Towards Deep Industrial Transfer Learning for Anomaly Detection on Time Series Data

TL;DR

This paper introduces a modular deep learning algorithm designed for anomaly detection in time series data, leveraging transfer learning to adapt to changing industrial environments and improve detection performance.

Contribution

It presents a novel modular deep learning framework that facilitates transfer learning specifically tailored for industrial anomaly detection in time series data.

Findings

Effective transfer learning in industrial anomaly detection demonstrated

Modular architecture enables easy integration of transfer learning capabilities

Proven on real-world manufacturing dataset

Abstract

Deep learning promises performant anomaly detection on time-variant datasets, but greatly suffers from low availability of suitable training datasets and frequently changing tasks. Deep transfer learning offers mitigation by letting algorithms built upon previous knowledge from different tasks or locations. In this article, a modular deep learning algorithm for anomaly detection on time series datasets is presented that allows for an easy integration of such transfer learning capabilities. It is thoroughly tested on a dataset from a discrete manufacturing process in order to prove its fundamental adequacy towards deep industrial transfer learning - the transfer of knowledge in industrial applications' special environment.