IWSHM 2017: Vibration-based damage localization and estimation via the stochastic functional model-based method: An overview

The functional model-based method is a unique vibration-based approach to damage localization and magnitude estimation in structures under normal operation, capable of operating on naturally arising random vibration signals without human intervention. It employs an innovative data-based stochastic functional model representing the partial dynamics under various damage scenarios, estimation techniques, and statistical hypothesis testing. Unlike alternative vibration methods employing data-based models but treat the localization and magnitude estimation approximately, as classification problems, it handles them properly as estimation problems. Among its remarkable properties is its ability to operate with a low number of vibration response sensors and low- and limited-frequency bandwidth, characteristics important in many applications. In this article, an overview of the method and its principles are presented, along with three validation case studies.