Abstract: For safe and economical operation of nuclear power plants (NPPs), the occurring anomalies should be promptly and correctly identified. In this paper, the transients were identified by processing time series of critical variables. First, based on its ability to measure the complexity of time series, the fuzzy entropy (FuzzyEn) was used to determine whether the system was in normal state. Then cross fuzzy entropy was employed for classifying the occurring transients, using its ability to characterize the similarity between two time series. The feasibility and effectiveness were verified by simulator data of pebble-bed modular high temperature gas-cooled reactor nuclear power plant (HTR-PM). It is demonstrated that the proposed method is effective for transient identification and it dispenses with a complex training phase.