A Resilience Modelling Framework for Improved Nuclear Safety (NuRes)

Accidents at nuclear plants, such as those at Fukushima and Chernobyl, have increased the public awareness of the severe consequences that can result when system failures occur. However, as the demand for energy increases and low-carbon sources are required, many countries, including the UK and India, see nuclear power generation as an important contributor to meeting these needs.

Risk analysis methods, which originated in the 1970s, require the evaluation of the frequency and consequences of the potential hazards which can occur on nuclear systems. It is these methods which have been used, and are still used, to ensure the safety of nuclear power generation. However, since the conception of the risk analysis approaches, the characteristics of engineering systems have undergone significant changes due to the advances which have occurred in technology. Computer control systems and the use of autonomous systems are now common and this introduces new vulnerabilities to the system. The range of failures and threat events which can cause safety issues is increasing with newly emerging threats due to the severe weather conditions associated with global warming and deliberate terrorist attacks and cyber-attacks of increasing concern. It is likely that new, currently unknown, threat types will continue to emerge. These changes in the systems, their vulnerabilities and their threats mean that new approaches, capable of dealing with these new requirements are needed to ensure the safety and security of nuclear energy production for future reactors.

Resilience engineering is considered to offer significant benefits when considering the effectiveness of safety critical systems on potentially hazardous plants. This approach looks at designing systems which are capable of experiencing threats and have several approaches (known as dimensions) which enable the system to avoid, withstand, adapt to or recover from their effects.

This project examines the benefits that resilience engineering could offer in the context of nuclear safety systems. It indicates the models and data required to predict the resilience of a nuclear power generation plant. Such models will be formulated and applied to a demonstrator system. Through this predictive tool modern nuclear systems can be designed and operated to achieve the high levels of safety demanded. Special attention in the framework will be given to deliberated, intended cyber-attacks and also the role in which humans can play in the recovery of the system following a threat.

This research has the potential to contribute to and have impact in three different areas: the advancement of academic knowledge, the safe and cost effective operation of nuclear power plant and the safe and efficient operation of safety critical systems in other industries. Specifically the impact will be:

i. to facilitate the safe, reliable and cost-effective operation of nuclear power plant in the UK and India as a result of a resilience engineering approach which has the potential to accommodate the characteristics of modern engineering systems.
ii. to enable a safe response to be formulated to the full range of treats that can be experienced by nuclear power plant systems. This includes known and unknown threat types. The frequency of unknown threat types has no meaning when considering the traditional risk based approach to safety assessment and an alternative approach to such disruptive occurrences has to be found.
iii. to define resilience, the dimensions of resilience and the key performance metrics in the context of nuclear engineering.
iv. to produce a modelling framework which represents a step change in nuclear safety quantification through a resilience engineering approach.
v. to better understand the effects that cyber-attacks can produce on a nuclear power plant along with the most effective strategies to make the system robust and resilient to their occurrence.
vi. to better understand how human actions can help mitigate or prevent the effects of threats on the nuclear systems. In addition an understanding will be provided as to how these can be modelled to establish the most effective options.
vii. to enable the most effective use of the decision support tools to fix the design of future reactors along with the threat recovery strategies. This requires a good understanding of the uncertainties which exist in the predictions made, especially when considering that the threat types experienced are constantly changing and may currently be unknown. The understanding of the uncertainties contributes to the effective definition of a robust power generation process.
viii. All aspects outlined in the points above which are expected to yield impact from the project, such as: resilience quantification, cyber-threats, unknown threats, human factors and uncertainty are not unique to the nuclear industry. As such the broader impact from this project will be significant, being directly applicable to any other safety critical industry including: aeronautical, oil and gas, chemical processing, transport, water distribution, power distribution, and other forms of power generation.
ix. The general public will be direct beneficiaries of this research through the development of safer, reliable and cost effective energy regardless of new threats which emerge.
x. An increase in the UK/India knowledge base and the training of skilled young scientists for potential recruitment by the nuclear industry or academia.