A Review on AI and Machine Learning for Predictive Maintenance and FDD in RAC Systems

Year : 2026 | Volume : 13 | Issue : 01 | Page : 15 25
    By

    S. Ramabalan,

  • S. Mahalakshmi,

  • V. Sathiya,

  • N. Godwin Raja Ebenezer,

  1. Professor, Department of Mechanical Engineering, Selvam College of Technology, Namakkal, Tamil Nadu, India
  2. Professor, Department of Information Technology, Selvam College of Technology, Namakkal, Tamil Nadu, India
  3. Professor, Department of Computer Science and Engineering, Selvam College of Technology, Tamil Nadu, India
  4. Professor, Department of Mechanical Engineering, MAM College of Engineering, Trichy, Tamil Nadu, India

Abstract

The paper reviews the existing AI/ML methods first in the general context of predictive maintenance and FDD of RAC systems, then specifically focusing on granular cooling appliances. Perspectives and insights are provided on the reasons why potentially valuable models do not make it into practice more often, and where future research and development should be headed. New emerging topics for decision support systems to include domain knowledge and physics-based modeling to enhance data-driven methods are also highlighted. Finally, the paper ends by making a reasoned comparison among the different generations of methods, introducing the current state of the art regarding other standard comparative metrics, and discussing the most promising solutions. This paper aims to inspire collaborative research in intelligent refrigeration and air conditioning (RAC) maintenance systems by reviewing current literature on state-of-the-art technologies, key conversations, and emerging trends. In addition, the review systematically categorizes existing approaches into data-driven, knowledge- driven, and hybrid frameworks, outlining their respective advantages, limitations, and applicability across various operational scenarios. Emphasis is placed on the role of high-quality datasets, sensor deployment strategies, and feature engineering in improving model robustness and generalization capability. The challenges associated with real-time implementation, scalability, cybersecurity, and integration with building management systems are also discussed in detail. Furthermore, the paper examines the impact of explainable AI, transfer learning, federated learning, and digital twin technologies in advancing predictive maintenance and fault detection and diagnosis (FDD) performance. Practical considerations such as computational cost, model interpretability, deployment feasibility, and lifecycle cost-benefit analysis are evaluated to bridge the gap between academic research and industrial adoption. By synthesizing theoretical advancements with practical constraints, the study provides a comprehensive roadmap for researchers, practitioners, and policymakers aiming to develop resilient, energy-efficient, and intelligent RAC systems for next-generation smart buildings and sustainable infrastructure.

Keywords: Predictive Maintenance, Fault Detection and Diagnosis, Artificial Intelligence, Refrigeration Systems, Energy Efficiency, Digital Twin

[This article belongs to Journal of Refrigeration, Air conditioning, Heating and ventilation ]

How to cite this article:
S. Ramabalan, S. Mahalakshmi, V. Sathiya, N. Godwin Raja Ebenezer. A Review on AI and Machine Learning for Predictive Maintenance and FDD in RAC Systems. Journal of Refrigeration, Air conditioning, Heating and ventilation. 2026; 13(01):15-25.
How to cite this URL:
S. Ramabalan, S. Mahalakshmi, V. Sathiya, N. Godwin Raja Ebenezer. A Review on AI and Machine Learning for Predictive Maintenance and FDD in RAC Systems. Journal of Refrigeration, Air conditioning, Heating and ventilation. 2026; 13(01):15-25. Available from: https://journals.stmjournals.com/jorachv/article=2026/view=241866


References

  1. (2022). The Future of Cooling. International Energy Agency.
  2. (2021). Cooling Emissions and Policy Synthesis Report. United Nations Environment Programme.
  3. Mobley, R. K. (2002). An Introduction to Predictive Maintenance. Butterworth- Heinemann.Elsevier. USA.
  4. Selcuk, Sule. “Predictive Maintenance, Its Implementation and Latest Trends.” Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, (2017). Accessed February 7, 2026. https://doi.org/10.1177/0954405415601640.
  5. Katipamula, S., & Brambley, M. R. (2005). Methods for Fault Detection, Diagnostics and Prognostics for Building Systems – A Review Part. HVAC&R Research, 11(1), 3-25.
  6. Breuker, M.S. & Braun, J.E., 1998a, Evaluating the performance of a fault detection and diagnostic system for vapor compression equipment. HVAC&R Res., 4 (4) (1998) 401–425. https://doi.org/10.1080/10789669.1998.10391412.
  7. Beghi, , Cecchinato, L., Corradi, M., & Rampazzo, M. (2016). A data-driven fault detection and diagnosis strategy for rooftop air conditioning units based on support vector machines. International Journal of Refrigeration, 69, 138–152
  8. Kim, W., & Katipamula, S. (2018). A review of fault detection and diagnostics methods for building systems. Science and Technology for the Built Environment, 24(8), 847–869.
  9. Wang, H., Chen, Y., Li, C., & Li, H. (2020). A hybrid fault diagnosis method for HVAC systems based on deep belief network and Dempster–Shafer evidence theory. Measurement, 151, 107199..
  10. Shi, , Xie, X., Chu, X., & Gadh, R. (2016). Data-driven methods for fault detection and diagnosis in HVAC systems: A review. IEEE Access, 4, 2711–2738..
  11. Pritoni, M., Li, H., Hong, T., Peffer, T., & Granderson, J. (2021). A synthetic building operation dataset. Scientific Data, 8(1), Article 213, 1–16
  12. Granderson, J., Lin, G., Casillas, A., Harding, A., Im, P., & Chen, Y. (2017). Curation and evaluation of ground-truth validated benchmarking datasets for fault detection and diagnostics in building systems.
  13. Mirnaghi, M. S., & Haghighat, F. (2020). [Title not indexed in major open sources] Journal of Building Engineering, 31, 101383.
  14. House, J. , Vaezi-Nejad, H., & Whitcomb, J. M. (2001). An expert rule set for fault detection in air-handling units. HVAC&R Research, 7(4), 347–369. 2001
  15. Schein, J., Bushby, S. T., Castro, N. S., & House, J. M. (2006). A rule-based fault detection and diagnostic method for air-handling units. ASHRAE Transactions, 112(2), 554–567.
  16. Li, H., & Braun, J. E. (2007). An overall performance index for characterizing the economic impact of faults in direct-expansion cooling equipment. International Journal of Refrigeration, 30(2), 299–310.
  17. Du, Z., Fan, B., Jin, X., & Chi, J. (2014). Sensor fault detection and its efficiency analysis in air- handling units using combined neural networks. Energy and Buildings, 85, 441–447.
  18. Zhao, Y., Xiao, F., Wen, J., Lu, Y., & Wang, S. (2015). A data-driven fault detection and diagnosis method for building energy systems using pattern recognition. Energy Conversion and Management, 103, 180–191.
  19. Swider, D. J. (2003). Fault detection and diagnosis in refrigeration systems — A modeling approach. International Journal of Refrigeration, 26(4), 434–444.
  20. Yan, R., Ma, Z., Zhao, Y., & Kokogiannakis, G. (2016). A decision-tree based data-driven diagnostic strategy for air-handling units. Energy and Buildings, 133, 37–45
Regular Issue Subscription Review Article
Volume 13
Issue 01
Received 06/02/2026
Accepted 07/02/2026
Published 26/02/2026
Publication Time 20 Days
5

Login


My IP

PlumX Metrics