Efficient multidisciplinary modeling of aircraft undercarriage landing gear using data-driven Naïve Bayes and finite element analysis

• Published in: Multiscale and Multidisciplinary Modeling, Experiments and Design Vol. 7; no. 4; pp. 3187 - 3199
• Main Authors: Al-Haddad, Luttfi A., Mahdi, Nibras M.
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.09.2024
• Subjects: Characterization and Evaluation of Materials; Engineering; Mathematical Applications in the Physical Sciences; Mechanical Engineering; Numerical and Computational Physics; Original Paper; Simulation; Solid Mechanics; Naïve Bayes; Finite element analysis; Artificial intelligence; Landing gear; Machine learning
• ISSN: 2520-8160; 2520-8179
• DOI: 10.1007/s41939-024-00389-4

Advancements in aircraft design and production necessitate exhaustive simulations of critical components, such as landing gear, to ensure optimal performance and safety. This paper presents a novel approach that combines finite element analysis (FEA) and machine learning (ML) for comprehensive simulations of aircraft landing gear during the landing phase, taking into account aircraft weight and wind force. In this paper, FEA-based simulations are performed accordingly where the deformation, stress, strain, and strain energy are all computed. A duration of 4 h and 42 min was the computed time for one run of ANSYS simulation for the landing touch methodology. Depending on the simulated deformation results, the values of stress, strain, and strain energy are forecasted based on the ML Naïve Bayes (NB) model. The proposed method is highly time effective as the forecasts would endure 1 s instead of around 5 h of calculation. To strengthen and present the model’s dependability, root mean square error (RMSE) and the coefficient of determination are called out, with values corresponding to 0.9% and 1.000, respectively. Overall, it has been evidenced that FEA–ML-based analyses are far more time effective and preferable over traditional FEA-based simulations. The proposed methodology is geometrical friendly as it might be applied over any part of an aircraft, hence advancing aircraft design capabilities.