A hybrid frequency-time domain method for predicting multiaxial fatigue life of 7075-T6 aluminium alloy under random loading

2014 ◽  
Vol 38 (3) ◽  
pp. 247-256 ◽  
Author(s):  
J. Ge ◽  
Y. Sun ◽  
S. Zhou ◽  
L. Zhang ◽  
Y. Zhang ◽  
...  
Author(s):  
Yalin Liu ◽  
Shangguan Bo ◽  
Zili Xu

An improved hybrid frequency-time domain method is proposed to efficiently study the nonlinear response of blade systems subject to dry friction damping. In this method, the friction force is calculated in time domain by tracing the motion trajectory of friction contact pairs. The Newton iterative algorithm is used to solve the algebraic equations of blade system in frequency domain. The parabolic-interpolation fast Fourier transform is used in time-frequency domain transformation. It can improve computation accuracy and enlarge spectral analysis range of the original alternating frequency/time domain (AFT) method. The multi-modal analysis technique is employed to reduce linear degree of freedoms (DOFs), and only the nonlinear DOFs are brought in iterative procedure. It can significantly reduce computation time. The improved method combined with three dimensional friction contact model is used to predict the vibration response of a shrouded blade with nonlinear friction contact. The effect of parameters on resonant frequency and response is investigated and discussed. The computation time of the improved method is only 42.25 percent of the original AFT method, while the relative difference between the results of the two methods is no more than 0.01%. The improved method is shown to be superior to the original AFT method in computation efficiency and accuracy, and it can be used in analyzing dynamic characteristics of complex frictionally damped blade systems.


2013 ◽  
Vol 569-570 ◽  
pp. 900-907
Author(s):  
Matjaz Mrsnik ◽  
Janko Slavič ◽  
Miha Boltezar

It is a common practice in the automotive industry to expose products to accelerated vibration tests, that simulate the load, predicted to occur during the products service time. To avoid long testing times, higher amplitudes are used. Usually such tests come late in the development process, and can result in unexpected costs. A common tool for predicting time-to-failure or expected fatigue-life of the product is the time-domain method, using the rainflow counting algorithm and the Palmgren-Miner summation method. However, if one chooses to apply this method inside a FEM environment on a large amount of nodes with different time histories dependent on the structure excitation, the time-domain method becomes computationally complex. This has led to more effective methods, that estimate the time-to-failure in frequency-domain but are less accurate, compared to the time-domain approach. In this research, a group of such methods is presented and compared using real signals, namely: Tovo-Benasciutti, Wirsching-Light, Petrucci-Zuccarello, empirical α0.75, Dirlik and Gao-Moan method. Separately, only some of those methods were already compared side by side. Usually the comparison was made on simulated random signals, while this research compares them based on a real signal, collected by measuring different groups of spectra (e.g.typical vibration test profiles, different background noise levels, spectral width, number of modes etc.). In existing studies, Dirlik is usually identified as most accurate but in this research, conclusions show, that the Tovo-Benasciutti and Zhao-Baker methods can be more accurate than the Dirlik method and should therefore also be considered for vibration fatigue analysis.


2021 ◽  
Vol 75 ◽  
pp. 102869
Author(s):  
Zhenguo Gao ◽  
Mike Efthymiou ◽  
Liang Cheng ◽  
Tongming Zhou ◽  
Matthieu Minguez ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document