Design of a multi-component system-based fixed-wing unmanned aerial vehicle maintenance policy and its case study

被引:1
作者
Liu, Guangshuai [1 ]
Li, Xurui [1 ]
Sun, Si [2 ]
Zhao, Xing [3 ]
Li, Bailin [1 ]
机构
[1] Southwest Jiaotong Univ, Sch Mech Engn, Chengdu 610031, Sichuan, Peoples R China
[2] Chinese Acad Sci, Inst Opt & Elect, Chengdu 610209, Sichuan, Peoples R China
[3] Chenghang Innovat Inst Intelligent Aerocraft Engn, Chengdu 600100, Sichuan, Peoples R China
关键词
Unmanned aerial vehicle; Knowledge-driven; Predictive maintenance; Multi-component system; Maintenance optimization; FAULT-DIAGNOSIS; MODEL;
D O I
10.1016/j.cie.2024.110701
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
At present, flight vehicle structure maintenance is undergoing a paradigm shift from preventive maintenance depending on calendar time and flight cycle to condition-based maintenance (CBM) using structural condition monitoring data. This shift is particularly relevant as unmanned aerial vehicle (UAV) systems increasingly undertake aerial missions. However, unlike CBM for single components, most UAV systems consist of multiple interconnected components. Optimizing pre-flight maintenance and ensuring traceability of inspections from a system-level perspective remains challenging. This study introduces a multi-layered decision-making policy for UAV reliability and stability multi-component evaluation, enhancing service condition monitoring and maintenance evaluation. The system adopts a knowledge-driven approach to develop a comprehensive maintenance framework for multi-component UAVs. It enables detailed damage assessment, effective management, predictive capabilities, and optimization strategies. By integrating and reasoning across knowledge-based, geometric, and decision-making models, the system supports dynamic maintenance and continuous iterative enhancements. To further reduce system complexity, this paper created a risk grating evaluation system. Within the framework of individual components, decision-making rules are established to optimally determine the components needing preventive maintenance when activated by decisions from risk assessments. Additionally, a new soft failure threshold model to determine the optimal maintenance decision variables is defined. This model incorporates general knowledge of the system and subsystems, further optimizing predictive reliability and economic dependency. Finally, the effectiveness and feasibility of this policy are validated through a fixed-wing UAV maintenance case study. The results demonstrate that the proposed framework holds significant promise for maintenance management in aircraft systems.
引用
收藏
页数:19
相关论文
共 44 条
[1]  
Ahmad M. W., 2020, AutoNet: Meta-model for Seamless Integration of Timed Automata and Colored Petri Nets, P307, DOI [10.1007/978-3-030-47679-326, DOI 10.1007/978-3-030-47679-326]
[2]   Intelligent framework for automated failure prediction, detection, and classification of mission critical autonomous flights [J].
Ahmad, Muhammad Waqas ;
Akram, Muhammad Usman ;
Ahmad, Rashid ;
Hameed, Khurram ;
Hassan, Ali .
ISA TRANSACTIONS, 2022, 129 :355-371
[3]   Computational framework for real-time diagnostics and prognostics of aircraft actuation systems [J].
Berri, Pier Carlo ;
Dalla Vedova, Matteo D. L. ;
Mainini, Laura .
COMPUTERS IN INDUSTRY, 2021, 132
[4]   An RTK UAV-Based Method for Radial Velocity Validation of Weather Radar [J].
Chen, Yubao ;
Li, Lu ;
Ye, Fei ;
Kang, Boshi ;
Wang, Xiaopeng ;
Bu, Zhichao ;
Zhu, Moyan ;
Yang, Qian ;
Shao, Nan ;
Zhang, Jianyun .
REMOTE SENSING, 2024, 16 (07)
[5]  
Cui A., 2020, Intelligent health management of fixed-wing UAVs: A deep-learningbased approach, DOI 10.1109ICARCV50220.2020.9305491
[6]   Micro aerial vehicle with basic risk of operation [J].
de Lucena, Alysson Nascimento ;
da Silva, Bruno Marques Ferreira ;
Goncalves, Luiz Marcos Garcia .
SCIENTIFIC REPORTS, 2022, 12 (01)
[7]   Modeling and Analysis of Multiproduct Multistage Manufacturing System for Quality Improvement [J].
Du, Shichang ;
Xu, Rui ;
Li, Lin .
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS, 2018, 48 (05) :801-820
[8]   Towards operational UAV-based forest health monitoring: Species identification and crown condition assessment by means of deep learning [J].
Ecke, Simon ;
Stehr, Florian ;
Frey, Julian ;
Tiede, Dirk ;
Dempewolf, Jan ;
Klemmt, Hans-Joachim ;
Endres, Ewald ;
Seifert, Thomas .
COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2024, 219
[9]   Training of physics-informed Bayesian neural networks with ABC-SS for prognostic of Li-ion batteries [J].
Fernandez, Juan ;
Corbetta, Matteo ;
Kulkarni, Chetan S. ;
Chiachio, Juan ;
Chiachio, Manuel .
COMPUTERS IN INDUSTRY, 2024, 155
[10]  
Gilchrist W., 1993, International Journal of Quality Reliability Management, V10, P16, DOI [DOI 10.1108/02656719310040105, 10.1108/02656719310040105]