Multi-objective optimization of maintenance program in multi-unit nuclear power plant sites. (August 2019)

• Record Type: Journal Article
• Authors: Zhang, Sai; Du, Mengyu; Tong, Jiejuan; Li, Yan-Fu
• Is Part Of: Reliability engineering & system safety. Volume 188(2019)
• Page Start: 532
• Page End: 548
• Publication Date: 2019-08
• Subjects: Multi-Unit Probabilistic Risk Assessment (MUPRA) -- Multi-Objective Optimization (MOO) -- Fast Non-dominated Sorting Generic Algorithm (NSGA-II) -- Maintenance optimization -- Nuclear Power Plant (NPP) -- High-Temperature Gas-cooled Reactor (HTGR); Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2019.03.034
• Languages: English
• ISSNs: 0951-8320

Abstract

Highlights: Maintenance program in multi-unit nuclear power plant sites is selected as the target of optimization. A tri-objective optimization problem is formulated to minimize unavailability, risk and cost. Multi-Unit Probabilistic Risk Assessment is utilized to assess single-unit and double-unit risks. Problem-specific Fast Non-dominated Sorting Generic Algorithm (NSGA-II) is designed to solve the optimization problem. Optimization problem is formulated and solved under aleatory and epistemic uncertainties. Abstract: Maintenance program optimization of nuclear power plants (NPPs) has been a research focus over the past decades, and the existing works are mostly conducted with a one-reactor-at-a-time presumption. Multi-unit NPP site design (i.e., a single site houses multiple reactors), however, is a common case, in which the reactors are not independent from each other, rather, connected by complex intra- and inter-unit mechanisms. To bridge the research gap and generate practically useful results, a methodology of conducting multi-objective optimization for maintenance program in the context of multi-unit NPP sites is proposed in this research. The maintenance optimization is formulated as a tri-objective scheme aiming at minimizing multi-unit unavailability, site-wide risk and cost. Case studies are conducted on feedwater systems adapted from a real-world two-unit NPP site with and without uncertainties. It can be concluded that, for the case studies in this paper, (i) risk attitudes, in the form of weighting factors of risk types with radiological consequences of different severities, of NPP decision makers and regulators could notably affect the optimal maintenance scheduling and the projected objective values; (ii) the optimization model under uncertainties can be taken as a generalization of that without uncertainties and could have the chance of discovering new Pareto-optimal solutions leading to lower costs without compromising multi-unit unavailability or risk.