Fault detection and diagnosis of marine diesel engines: A systematic review

• Published in: Ocean engineering Vol. 294; p. 116798
• Main Authors: Lv, Yaqiong, Yang, Xueting, Li, Yifan, Liu, Jialun, Li, Shijie
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15.02.2024
• Subjects: Data-driven method; Fault detection and diagnosis (FDD); Hybrid method; Knowledge-based method; Marine diesel engine; Model-based method; Marine diesel engine; Hybrid method; Model-based method; Fault detection and diagnosis (FDD); Knowledge-based method; Data-driven method
• ISSN: 0029-8018
• DOI: 10.1016/j.oceaneng.2024.116798

Marine diesel engines play a pivotal role in ensuring the smooth operation of maritime vessels. However, given the rigorous operational conditions and the natural wear and tear of engine components, the likelihood of diesel engine failures during voyages is an inherent concern. Preserving the optimal functionality and performance of marine diesel engines is of paramount importance, as it averts the potential repercussions of in-service failures and accidents. Hence, timely and effective fault detection is imperative to minimize operational disruptions and uphold safety standards. Intelligent fault detection and diagnosis (FDD) holds substantial promise in both academic and industrial contexts. While the monitoring of marine diesel engines using various FDD techniques has been a subject of study for decades, various new techniques emerges and there is a notable absence of comprehensive analysis and summarization of the methodologies developed. This paper commences by introducing the operational principles of marine diesel engines and elucidating potential faults. Subsequently, it provides an overview of recent advancements in FDD methods tailored for marine diesel engines, categorizing them into four distinct sections. In alignment with the conventional FDD workflow, the paper delves into discussions concerning model-based, data-driven, knowledge-based, and hybrid approaches. Moreover, it offers insights into the potential future directions of this field. •This paper meticulously categorizes Fault Detection and Diagnosis (FDD) methods for marine diesel engines into model-based, data-driven, knowledge-based, and hybrid approaches, providing a clear framework of each category.•Each FDD method category is thoroughly analyzed, with detailed workflows that guide the reader from data collection and model construction through to validation and optimization.•Future research directions are highlighted, underscoring the importance of integrating sophisticated data processing techniques to enhance the accuracy and reliability of FDD systems in marine applications.•The paper serves as an authoritative reference, collating a broad spectrum of studies and offering an in-depth evaluation of the strengths and limitations of various FDD methodologies in the context of the maritime industry.