Trend prognosis of aero-engine abrupt failure based on affinity propagation clustering

Abstract

This paper presented a new method for trend prognosis of aero-engine abrupt failure which was based on clustering analysis. We used affinity propagation method to cluster vibration data from aero-engine rotating machinery to find the failure trend by means of features extracted from it. This system could remind rapidly which failure will happen next. It is possible to be applied to trend prognosis of aero-engine abrupt failure in airlines. © 2014 Springer-Verlag.