It is crucial to detect, characterize and model events of interest in a new propulsion system. In this paper, a novel framework is established within which an adaptive clustering approach that uses fuzzy techniques and hierarchical methodologies is proposed to characterize specific events of interest using the measured and processed features. Raw engine measurement data is first analyzed using the wavelet transform to provide localization of frequency information before they are fed to the classification system. A method combining hierarchical clustering and fuzzy k-mean clustering in an adaptive learning manner is then applied to find the events of interest during the operation of engine, thus addressing two major issues associated with conventional partitional clustering: sensitivity to initialization and difficulty in determining the optimal number of clusters. Experimental results show that the proposed approach is computationally feasible and effective in learning, and detecting and classifying events of interest.