Intelligent Modeling for Nuclear Power Plant Accident Management

Abstract

This paper explores the viability of using counterfactual reasoning for impact analyses when understanding and responding to “beyond-design-basis” nuclear power plant accidents. Currently, when a severe nuclear power plant accident occurs, plant operators rely on Severe Accident Management Guidelines. However, the current guidelines are limited in scope and depth: for certain types of accidents, plant operators would have to work to mitigate the damage with limited experience and guidance for the particular situation. We aim to fill the need for comprehensive accident support by using a dynamic Bayesian network to aid in the diagnosis of a nuclear reactor's state and to analyze the impact of possible response measures. The dynamic Bayesian network, DBN, offers an expressive representation of the components and relationships that make up a complex causal system. For this reason, and for its tractable reasoning, the DBN supports a functional model for the intricate operations of nuclear power plants. In this domain, it is also pertinent that a Bayesian network can be composed of both probabilistic and knowledge-based components. Though probabilities can be calculated from simulated models, the structure of the network, as well as the value of some parameters, must be assigned by human experts. Since dynamic Bayesian network-based systems are capable of running better-than-real-time situation analyses, they can support both current event and alternate scenario impact analyses.

Technologies to Reactors: Enabling Accelerated Deployment of Nuclear Energy Systems Workshop Report

“…An advanced AI algorithm can provide efficient fault diagnosis and help in determining efficient layout for sensor placement. Darling et al 24 have illustrated a technique to determine the key target variables for sensory output using information metrics such as Kullback-Leibler (KL) divergence with machine learning techniques (dynamic Bayesian network).…”

Building and using dynamic risk-informed diagnosis procedures for complex system accidents

“…For example, if the system is in a particular state, what would happen if the control engineers take specific steps to protect the reactor from a dangerous situation such as LOF? Additional examples of prognostic reasoning using this model can be found in Darling et al 42 Key next steps include examining the impact of time discretization and reactor parameter discretization 43 and to assess the impact of different modeling choices with regard to predictive power, accuracy, solution time, and error rate of the predictions. There are many changes that should be explored quantitatively, including variations in the number and nature of scenarios and components, the number and parameters of simulations, the source of the PRA data, and more.…”

“…For example, if the system is in a particular state, what would happen if the control engineers take specific steps to protect the reactor from a dangerous situation such as LOF? Additional examples of prognostic reasoning using this model can be found in Darling et al 42…”

Digital-twin-based improvements to diagnosis, prognosis, strategy assessment, and discrepancy checking in a nearly autonomous management and control system

“…K. Hu and J. Yuan [6] developed a multi-model predictive control method for nuclear steam generator water level by approximating the system behaviors with a polytopic uncertain linear parameter varying model. Darling et al [7] developed a fault detection and management system for sodium fast reactors using dynamic probabilistic risk assessments and counterfactual reasoning. Cetiner et al [8] developed a supervisory control system for the autonomous operation of advanced small modular reactors.…”