An Integrated Risk Assessment Process for Digital Instrumentation and Control Upgrades of Nuclear Power Plants

Abstract

Most of the existing nuclear power plants (NPPs) in the world rely on traditional analog instrumentation and control (I&C) systems for monitoring, control, and protection functions. With the industrial base largely moving to digital systems, the operation and maintenance of plants involves managing issues including lack of needed analog spare parts, increasing maintenance costs, and the loss of vendor support. Compared with existing analog I&C systems, digital I&C systems have significant functional advantages, such as reliable system performance in terms of accuracy and computational capability, high data handling and storage capabilities to fully measure and display operating conditions, and improved capabilities (e.g., fault tolerance, self-testing, signal validation, process system diagnostics). Therefore, the U.S. nuclear power industry has initiated the replacement of existing, aging analog systems with digital I&C technology, and is developing new designs for advanced plants using digital systems in integrated control rooms to provide modern control and protection systems. However, the qualification of digital I&C systems remains a challenge, especially the issue of software common cause failure (CCF), which has been difficult to address.Upgrading digital I&C (safety and non-safety-related) systems in existing NPPs within a cost-effective and regulatory acceptable way offers the foremost means of performance improvements and cost-reductions for existing NPPs. One key outcome of this project is to perform plant-specific risk assessment to provide a sustainable scientific support for enabling industry to balance the digital-related risks, costs, reliability, and safety.

Uncertainty quantification and software risk analysis for digital twins in the nearly autonomous management and control systems: A review

“…However, some trivial assumptions made or restrict requirements followed during software development life cycle (SDLC) may not meet some extreme conditions or unexpected conditions in new operating environment due to upgrades. It is shown that a majority of software issues causing serious accidents have been investigated as design defects or incompleteness of requirements in SDLC, other than in the implementation process of these requirements [84].…”

“…Light Water Reactor Sustainability (LWRS) Program launched a project that aims to provide solid technical supports and risk-informed insights for the development and deployment of safe, secure, efficient and licensable digital I&C technologies in NPPs. [84] [85] An integrated Risk Assessment process for Digital I&C (RADIC) was proposed in this project including two phases: risk analysis and risk evaluation. Risk analysis aims to identify hazards of digital-based SSCs, estimate their failure probabilities, and analyze relevant consequences by performing hazard analysis, reliability analysis, and consequence analysis.…”

“…The respective acceptance criterion is defined as, "How reliable is the digital system with the identified digital failures existing?" According to [84], quantification of failure probabilities is the main outcome for both single failure events and consequences. For different designs and requirements from licensing regulators, the set point for reliability probability should integrate the efforts and experiences from industry, regulators, and researchers.…”

Development and assessment of a nearly autonomous management and control system for advanced reactors

“…Compared to the classical training for ML-based DTs, denoted as the base learner, there is also a meta model that optimizes the base learner by updating its parameters via a meta-knowledge base [27]. In addition to the separate DTs development and assessment, the quantitative software reliability method [28] and digital instrumentation & control (I&C) assessment [29] evaluate the reliability and risk of the autonomous control system and digital I&C by quantifying the software failure rates and the risks to reactor component and system. As a result, at the refinement stage, the impacts of uncertainty and bias on operators' decisions and directly-connected DTs are known, and resources can be directed to areas with insufficient knowledge.…”

Digital Twin Concepts with Uncertainty for Nuclear Power Applications

“…Further, the NRC also recently updated their standard review plan for digital instrumentation and control systems that would necessarily be a part of DTs. The DOE has also recently launched the Risk-Informed Systems Analysis project that developed the Risk Assessment process for Digital I&C (RADIC) [73] which discusses digital-based systems, structures, and components that would exist within a DT. Lin et al [19] discuss these and other approaches to the integration of DTs with risk assessments in detail.…”