Development of a new frequency-domain method for fatigue damage assessment in bimodal Gaussian random processes

Abstract:

To accurately evaluate the fatigue damage of bimodal Gaussian random process, a new frequency-domain analysis method called as the improved bands method was proposed. The basic idea of the proposed method was developed from the spectral decomposition approach. Firstly, the power spectrum density functions of low-frequency and high-frequency modes were split into a large number of infinitesimal frequency bands. According to the damage equivalence principle, the equivalent conversions were separately performed for the low-frequency and high-frequency components to obtain the corresponding equivalent narrow-band processes. Subsequently, another equivalent conversion from high-frequency equivalent narrow-band process to low-frequency mode was conducted. In order to account for the interaction between the low-frequency and high-frequency modes, a correction factor, which depends on characteristic frequency ratio, energy ratio and material parameter m of the S-N curves, was introduced to modify the summed zero-order spectral moment. When the final equivalent narrow-band process was obtained, the total damage of bimodal Gaussian process could be calculated by the analytical solution of narrow-band fatigue damage. Through numerical tests of ideal rectangular bimodal spectrum and real bimodal spectrum, by taking the fatigue damage estimated by time-domain rain-flow counting method as reference, the accuracy and robustness of the present method was validated against several popular frequency-domain methods. Results showed that the improved bands method is not only more accurate than other traditional frequency-domain methods but also easier to be implemented in computer programming, which also indicates its great potential in actual engineering practice.