Maintenance strategy design for nuclear reactors safety systems using a constraint particle swarm evolutionary methodology

Nuclear power plants (NPPs) in the world should be equipped with many standby safety systems. To keep standby safety systems at a high level of availability once on demand, most of them are tested and maintained regularly. These test and maintenance activities sometimes are cost-benefit issues. For stakeholders of a NPP, two maintenance strategies are typically desired: 1) for a given budget, the safety systems’ unavailability (i.e., risk) is minimal; or 2) under the premise of meeting a risk threshold, the maintenance cost is minimal. Yet these maintenance strategies are often involved in optimization issues with nonlinear constraints and mixed integers. To solve these issues, in this contribution, an integrated optimization framework based on a constraint particle swarm evolutionary methodology is developed for optimal design of maintenance strategies for NPPs. The main contribution of this work lies in: 1) providing an integrated optimization framework for designing strategies of maintenance for NPPs standby safety systems; 2) handling the problem of mixed integers; 3) dealing with nonlinear constraints. The case study demonstrated that the proposed methodology is more effective than genetic algorithms and can be used as a candidate method for designing maintenance strategies for nuclear reactors standby safety systems in the future.