Random vibration fatigue behavior of directionally solidified superalloy: Experiments and evaluation of life prediction methods

Abstract

Vibration is closely related to structural dynamics, and the study of random vibration fatigue loads is typically conducted in the frequency domain. In this study, various frequency domain methods were employed to predict the random vibration fatigue life of DZ125L directionally solidified superalloy under different vibration signal intensities. Experimental analysis was used to examine the fatigue life and random vibration response of DZ125L alloy specimens under different vibration signal intensities. The S-N curve derived from sinusoidal vibration tests was used to replace the traditional S-N curve obtained from tension-compression tests in random vibration fatigue analysis, leading to improved life prediction accuracy. Additionally, an intensity function fξ was established using the spectral width parameters α1 and α2 of the random process and the vibration signal intensity ξ, which was used to modify the traditional Rayleigh method. A new vibration fatigue life prediction model based on frequency domain method is obtained. The results indicate that the intensity function fξ emphasizes the elastic or plastic stage of the specimen during the random process, leading to highly accurate life prediction values using the Modified Rayleigh method across different vibration signal intensities, with all prediction points are within 1.5 times the scatter band.