Dimensional Optimization of Parallel Mechanisms Based on Differential Evolution

Abstract Dimensional optimization of parallel mechanisms forms one of the theoretical foundations for prototype development and is critical to effective prototype design. In this study, a three-degree-of-freedom remote-center parallel mechanism is examined. The problem of maximizing the average transfer index within the desired conical workspace is investigated using the motion/force transfer index and an intelligent algorithm. An optimization model for the mechanism's dimensional parameters is formulated, with the maximization of the average transmission index as the objective. Feasibility rules are applied to handle constraints. The escape differential evolution algorithm with sine-selection-based dual mutation strategies (SDEDE) is employed to address the optimization problem. To evaluate the generality and effectiveness of the SDEDE algorithm, numerical experiments were conducted on three parallel mechanism dimensional parameter optimization cases. The results indicate that the SDEDE algorithm provides higher solution accuracy and faster convergence and generally outperforms other comparable algorithms. Consequently, the performance of the optimized parallel mechanism shows substantial improvement.