Markov Approach to Evaluate Physical Protection of Nuclear Energy Systems

A Markov-based methodological approach has been developed to assess the probabilistic outcomes of various damage states as a consequence of security-related events at nuclear energy systems. Sabotage on an advanced small modular reactor (SMR) is used to demonstrate the Markov approach to model security risk. The methodology adapts the proliferation resistance and physical protection (PR&PP) evaluation framework developed by the PR&PP Working Group of the Generation IV International Forum (GIF). In the PR&PP framework, a PP pathway analysis identifies potential pathways by which the threat could access and disable the vital equipment, evaluates the response of the physical protection system (PPS), and assesses the PP measures to determine the attractiveness of the pathways to potential PP adversaries. A Markov chain is a systematic approach to implement and quantify the pathway analysis embedded in the GIF PR&PP methodology. Using the Markov approach, security and safeguards scenarios can be modeled by discrete stages that represent major activity modules and end points in the system. As the sabotage evolves over time the Markov model provides the probability of each state of the plant as a function of time. As a feasibility demonstration of the Markov approach to PP, it has been applied in this study to a hypothetical advanced SMR, an Example Sodium Fast Reactor (ESFR), developed by the GIF PR&PP Working Group for methodology development and demonstration. The Example Sodium Fast Reactor (ESFR) is a modular nuclear system. Each plant system consists of four sodium-cooled fast reactors of medium size (300 MWe each) co-located with a dry fuel storage facility and a pyrochemical spent-fuel reprocessing facility. The sabotage scenario involves the disabling of the decay heat removal function at one of the four ESFR reactors and the subsequent failure of multiple barriers to radiological releases from the nuclear fuel. The failures of intrinsic barriers to radiological releases include the losses of fuel and primary coolant boundary integrity and the containment integrity. In the Markov model it is assumed that the targets of attack have been identified by the defined adversary with enough resources (knowledge, hardware, and manpower) to penetrate the protective barriers, both intrinsic and extrinsic. The response of the system is evaluated considering the potential damages caused by the physical attack, available mitigation, and interdiction of the adversary by the response force. For the analysis of a sabotage event the Markov model is constructed by overlaying the PPS (extrinsic barriers) over the plant systems. Different pathways are used to model the physical attack on the plant systems and the responses of the PPS in terms of detection, delay and interdiction. Separately but embedded in the overall Markov model is a phenomenology based plant degradation model that represents the intrinsic barriers and identifies different pathways that can lead to potential off-site consequences. Application of the Markov model to evaluate the sabotage on the ESFR demonstrates the integration of the PPS and the plant degradation model for analyzing the response of the overall plant system to a security event. This study also shows the Markov approach is completely compatible with the conventional source term methodology. Pathways that can lead to potential off-site consequences are formed by connecting segments that constitute a phenomenology-based plant degradation model. In addition, different categories of radiological consequences, such as early and late releases, can be mapped to specific Markov states based on the pathways driven by the melt/release progression. The Markov model has the potential to inform the development of strategies for emergency planning efficiencies in terms of size and timing of on-site and off-site actions and resources. This can be accomplished by binning the Markov states to the emergency action levels. In essence the Markov model provides a holistic approach for stepping through the emergency planning process, taking into consideration the potential evolution of a threat into multiple outcomes. (author)