A Comparative Study of Optimization Algorithms for 6DOF Design in Electric Vehicle Powertrain Mounting Systems

<div class="section abstract"><div class="htmlview paragraph">Optimizing engine mounting systems is a complex task that requires balancing the isolation of vehicle vibrations with controlling powertrain movement within a limited dynamic envelope. Six Degrees of Freedom (6DOF) optimization is widely used for mounting stiffness and location optimization. This study investigates the application of various optimization algorithms for 6DOF analysis in engine mount design, where the system’s stochastic behaviour and probabilistic characteristics present additional challenges. Selecting an appropriate optimization framework is essential for achieving accurate and efficient NVH results.</div><div class="htmlview paragraph">Recent advancements in research have introduced several 6DOF optimization algorithms to determine the optimal stiffness and location of engine mounts. The study evaluates a range of optimization methods, including Simultaneous Hybrid Exploration that is Robust, Progressive and Adaptive (SHERPA), Quadratic Programming (QP), Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Non-dominated Sorting Genetic Algorithm III (NSGA-III), Nelder-Mead Simplex (NMS), Multi-Start Local Search (MSLS), Response Surface Method (RSM), and Simulated Annealing (SA). This paper conducts a comparative analysis of these algorithms through a detailed case study, focusing on key parameters required to achieve convergence.</div><div class="htmlview paragraph">By systematically comparing these optimization methods based on criteria such as time efficiency, accuracy, data generation, and robustness, this paper provides comprehensive guidelines for selecting the optimal approach to engine mount design. The findings contribute significantly to mounting system design, offering efficient solutions for enhancing vehicle performance, comfort, and durability.</div></div>