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

TL;DR

This paper explores probabilistic inference methods for structural health monitoring, emphasizing their robustness and adaptability to noisy, incomplete data, with applications demonstrated through three case studies involving semi-supervised, active, and multi-task learning.

Contribution

It introduces novel probabilistic learning techniques tailored for SHM, addressing challenges of data noise, incompleteness, and lack of labels, with practical case studies.

Findings

Probabilistic methods improve robustness in SHM data analysis.

Semi-supervised, active, and multi-task learning enhance SHM performance.

Case studies demonstrate practical effectiveness of proposed techniques.

Abstract

In data-driven SHM, the signals recorded from systems in operation can be noisy and incomplete. Data corresponding to each of the operational, environmental, and damage states are rarely available a priori; furthermore, labelling to describe the measurements is often unavailable. In consequence, the algorithms used to implement SHM should be robust and adaptive, while accommodating for missing information in the training-data -- such that new information can be included if it becomes available. By reviewing novel techniques for statistical learning (introduced in previous work), it is argued that probabilistic algorithms offer a natural solution to the modelling of SHM data in practice. In three case-studies, probabilistic methods are adapted for applications to SHM signals -- including semi-supervised learning, active learning, and multi-task learning.