Graph Structural Residuals: A Learning Approach to Diagnosis
Abstract
Traditional model-based diagnosis relies on constructing explicit system models, a process that can be laborious and expertise-demanding. In this paper, we propose a novel framework that combines concepts of model-based diagnosis with deep graph structure learning. This data-driven approach leverages data to learn the system's underlying structure and provide dynamic observations, represented by two distinct graph adjacency matrices. Our work facilitates a seamless integration of graph structure learning with model-based diagnosis by making three main contributions: (i) redefining the constructs of system representation, observations, and faults (ii) introducing two distinct versions of a self-supervised graph structure learning model architecture and (iii) demonstrating the potential of our data-driven diagnostic method through experiments on a system of coupled oscillators.
Cite
@article{arxiv.2308.06961,
title = {Graph Structural Residuals: A Learning Approach to Diagnosis},
author = {Jan Lukas Augustin and Oliver Niggemann},
journal= {arXiv preprint arXiv:2308.06961},
year = {2023}
}
Comments
11 pages, added missing section heading