Optimization of tooth modifications for spur and helical gears using an adaptive multi-objective swarm algorithm
Résumé
Metaheuristic methods have proved to be suitable for solving complex multi-criteria optimization problems. In this paper, a modified particle swarm algorithm has been implemented in order to improve the quasi-static behavior of a power transmission gearbox, thus optimizing various objectives such as the maximum contact pressure on the gear flanks, the root-mean-square of the loaded transmission error signal, the tooth bending stress, and/or the pressure-speed factor. For narrow-faced spur gears, the comparison between optimal solutions found by the algorithm and the so-called master curve shows quite good agreements. The chosen form of the profile modifications, linear or quadratic, is then discussed. Finally, the robustness of the optimal solutions is tested to guarantee their efficiency against variable shaft misalignments.