Bump extraction algorithm for variable amplitude fatigue loading

Abstract

This paper presents the development of a fatigue mission synthesis algorithm, called Wavelet Bump Extraction (WBE), for summarising long records of fatigue road load data. This algorithm is used to extract fatigue damaging events or bumps in the record that cause the majority of the fatigue damage, whilst preserving the load cycle sequences. Bumps are identified from characteristic frequency bands in the load spectrum using the 12th order Daubechies wavelet. The bumps are combined to produce a mission signal which has equivalent signal statistics and fatigue damage to the original signal. The WBE accuracy has been evaluated by observing the cycle sequence effects of the bump loadings. The WBE was compared with the time domain fatigue data editing method, so that the effectiveness of WBE can be verified. Using WBE, a substantial compression of the load-time history could be achieved for the purpose of accelerated fatigue tests in the automotive industry.