On the Use of Model-Based Versus Data-Based Approaches for Virtual Sensing in SHM

Abstract

Structural Health Monitoring (SHM) typically aims at producing methodologies for the periodic, and often online, assessment of structures. While there are different approaches to SHM, there is one common necessary element if applied in practice - data acquired from sensors. In an ideal scenario, sensors would be deployed anywhere on a structure of interest and could be added retrospectively, if not included in the original design. In practice, there are various limitations to the availability and installation of sensors, such as physical access and cost. Virtual sensing has thus been proposed as a solution to the problem of sensor availability and different approaches have been applied in various fields. This paper applies and compares two approaches for virtual sensing in structural dynamics: modal expansion which is model-based and Gaussian Process (GP) regression which is data-based. The approaches are demonstrated on data from a Piper PA-28 tailplane structure, which was tested with the help of a scanning-laser vibrometer under laboratory conditions.