Presentation
CANCELED - Context-aware Procedure System for Advanced Reactors Operations Using Large Language Models
SessionPoster Session 2
DescriptionAdvances in digital human-machine interfaces (HMIs) for nuclear process control typically use a static windowed display approach to present system information to operators. Future advanced control systems are beginning to incorporate machine learning to classify plant states, automate certain control actions, or recommend courses of action. These automatic functions can offload many of the manually performed monitoring and control functions performed by operators. However, increased digitalization and automation do not obviate the need for human operators in nuclear power plants. In fact, these new technologies impose new demands on operators as they must now oversee the automatic systems within the context of available process parameters to determine if they concur with its classification and decisions for executed or recommend actions. Furthermore, aims to reduce staffing may entail operators overseeing multiple reactors. As such, these new demands necessitate interfaces capable of dynamically presenting pertinent plant state information to reduce the workload and allow the operator to maintain oversight of the process. A context-aware procedure system (CAPS) is proposed to illustrate what must be considered to adopt machine learning within the control room as a steppingstone for aligning the role of the human operator with emerging artificial intelligence technologies for plants.
Event Type
Poster
TimeWednesday, October 15th5:30pm - 6:30pm CDT
LocationRiverside East
Similar Presentations