Population-based structural health monitoring

One of the dominant challenges in data-based structural health monitoring (SHM) is the scarcity of measured data corresponding to different damage states of the structures of interest. A new arsenal of advanced technologies is described here that can be used to solve this problem. This new generation of methods is able to transfer health inferences and information between structures in a population-based environment—population-based SHM (PBSHM). In the category of homogenous populations (sets of nominally identical structures), the idea of a Form can be utilised, as it encodes information about the ideal or typical structure, together with information about variations across the population. In the case of sets of different structures and thus heterogeneous populations, technologies of transfer learning are described as a powerful tool for sharing inferences (technologies that are also applicable in the homogeneous case). In order to avoid negative transfer and assess the likelihood of a meaningful inference, an abstract representation framework for spaces of structures will be analysed as it can capture similarities between structures via the framework of graph theory. This chapter presents and discusses all of these very recent developments and provides illustrative examples.