Multi-dimensional Digital Twins for Nuclear Power Plants

该提案旨在深入调查、开发和使用数字双胞胎在核电站(NPP)系统结构和组件(SCCs)资产管理的背景下。这项提议的主要目标是：(1)为核电厂关键资产的数字孪生兄弟开发一个合适的系统架构，包括收集和融合信息所需的软件和硬件框架；以及(2)根据当前和正在出现的数字孪生兄弟在核电厂SSC中的使用情况开发一个决策支持系统。为了实现这些目标，我们将从一个全面的数据收集阶段开始，在这个阶段将确定安全和安全部门内的关键资产、系统和程序之间的中心关系。在此之后，将进行全面的风险分析和信息价值研究，以确定哪些SSC需要数字孪生兄弟，以及应在这种虚拟表示中包括哪些信息。在这项活动之后，将进行深入研究，将感觉信息融合到数字孪生表示中，这将作为原始信息源和资产决策支持工具之间的节点。最后，将开发一个框架，通过使用基于机械和物理的模型来丰富数字双胞胎。这将包括开发算法，将感觉信息直接融合到机械模型(例如，有限元)中，使用检查信息定期更新此类模型(以促进条件评估和退化模型的开发)，以及计算高效的重新分析技术。拟议的工作将直接支持我们的行业合作伙伴在其整个生命周期中优化资产管理活动，同时保持核电厂SSC所需的高水平安全和可靠性。此外，这样的数字双胞胎最终将补充我们的行业合作伙伴在模型设施上的约7亿美元投资，同时还确保有效的风险管理、公共问责、资产管理和规划。最终，该项目将向几名研究生传授高质量的跨学科培训和技能，并为他们在资产管理领域的下一代工作做好准备。

This proposal is aimed at an in-depth investigation, development, and the use of digital twins in the context of nuclear power plant (NPP) systems structures and components (SCCs) asset management. The main objectives of this proposal are to: (i) develop a suitable system architecture for digital twins for key NPP assets, including the requisite software and hardware framework for collecting and fusing information; and, (ii) develop a decision support system based on current and emerging uses for digital twins in NPP SSCs. In pursuit of these objectives we will start with a comprehensive data collection phase, where the central relationships between key assets, systems and processes within the SSCs will be defined. Following this, a comprehensive risk analysis and value-of-information study will be carried out to prioritize which SSCs warrant a digital twin and what information should be included in such a virtual representation. This activity will be followed by an in-depth study to fuse sensory information into the digital twin representation, which will act as a node between raw information sources and asset decision support tools. Finally, a framework to enrich digital twins with the use of mechanistic and physics based models will be developed. This will involve the development of algorithms to directly fuse sensory information into mechanistic models (e.g., finite-element), update such models periodically using inspection information (to facilitate condition assessment and degradation model development) and computationally efficient re-analysis techniques. The proposed work will directly support the optimization of asset management activities for our industry partner throughout its life cycle, while maintaining a high level of safety and reliability required in NPP SSCs. In addition, such digital twins will ultimately complement the ~ $700M investment by our industry partner in mock-up facilities, while also ensuring effective risk management, public accountability, asset management, and planning. Ultimately, this project will impart high-quality inter-disciplinary training and skills to several post-graduate students and will prepare them for the next-generation jobs in asset management