Genetic algorithm-based optimization of helical gear pair with non-standard center distances: validated through FEA and strain gauge technique

In mechanical systems, helical gears are essential for transmitting power and motion between parallel shafts. Optimizing key parameters like addendum, center distance, tooth profile, and material selection is critical to improve performance and durability. Multi-objective optimization using genetic algorithm (MOO-GA) is employed in this research to optimize the design of innovative non-standard center distance helical gear pairs. The MOO-GA approach optimizes for three main objectives simultaneously: (i) balancing tooth root strength to maximize load-carrying capacity, (ii) optimizing the specific sliding ratio during tooth engagement and disengagement (approach and recess actions) for reduced noise and improved meshing efficiency, and (iii) maximizing the sum of the addendum modification coefficients. MOO-GA iteratively searches for optimal solutions by manipulating key design variables (x₁ and x₂), leading to consistent convergence and significant addendum modifications. Finite element analysis (FEA) with ANSYS software is used to evaluate how optimization reduces tooth stress. Experimental strain gauge technique involves strategically placing strain gauges on gear teeth identified by the FEA model. These strain gauges directly measure real-world strain under load, validating the predicted strains from FEA. Validation with a real-time CAD model confirms the optimized design exhibits reduced tooth root stress and contact stress compared to helical gear pairs with standard distance. The research demonstrates the effectiveness of MOO-GA in creating superior helical gear designs that meet performance requirements and potentially offer weight and space savings. Strain gauge validation strengthens confidence in FEA results and provides valuable data for further optimization refinement.