作者: Khalil Moshkbar-Bakhshayesh , Mohammad B. Ghofrani
DOI: 10.1016/J.PNUCENE.2013.03.017
关键词: Genetic algorithm 、 Identification (information) 、 Particle swarm optimization 、 Expert system 、 Computer science 、 Artificial intelligence 、 Hidden Markov model 、 Machine learning 、 Soft computing 、 Transient (computer programming) 、 Fuzzy logic
摘要: Abstract A transient is defined as an event when a plant proceeds from normal state to abnormal state. In nuclear power plants (NPPs), recognizing the types of transients during early stages, for taking appropriate actions, critical. Furthermore, classification novel “don't know”, if it not included within NPPs collected knowledge, necessary. To fulfill these requirements, identification techniques method recognize and classify conditions are extensively used. The studies revealed that model-based methods suitable candidates in NPPs. Hitherto, data-driven methods, especially artificial neural networks (ANN), other soft computing such fuzzy logic, genetic algorithm (GA), particle swarm optimization (PSO), quantum evolutionary (QEA), expert systems mostly investigated. hidden Markov model (HMM), support vector machines (SVM) considered By modern techniques, safety, due accidents recognition by symptoms rather than events, improved. Transient expected become increasingly important next generation reactors being designed operate extended fuel cycles with less operators' oversight. this paper, recent related advanced presented their differences illustrated.