Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Unraveling Anomalies in Time: Unsupervised Discovery and Isolation of Anomalous Behavior in Bio-regenerative Life Support System Telemetry

Ferdinand Rewicki, Jakob Gawlikowski, Julia Niebling, Joachim Denzler

Abstract

The detection of abnormal or critical system states is essential in condition monitoring. While much attention is given to promptly identifying anomalies, a retrospective analysis of these anomalies can significantly enhance our comprehension of the underlying causes of observed undesired behavior. This aspect becomes particularly critical when the monitored system is deployed in a vital environment. In this study, we delve into anomalies within the domain of Bio-Regenerative Life Support Systems (BLSS) for space exploration and analyze anomalies found in telemetry data stemming from the EDEN ISS space greenhouse in Antarctica. We employ time series clustering on anomaly detection results to categorize various types of anomalies in both uni- and multivariate settings. We then assess the effectiveness of these methods in identifying systematic anomalous behavior. Additionally, we illustrate that the anomaly detection methods MDI and DAMP produce complementary results, as previously indicated by research.

Unraveling Anomalies in Time: Unsupervised Discovery and Isolation of Anomalous Behavior in Bio-regenerative Life Support System Telemetry

Abstract

The detection of abnormal or critical system states is essential in condition monitoring. While much attention is given to promptly identifying anomalies, a retrospective analysis of these anomalies can significantly enhance our comprehension of the underlying causes of observed undesired behavior. This aspect becomes particularly critical when the monitored system is deployed in a vital environment. In this study, we delve into anomalies within the domain of Bio-Regenerative Life Support Systems (BLSS) for space exploration and analyze anomalies found in telemetry data stemming from the EDEN ISS space greenhouse in Antarctica. We employ time series clustering on anomaly detection results to categorize various types of anomalies in both uni- and multivariate settings. We then assess the effectiveness of these methods in identifying systematic anomalous behavior. Additionally, we illustrate that the anomaly detection methods MDI and DAMP produce complementary results, as previously indicated by research.
Paper Structure (25 sections, 13 equations, 23 figures, 3 tables)

This paper contains 25 sections, 13 equations, 23 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Methodology
  4. Anomaly Detection
  5. Feature Extraction
  6. Time Series Clustering
  7. Quality Measures
  8. Experimental Results
  9. Dataset
  10. RQ1: Are the results of MDI and DAMP complementary?
  11. RQ2: Which features yield the best results?
  12. Univariate anomalies
  13. Multivariate Anomalies
  14. RQ3: Which algorithm yields better results?
  15. Univariate Anomalies
  16. ...and 10 more sections

Figures (23)

  • Figure 1: Overview of our approach to derive different types of anomalous behavior from unlabelled time series. 1 MDI and DAMP are applied to time series data to obtain anomalous sequences. 2 Features are extracted from sequences and 3 clustering is applied to derive different anomaly types, marked by color.
  • Figure 2: Amount of (a) univariate and (b) multivariate anomalies found by MDI and DAMP. The highlighted area marks anomalies found by both algorithms.
  • Figure 3: (a), (c) $SAAI$ and $SSC$ aggregated over all sensor types for the four different feature extraction methods for (a) K-Means and (c) HAC for univariate anomalies. (b), (d): $SSC$ over all subsystems in the multivariate case.
  • Figure 4: Gini-Index of cluster sizes for K-Means Clustering and HAC in (a) the univariate and (b) the multivariate case. Lines represent mean values, while the opaque bands span the second and third quartiles.
  • Figure 5: Anomaly types found in ICS temperatures with K-Means clustering for $SAAI$-optimal $K$ per feature set. Anomaly types consistent across different feature sets have the same color. T/B: DenoisedCrafted, Rocket, Catch22
  • ...and 18 more figures

Theorems & Definitions (2)

  • Definition 1
  • Definition 2