Bladed disk crack detection through advanced analysis of blade time of arrival signal

Abstract

Health condition monitoring and fault diagnostics of turbo fan engines play significant roles in overall cost reduction and reliability enhancement of the aircraft system. Among various types of potential faults in a turbo fan engine, crack initiation and propagation in the bladed disks of engines caused by high-cycle fatigue under cyclic loads are typical ones that could result in the breakdown of the engines if not detected at an early stage. Reliable fault detection techniques are therefore required to detect impending engine malfunctions as well as unexpected failures that could otherwise lead to costly and/or catastrophic consequences. Although a number of approaches have been reported in literature, it still remains very challenging to develop a reliable technique to accurately estimate the health condition of bladed disks of engines. As such, this paper presents a new technique for engine bladed disk crack detection through advanced analysis of blade time-of-arrival signal. Two stages of signal processing are involved in this technique: 1) signal preprocessing for removing the noise caused by rotor imbalance; and 2) signal post-processing for identifying the location of the crack. The effectiveness of the developed technique is validated experimentally in a spin rig test.