(332b) The Role of Modeling and Scaled Experiments in Advanced Nuclear Reactor Design and Safety Analysis

The pebble-bed fluoride-salt-cooled, high-temperature nuclear reactor (PB-FHR) concept is currently under development at UC Berkeley, in collaboration with other universities and national laboratories. It is an advanced nuclear reactor design, with the objective of providing an energy technology with a short commercialization timeline and significant safety advantages compared to advanced light water nuclear reactors currently under construction, at costs that are competitive with natural gas power plants.

As the design of large complex engineered systems progresses, changes become increasingly costly and challenging to implement. Methodologies that are common practice for safety analysis of nuclear reactors, such as Code Scaling, Applicability and Uncertainty (CSAU), Phenomena Identification and Ranking Table (PIRT), and Hierarchical, Two-Tiered Scaling (H2TS), while flexible in the sense that they are technology-neutral, require well defined designs. This pushes system designers towards incremental changes to evolutionary systems, which rely on experience with previous technology to design for safety and other regulatory performance requirements. Thus, tools that enable integration of safety analysis and other types of risk analysis very early in the design process of a complex engineered system are needed for the development of revolutionary technologies. This presentation addresses the question of how safety analysis is integrated very early in the design phase of a nuclear reactor, such as the PB-FHR, and the role of modeling and experimentation in this process.

Nuclear reactor design and safety analysis activities can be grouped into three categories: system design, modeling of system response to specific scenarios, and characterization of individual phenomena. The role of modeling and experimentation in demonstrating reactor safety and developing a reactor design is illustrated by the interaction among these three activities. System design is iterative with identification of system states and transient scenarios. Accurate modeling of the overall system response to specific scenarios relies on simplified models for the behavior of subsystems and components. System transient models must be validated by experiments termed integral effects tests (IETs), which demonstrate that the system model accurately predicts the overall behavior of the system. Characterization of phenomena focuses on establishing the knowledge base for the underlying individual phenomenology. It can be based on empirical measurements, analytical or numerical models that require experimental validation, or semi-empirical models. Experiments in support of phenomena characterization are termed separate effects tests (SETs). Interactions among these activities follow a combination of top-down and bottom-up approaches, which are discussed here.