Failure analysis and finite element simulation for structural systems in an unmanned aerial vehicle

• Main Authors: Wang, Bing, Ng, Poh Huat, Elhadidi, Basman Mohamed Nabil Ahmed Khairat, Ang, Hui Shan, Moon, Seung Ki
• 其他作者: School of Electrical and Electronic Engineering
• 格式: Conference or Workshop Item
• 語言: English
• 出版: 2020
• 主題: Engineering; UAV; Failure analysis
• 在線閱讀: https://hdl.handle.net/10356/141497
• 機構: Nanyang Technological University

In this paper, failure analysis of an unmanned aerial vehicle (UAV) has been conducted through failure mode and effects analysis (FMEA) and finite element (FE) simulation. For FMEA, a system is broken down into subsystems, and possible failures are considered for each subsystem. Each failure is then ranked in 3 categories: severity, occurrence, and difficulty of detection. Each category has 1 of 5 ranks: very low, low, moderate, high, and very high. Since these ranks are linguistic variables that cannot be properly assigned a definitely value, fuzzy logic is applied. Using a membership function and a defuzzification formula, the variables are defuzzified. Grey relational analysis is applied to factor in the relationship between the variables. With these methods applied, the battery failure and rotor arm failure (including motors, ESC arm structure) were found to be the most critical failures followed by avionics failure and structural failure. For the FE simulation, a 3D model is created. The stress distribution and fatigue life of fuselage are simulated by using the commercial software Comsol. The fuselage was made of Al 7076 T6 with an ultimate tensile strength of 572 MPa. The stress distribution at different loads (5 kg, 50 kg 75 kg) are simulated. According to the simulation results, the maximum stress occurs at the corner of the fuselage. The fatigue life under different loads were also simulated. The potential failure mode and risk level analysis enables the identification of critical component and prediction of system failures. With FE simulation, the limit of the structure is simulated. Maximum load the structure can support is determined, the simulation of fatigue life cycles enables the prediction of structure failures.