An on-line combined linear–nonlinear fatigue crack detection technique
暂无分享,去创建一个
Abstract To monitor the structural health during fatigue tests, classical nondestructive tests (ultrasonic inspection, liquid penetration, eddy current, etc.) are usually performed at regular time instances. Unfortunately, the fatigue tests should be interrupted to use these techniques. In addition, a large amount of user interaction is required. In this article, vibration features are used to detect cracks on-line with the execution of a fatigue test. To perform this task, an experimental strategy is developed to simultaneously estimate static and dynamic as well as linear and nonlinear vibration features. By means of these features the sensitivity of static versus dynamic and linear versus nonlinear damage detection techniques will be qualified. Finally, it will be shown that by using nonlinear identification techniques, additional information on the damage scenario can be extracted. The validation will be done on a steel beam with a propagating fatigue crack.
[1] David E. Goldberg,et al. Genetic Algorithms in Search Optimization and Machine Learning , 1988 .
[2] Yves Rolain,et al. SIMPLE METHODS AND INSIGHTS TO DEAL WITH NON-LINEAR DISTORTIONS IN FRF-MEASUREMENTS , 1998 .
[3] P. Verboven,et al. On-line detection of fatigue cracks using an automatic mode tracking technique , 2003 .