feedback to Standards Development Organizations for improving the quality of the standards.
These products will address core measurement science challenges by providing a system-of-systems framework to integrate multiple subsystems into a coherent digital twin, ensuring consistency and credibility. In parallel, the project will develop a lifecycle integration approach to link digital twins across design, production, and maintenance stages using digital threads, reducing redundant data exchange and improving traceability. Together, these products will create the foundation for a marketplace where digital twin users and technology providers can confidently exchange solutions. The resulting standards and tools will improve the agility and flexibility of advanced manufacturing systems and strengthen the competitiveness of U.S. industry.
Research Plan
This project seeks to continuously lead and contribute to the development and testing of new digital twin-related standards and use the digital twin testbed for validating and testing these standards.
- New work item for a guideline standard “Verification, Validation and Uncertainty Quantification (VVUQ) Framework for Digital Twins in Manufacturing”, Part 7 of ISO 23247 Digital Twin Framework for Manufacturing – Standard.
- User case implementation based on the proposed VVUQ framework to integrate VVUQ processes into digital twin workflows, ensuring model credibility, quantified uncertainty, and traceable results in manufacturing applications - Reference implementation.
- Use case implementation of ISO 23247 Part 5 - Digital Thread for Digital Twins to demonstrate how to implement a digital thread that supports digital twins in manufacturing, specifically focusing on data flow, traceability, and lifecycle integration across multiple systems - Reference implementation.
- MTConnect interface development for digital twin implemented and tested in the Lab.
Highlights
- ISO 23247 (“Digital Twin Framework for Manufacturing”) was published in 2021.
- Shao, G. and Helu, M. 2020. Framework for a digital twin in manufacturing: Scope and requirements. Manufacturing Letters 24, 105-107. https://doi.org/10.1016/j.mfglet.2020.04.004
- Shao, G., Hightower, J. and Schindel, W. 2023. Credibility consideration for digital twins in manufacturing. Manufacturing Letters 35, 24-28.
- Shao, G. 2021. Use case scenarios for digital twin implementation based on ISO 23247. NIST Advanced Manufacturing Series (NIST AMS) - 400-2
- Lin, SW., Watson, K., Shao, G., Stojanovic, L., and Zarkout, B. 2023. Digital Twin Core-Essential Models for Interoperability. An OMG IIC Technical Report
来源：NIST · 制造研究 · nist.gov