Towards semi-supervised and probabilistic classification in structural health monitoring

Abstract

In practical applications of data-driven Structural Health Monitoring (SHM), recording labels for each of the measured signals can be infeasible and expensive. In consequence, conventional methods for (supervised) machine learning can become irrelevant in certain applications of damage classification. Semisupervised methods, however, allow algorithms to learn from information in the available unlabelled measurements as well a limited set of labelled data. As such, this paper suggests a semi-supervised Gaussian mixture model for probabilistic damage-classification, informed by both labelled and unlabelled signals. The generative statistical model is shown to improve the classification performance, compared to supervised learning, with simulated and experimental SHM data, while requiring no further inspections of the system. Specifically, semi-supervised learning leads to 3.87% and 3.83% reductions in the classification error for the simulated and experimental datasets respectively. These results indicate that, through semi-supervised learning in SHM, the cost associated with labelling data could be managed, as the information in a small set of labelled signals can be combined with larger sets of unlabelled data.

Probabilistic Inference for Structural Health Monitoring: New Modes of Learning from Data

“…Missing label information is especially relevant to practical applications of SHM: while fully labelled data are often infeasible, it can be possible to include labels for a limited set (or budget) of measurements. Typically, the budget is limited by some expense incurred when investigating the signals; this might include direct costs associated with inspection, or loss of income due to down-time (Bull et al 2020b).…”

“…A generative classifier is used to demonstrate probabilistic active learning. In this example -originally shown in (Bull et al 2020b) -a Gaussian mixture model (GMM) is used to monitor streaming data from a motorway bridge, as if the signals were recorded online. The model defines a multi-class classifier, to aid both damage detection and identification, while limiting the number of (costly) system inspections.…”

A Brief Introduction to Recent Developments in Population-Based Structural Health Monitoring

“…Firstly, it is important to confront the "myth" that one cannot utilize machine learning for SHM problems like localization, or classification of damage unless one always has millions of labeled examples for supervised learning. To dispel this idea, the reader is referred to Bull et al (2019) and Bull et al (2020c), where techniques like active and semi-supervised learning are demonstrated on SHM problems. This is an important point, because one of the major problems for SHM is the scarcity of labeled data.…”

Application of Bayesian Fe Model Selection for SHM of a Monumental Masonry Palace Through Semi-Supervised Learning

“…Indeed, only a few recent contributions deal with long-term SHM data and damage detection so far [4,3,23,2,12]. In this context, SHM-based technologies are encompassing machine learning (ML) techniques [18,17,24], devoting the attention to the semi-supervised learning paradigm, which allows the use of large amount of monitoring data with and without descriptive labels [5,6]. The process of prioritizing the acquired data can be accomplished by using numerical models, i.e.…”