Research overview and prospect in condition monitoring of compressors

• Published in: Expert systems with applications Vol. 277; p. 127284
• Main Author: Kumar, Anil
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 05.06.2025
• Subjects: Artificial Intelligence in Maintenance; Compressor; Digital Twins; Predictive Maintenance; Compressor; Artificial Intelligence in Maintenance; Predictive Maintenance; Digital Twins
• ISSN: 0957-4174
• DOI: 10.1016/j.eswa.2025.127284

Compressors are critical components in industries such as manufacturing, energy production, and automotive systems. Compressors operate in high-pressure environments. Any failures, especially in high-pressure components, can lead to serious safety risks. Therefore, effective condition monitoring is essential for early fault detection and risk reduction. Despite the importance of air compressors, research in this area is often undervalued and needs more attention. This has motivated the author to advocate for a deeper exploration of the subject, highlight research overview and prospects in condition monitoring of air compressors. This paper provides an overview of key methods such as vibration analysis, acoustic emission, thermal imaging, oil analysis, and pressure and flow monitoring, each of which offers unique advantages in detecting specific fault types. Additionally, the paper explores advancements in feature extraction, dynamic modeling, and the application of both conventional and next-generation artificial intelligence techniques for predictive maintenance. The paper also discusses several industrial case studies. This paper concludes by discussing the ongoing challenges and opportunities in the domain, offering a roadmap for future research and technological advancements in air compressor condition monitoring. As one of the first review of its kind, this work aims to fill a gap in the literature on the topic, which remains an important yet often overlooked area of study in compressor condition monitoring. Through this review paper, we also propose a continuous learning and model adaptation framework for compressor condition monitoring to enable real-time adaptation and improvement of predictive maintenance model.