Unsupervised Probabilistic Anomaly Detection over Nominal Subsystem Events through a Hierarchical Variational Autoencoder

This work develops a versatile approach to discover anomalies in operational data for nominal (i.e., non-parametric) subsystem event signals using unsupervised Deep Learning tech-niques. Firstly, it builds a neural convolutional framework to extract both intrasubsystem and intersubsystem patterns. This is done by applying banks of voxel filters on the charted data. Secondly, it generalizes the learned embedded regu-larity of a Variational Autoencoder manifold by merging latent space-overlapping deviations with non-overlapping synthetic irregularities. Contingencies like novel data, model drift, etc., are therefore seamlessly managed by the proposed data-augmented approach. Finally, it creates a smooth diagnosis probabilistic function on the ensuing low-dimensional distributed representation. The resulting enhanced solution warrants analytically strong tools for a critical industrial en-vironment. It also facilitates its hierarchical integrability, and provides visually interpretable insights of the degraded condition hazard to increase the confidence in its predictions. This strategy has been validated with eight pairwise-interrelated subsystems from high-speed trains. Its outcome also leads to further reliable explainability from a causal perspective.