A nearly autonomous management and control (NAMAC) system is designed to furnish recommendations to operators for achieving particular goals based on NAMAC's knowledge base. As a critical component in a NAMAC system, digital twins (DTs) are used to extract information from the knowledge base to support decision-making in reactor control and management during all modes of plant operations. With the advancement of artificial intelligence and data-driven methods, machine learning algorithms are used to build DTs of various functions in the NAMAC system. To evaluate the uncertainty of DTs and its impacts on the reactor digital instrumentation and control systems, uncertainty quantification (UQ) and software risk analysis is needed. As a comprehensive overview of prior research and a starting point for new investigations, this study selects and reviews relevant UQ techniques and software hazard and software risk analysis methods that may be suitable for DTs in the NAMAC system.
翻译:近乎自主的管理和控制系统(NAMAC)旨在向操作者提供建议,以便根据NAAC的知识基础实现特定目标,作为NAAC系统的一个关键组成部分,数字双胞胎(DTs)用于从知识库提取信息,以支持所有工厂作业模式中反应堆控制和管理的决策;随着人工智能和数据驱动方法的发展,机器学习算法被用于建立NAAC系统中各种功能的DT;评估DT的不确定性及其对反应堆数字仪器和控制系统的影响,需要不确定性量化(UQ)和软件风险分析;作为以前研究的全面概览和新调查的起点,本研究选择并审查了适合NAAC系统中的DT的UQ技术和软件危害及软件风险分析方法。