Towards Population-Based Structural Health Monitoring, Part I: Homogeneous Populations and Forms

Abstract

Data-driven models in Structural Health Monitoring (SHM) generally require comprehensive datasets, recorded from systems in operation, which are rarely available. One potential solution to this problem, considers that information might be transferred, in some sense, between similar systems. As a result, a population-based approach to SHM suggests methods to both model and transfer this valuable information, by considering different groups of structures as populations. Specifically, in this work, a method is proposed to model a population of nominally-identical systems, where (complete) datasets are only available from a subset of members. The framework attempts to build a general model, referred to as the population form, which can be used to make predictions across a group of homogeneous systems. First, the form is demonstrated through applications to a simulated population -with a single experimental (test-rig) member; secondly, the form is applied to data recorded from a group of operational wind turbines.

Towards Population-Based Structural Health Monitoring, Part II: Heterogeneous Populations and Structures as Graphs

“…If the population is described as strongly homogeneous, this also implies that the boundary conditions for the two structures are the same, for example all turbines in a wind farm with foundations on the same part of the seabed. A population where the IE representations are at least nominally-identical is a homogeneous population, and the idea of a form can be used in this case [4], in addition to other transfer learning methods.…”

“…This situation is typically the case for wind farms, and so far there has been work taking the first steps towards PBSHM in the Lillgrund wind farm [3]. There are many ways of performing PBSHM for a homogeneous population with one approach being to represent a homogeneous population with a single model called a form [4], and another approach being the use of transfer learning.…”

Towards Population-Based Structural Health Monitoring, Part III: Graphs, Networks and Communities

“…By examining a population of structures, rather than simply looking at each structure individually, it may be possible to share information about the normal condition and damage states, provided certain conditions are met. An example of this is the Lillgrund wind farm [4], where the nominally identical structures form a homogeneous population [5]. In the Lillgrund wind farm, it was possible to make the detection of performance anomalies robust to variations in the normal condition by sharing SCADA data between wind turbines.…”

“…In the Lillgrund wind farm, it was possible to make the detection of performance anomalies robust to variations in the normal condition by sharing SCADA data between wind turbines. In homogeneous populations, it is also possible to describe the population using a single model, called a form, which effectively captures the variation within the population [5].…”

An Ontological Approach to Structural Health Monitoring

“…Although applicable to SHM in a broad sense, an ontology could be particularly useful in a Population-based SHM (PBSHM) setting [4,5,6,7,8], where the goal is to develop general inference tools across a population. Here an ontology would identify useful, and perhaps overlooked, connections between objects such as Irreducible Element (IE) models and Atribbuted Graph (AG) representations of structures [5,6] and appropriate knowledge transfer methods like transfer learning [7] and 'forms' [7,4]. This may provide benefits in highlighting appropriate methods for each data source such that destructive phenomenon like negative transfer in transfer learning are avoided.…”