Applications of fault detection and diagnosis methods in nuclear power plants: A review

Abstract

Nuclear power industries have increasing interest in using fault detection and diagnosis (FDD) methods to improve safety, reliability, and availability of nuclear power plants (NPP). A brief overview of FDD methods is presented in this paper. FDD methods are classified into model-based methods, data-driven methods, and signal-based methods. While practical applications of model-based methods are very limited, various data-driven methods and signal-based methods have been applied for monitoring key subsystems in NPPs. In this paper, six areas of such applications are considered. They are: instrument calibration monitoring, instrumentation channel dynamic performance monitoring, equipment monitoring, reactor core monitoring, loose part monitoring, and transient identification. The principles of using FDD methods in these applications are explained and recent studies of advanced FDD methods are examined. Popularity of FDD applications in NPPs will continuously increase as FDD theories advance and the safety and reliability requirement for NPP tightens