Optimal structural design of helicopter components using tornado optimizer with Coriolis forces

Abstract To address complex mechanical engineering design issues, this study presents a novel optimization algorithm, the modified tornado optimizer with Coriolis forces (MTOC), which is enhanced by artificial neural networks (ANNs). To strike a balance between exploration and exploitation in high-dimensional search spaces, MTOC mimics the dynamic transformation of windstorms into tornadoes, drawing inspiration from the natural development of tornadoes influenced by Coriolis forces. The algorithm exhibits enhanced convergence behavior and optimization accuracy by incorporating ANN techniques for performance improvement and hyperparameter adjustment. Multiple mechanical component design challenges, such as those involving rolling element bearings, Ravigneaux planetary gears, Belleville springs, and helicopter hinge arms, are used to validate the efficacy of MTOC. Comparative results with existing metaheuristic algorithms show MTOC consistently outperforms others in terms of best fitness values, stability (low standard deviations), and computational efficiency, making it a powerful tool for multidisciplinary engineering optimization.