Abstract
A key challenge in Structural Health Monitoring (SHM) is the lack of availability of data from a full range of changing operational and damage conditions, with which to train an identification/classification algorithm. This paper presents a framework based on Bayesian non-parametric clustering, in particular Dirichlet Process (DP) mixture models, for performing SHM tasks in a semi-supervised manner, including an online feature extraction method. Previously, methods applied for SHM of structures in operation, such as bridges, have required at least a year’s worth of data before any inferences on performance or structural condition can be made. The method introduced here avoids the need for training data to be collected before inference can begin and increases in robustness as more data are added online. The method is demonstrated on two datasets; one from a laboratory test, the other from a full scale test on civil infrastructure. Results show very good classification accuracy and the ability to incorporate information online (e.g. regarding environmental changes).
Highlights
Structural Health Monitoring (SHM) [1,2] is an important area of research within engineering, seeking to detect and diagnose degradation in structures and systems before it can impede use or become a hazard
The application of the Dirichlet Process (DP) mixture model is explored here using a benchmark dataset from a three-storey building structure (Fig. 3), produced by Los Alamos National Laboratory [53] for identification of damage under changing system behaviour
The work presented in this paper introduces a methodology for incorporating a DP mixture model into an SHM system for online damage detection
Summary
Structural Health Monitoring (SHM) [1,2] is an important area of research within engineering, seeking to detect and diagnose degradation in structures and systems before it can impede use or become a hazard. Statistical models can be used to detect similarity (or difference) between sets of data collected from a structure, which is, in turn, used to infer its health/condition. It will not be possible to acquire data covering all healthy conditions and damage scenarios, the main limitation being the cost of producing and subsequently damaging large valuable structures, e.g. within the aerospace industry or civil infrastructure. The existence of a number of different damage scenarios comes from the multiple mechanisms for damage that a structure might experience. In certain cases it will be unsafe to operate the structure with a given type of damage present, meaning that collection of data from this damage state prior to operation of the structure is not possible. A structure will operate in a number of different operational and environmental
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
