Development of a residual-based anomaly detection system with persistence logic for marine diesel generators

Abstract

Diesel generator engines (DGEs) are critical safety and mission assets for naval platforms, supporting continuous electrical power for navigation, communications, habitability, and training operations. In practice, maintenance of auxiliary generation on training ships still relies predominantly on time-based tasks complemented by reactive corrective actions, despite the increasing availability of onboard operational data. This paper presents the development of a residual-based anomaly detection system with persistence logic for the diesel generator engines of a Colombian Navy training ship, focusing on the auxiliary generator sets as a case study. The proposed approach targets early detection of abnormal thermal behavior under scarce fault labels by combining (i) data-quality gates and traceable preprocessing, including plausibility filtering and multivariate inconsistency treatment during cleaning, (ii) target and feature definition for normal-behavior regression, (iii) residual-based monitoring with EWMA smoothing and time-varying control limits, and (iv) persistence rules for sustained event declaration. The methodology is organized as an end-to-end workflow aligned with CRISP-DM and a PHM detection-first strategy. Because historical fault labels are limited and not reliably aligned in time, offline evaluation combines predictive assessment on held-out healthy data with Monte Carlo validation under simulated fault scenarios. Results on historical generator monitoring data from the ship show that the normal-behavior models provide stable residual baselines for key thermal variables, while the detector can identify simulated abnormal scenarios across different severity levels. This paper provides a traceable workflow for anomaly detection in sparse and irregular shipboard monitoring data, discusses the limitations imposed by manual logs and scarce labels, and outlines future work toward health indexing, operational feedback, and more robust diagnostic support