2025, Vol. 12, No. s4. - go to content...

Permanent address of this page - https://resources.today/en/11faor425.html

Метаданные этой статьи так же доступны на русском языке

DOI: 10.15862/11FAOR425 (https://doi.org/10.15862/11FAOR425)

Full article in PDF format (file size: 824.4 KB)

For citation:

Lakhtin A.A., Goncharov A.V. Key algorithms for fault diagnosis and prevention in automated in-flight catering heating systems. Russian journal of resources, conservation and recycling. 2025; 12(s4). Available at: https://resources.today/PDF/11FAOR425.pdf (in Russian). DOI: 10.15862/11FAOR425

Key algorithms for fault diagnosis and prevention in automated in-flight catering heating systems

Lakhtin Alexey Artemovich
K.G. Razumovsky Moscow State University of Technologies and Management (the First Cossack University),
Moscow, Russia
E-mail: lakhtin.l@mail.ru

Goncharov Andrey Vitalievich
K.G. Razumovsky Moscow State University of Technologies and Management (the First Cossack University),
Moscow, Russia
E-mail: a.goncharov@mgutm.ru

Abstract. Modern requirements for safety and efficiency of aviation systems necessitate the development and implementation of advanced methods for fault diagnosis and prevention in critical aircraft components, including automated in-flight catering heating systems, which represent complex electromechanical complexes with a high degree of integration into the overall architecture of aviation equipment. The study is devoted to a comprehensive analysis of algorithmic approaches to fault diagnosis and prevention in in-flight catering heating systems through the prism of modern machine learning methods, predictive analytics, and reliability theory of technical systems. The subject of the study is mathematical models, algorithms, and hardware-software solutions for ensuring fail-safety of galley heating systems under operating conditions of modern aircraft. Classical methods of reliability analysis are considered, including FMEA (Failure Mode and Effects Analysis), FDIR (Fault Detection, Isolation and Recovery), as well as modern approaches based on machine learning and neural network technologies. The evolution of approaches to fault diagnosis from reactive maintenance strategies to proactive predictive analytics methods is analyzed, allowing to predict component degradation and prevent failures before they occur. Key factors affecting the effectiveness of diagnostic algorithms have been identified, including the quality and completeness of sensor data, monitoring system architecture, computational resources of onboard systems, and real-time information processing requirements. The specific features of applying machine learning methods in the aviation industry have been investigated, due to strict certification requirements, limited computational resources, and the need to ensure determinism of results. An integrated model of a fault diagnosis and prevention system has been developed, combining the advantages of classical reliability analysis methods with the capabilities of modern machine learning algorithms, providing a multi-level data processing architecture from primary signal filtering to the formation of control actions. The scientific novelty of the study lies in the development of a hybrid approach to fault diagnosis, combining deterministic algorithms with adaptive machine learning methods, as well as in formalizing the criteria for selecting optimal algorithmic solutions for various operating modes of in-flight catering heating systems. The practical significance of the work is determined by the possibility of applying the developed algorithms to improve the reliability of aviation systems, reduce maintenance costs, and minimize risks associated with failures of in-flight catering equipment.

Keywords: automated heating system; in-flight catering; fault diagnosis; predictive analytics; machine learning; FDIR; aviation safety; technical systems reliability; neural networks; monitoring algorithms; fail-safety

Download article in PDF format