Abstract
The application of machine learning within Structural Health Monitoring (SHM) has been widely successful in a variety of applications. However, most techniques are built upon the assumption that both training and test data were drawn from the same underlying distribution. This fact means that unless test data were obtained from the same system in the same operating conditions, the machine learning inferences from the training data will not provide accurate predictions when applied to the test data. Therefore, to train a robust predictor conventionally, new training data and labels must be recollected for every new structure considered, which is significantly expensive and often impossible in an SHM context. Transfer learning, in the form of domain adaptation, offers a novel solution to these problems by providing a method for mapping feature and label distributions for different structures, labelled source and unlabelled target structures, onto the same space. As a result, classifiers trained on a labelled structure in the source domain will generalise to a different unlabelled target structure. Furthermore, a holistic discussion of contexts in which domain adaptation is applicable are discussed, specifically for population-based SHM. Three domain adaptation techniques are demonstrated on four case studies providing new frameworks for approaching the problem of SHM.
Highlights
Data-driven approaches to Structural Health Monitoring (SHM), those utilising machine learning techniques, have achieved significant successes in a variety of applications [1,2,3,4]
These studies do not provide a holistic discussion of contexts in which domain adaptation is applicable for SHM, neither do they outline the specific domain adaptation methods utilised in the paper, namely Transfer Component Analysis (TCA), Joint Domain Adaption (JDA) and Adaptation Regularization based Transfer Learning (ARTL)
Domain adaptation has been demonstrated to be applicable in a variety of population-based SHM scenarios
Summary
Domain adaptation is a particular branch of transfer learning, where the focus is in reducing the distance between differing data distributions from source and target domains This approach to transfer learning is applicable to the SHM scenarios outlined above, namely when there is labelled data for a particular structure or operational environment and a set of unlabelled data for a different structure or operational environment. The algorithms assessed in this paper were transductive component analysis, geodesic flow kernel and maximum independence domain adaptation in combination with a Support Vector Machine (SVM), where the first two were found to provide better classification accuracies These studies do not provide a holistic discussion of contexts in which domain adaptation is applicable for SHM, neither do they outline the specific domain adaptation methods utilised in the paper, namely Transfer Component Analysis (TCA), Joint Domain Adaption (JDA) and Adaptation Regularization based Transfer Learning (ARTL).
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have