Foundations of population-based SHM, Part IV: The geometry of spaces of structures and their feature spaces. (August 2021)

• Record Type: Journal Article
• Authors: Tsialiamanis, G.; Mylonas, C.; Chatzi, E.; Dervilis, N.; Wagg, D.J.; Worden, K.
• Is Part Of: Mechanical systems and signal processing. Volume 157(2021)
• Publication Date: 2021-08
• Subjects: Population-Based Structural Health Monitoring (PBSHM) -- Differentiable manifolds -- Fibre bundles -- Confounding influences -- Graph Neural Networks (GNNs); Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2021.107692
• Languages: English
• ISSNs: 0888-3270

Abstract

Highlights: Population-based SHM allows knowledge transfer between structures of a population. Collecting manifolds of potential structural states, a fibre bundle is defined. The normal-condition cross section is of particular interest in the SHM discipline. A graph representation of structures in combination with graph neural networks is used to construct the section. Method is demonstrated successfully on a population of simulated structures. Abstract: One of the requirements of the population-based approach to Structural Health Monitoring (SHM) proposed in the earlier papers in this sequence, is that structures be represented by points in an abstract space. Furthermore, these spaces should be metric spaces in a loose sense; i.e. there should be some measure of distance applicable to pairs of points; similar structures should then be 'close' in the metric. However, this geometrical construction is not enough for the framing of problems in data-based SHM, as it leaves undefined the notion of feature spaces. Interpreting the feature values on a structure-by-structure basis as a type of field over the space of structures, it seems sensible to borrow an idea from modern theoretical physics, and define feature assignments as sections in a vector bundle over the structure space. With this idea in place, one can interpret the effect of environmental and operational variations as gauge degrees of freedom, as in modern gauge field theories. One can then regard data normalisation procedures like cointegration as gauge-fixing operations. This paper will discuss the various geometrical structures required for an abstract theory of feature spaces in SHM, and will draw analogies with how these structures have shown their power in modern physics. Having motivated a number of problems in Population-Based SHM (PBSHM) in geometrical terms, it remains to show how these problems might be solved. In the second part of the paper, the problem of determining the normal condition cross section of a feature bundle is addressed. The solution is provided by the application of Graph Neural Networks (GNN), a versatile non-Euclidean machine learning algorithm which is not restricted to inputs and outputs from vector spaces. In particular, the algorithm is well suited to operating directly on the sort of graph structures which are an important part of the proposed framework for PBSHM. The solution of the normal section problem is demonstrated for a heterogeneous population of truss structures for which the feature of interest is the first natural frequency. The GNN approach is shown to not only solve the normal section problem, but also to accommodate varying temperatures across the population; it thus provides a means of data normalisation.