Identification of anisotropy and asymmetry in flexible rotor systems using directional frequency response functions

Abstract

Recently, the significance of timely diagnosis of rotating machinery is growing because it can guarantee the reliability and safety of operation, and cut maintenance costs. A diagnosis method by assembling components is prone to cause another damages during the inspection process due to complicated structures of the machines. Moreover, the economical loss due to stopping the machine is inevitable. Thus, among other diagnosis methods, a vibration-based method, especially modal testing, has been considered as a reliable method for many years because it can perform diagnosis of machines without dismantling them. However, an effective method to extract physical parameters that show both degree and location of damage from modal testing results has not been researched yet. Moreover, with the interest in reducing costs, there has been growing demands to develop advanced diagnosis technique based on vibration that can identify damage information with a limited number of sensors. Therefore, the primary objective of this thesis is to develop effective methods that can identify damage and faults from modal testing of rotating machinery with a minimum number of sensors and exciters. The approach to identify damage information, this thesis employs the directional frequency response functions (dFRFs) in the complex coordinates, which can evaluate anisotropy and asymmetry of a rotating machine. The anisotropy and asymmetry indicate non-axisymmetry of stationary and rotating parts of rotating machines, respectively. Thus, they are directly linked to damages of stationary parts such as bearings and rotating shafts. As such, the dFRFs can be used as an excellent indicator to represent the damages of rotating machines. When a rotor system possesses both anisotropy and asymmetry, conventional transformation cannot eliminate periodically time-varying coefficients. Thus, it is difficult to perform conventional modal analysis and estimate the dFRFs accurately. Hence, this stu...