Litcius/Paper detail

Two-step residual-error based approach for anomaly detection in engineering systems using variational autoencoders

Ana González, Ignacio Díaz, Abel A. Cuadrado, Diego García-Pérez, Daniel Pérez

2022Computers & Electrical Engineering33 citationsDOIOpen Access PDF

Abstract

Anomaly detection is a crucial task in the engineering systems field. However, there is usually little or no information about all possible abnormal modes in systems. Hence, a common approach is to build a model of healthy behaviour, based on normal operation data, so that anomaly detection would depend on how well new data fit this model. According to this idea, we propose a residual-error based approach consisting of: a variational autoencoder, used to model the probability density function of the system’s healthy behaviour; and a two-step classification algorithm, which classifies the incoming samples based on their residuals, and reports not only their normal/anomalous nature but also that of their components. We have tested this proposal in three different engineering contexts and we have compared its performance with that of state-of-the-art approaches, demonstrating its capability to successfully detect and characterize anomalies.

Topics & Concepts

ResidualAutoencoderAnomaly detectionAnomaly (physics)Computer scienceArtificial intelligenceField (mathematics)Function (biology)Data miningPattern recognition (psychology)Probability density functionAlgorithmMathematicsArtificial neural networkStatisticsPhysicsEvolutionary biologyPure mathematicsCondensed matter physicsBiologyAnomaly Detection Techniques and ApplicationsFault Detection and Control SystemsBacillus and Francisella bacterial research
Two-step residual-error based approach for anomaly detection in engineering systems using variational autoencoders | Litcius