Spur Gear Design via Particle Swarm

This study aims to employ a nature inspired algorithm Particle Swarm (PS) to seek minimum center distance for desired life of a compact spur gear set to tranulsmit a given power satisfying a number of constraints imposed. The PS, which is relatively recent heuristic search technique, imitates social behavior of birds flocking by sharing information among their members in multiple dimensions in space looking for the best food sources. The PS algorithm employs the generation of guided random positions when they search for the global optimum point. The algorithm does not require the evaluation of gradients of the objective functions since it is a derivative free optimization technique. Heuristic optimization techniques are used to overcome most of the limitations found in the traditional techniques. In this study, an algorithm which is a variant of the PS called the Repulsive Particle Swarm (RPS) is utilized. In this algorithm, each particle can make a wider search in its neighborhood to optimize itself due to stronger local search abilities. It is demonstrated that the PS performed and obtained better convergence reliability on the global optimum point than a more traditional method (Feasible Direction Gradient Search-FDGS).