科研之友
微信
新浪微博
Facebook
分享链接
摘要
The main problem associated with pattern recognition based approaches to Structural Health Monitoring (SHM) is that damage localisation and quantification almost always require supervised learning. In the case of high-value engineering structures like aircraft, it is simply not possible to generate the training data associated with damage by experiment. It is also unlikely that data can always be generated by simulation as the models required would often need to be of such high fidelity that the costs of development and the run-times would again be prohibitive. The object of this paper is to explore the potential of a simple experimental strategy, which involves adding masses to the structure, in the attempt to extract features for novelty detection. The idea itself is not presented as revolutionary based on the fact that adding masses has been considered as a case of damage before, however, an in-depth investigation of its suitability for guiding feature selection is presented here. The approach is illustrated first on a simple structure by using data generated from a finite-element (FE) simulation and then validated experimentally on a more complicated laboratory structure. Simulated damage, in the form of a loss in the stiffness in the case of the numerical model and of a saw-cut in the case of the structure is used for comparison. The results show similar patterns in both cases which suggests a potential use of the method for higher level damage detection.
- 出版日期2010-6-7
