Design and multi-objective optimization of broadband sound insulation structure for mobile power supply vehicles

In response to the broadband noise control requirements of power supply vehicles, this paper proposes a multi-objective optimization framework for composite sound insulation structures, designed to simultaneously improve acoustic performance, reduce structural weight, and minimize manufacturing costs. An acoustic-solid coupling numerical model is established by integrating the Johnson-Champoux-Allard model with the Delany-Bazley empirical model, enabling accurate characterization of the sound absorption and sound insulation behaviors of each layer. Perfectly matched layers and Floquet periodic boundary conditions are employed to simulate non-reflective acoustic fields and periodic structural configurations, respectively. Based on this modeling approach, a multi-objective optimization model is formulated with sound insulation efficiency, structural weight, and production cost as objective functions. To effectively address this trade-off, an improved NSGA-III algorithm is proposed. Key enhancements include a novel grouping crossover strategy (GCS) based on acoustic features and an adaptive mutation operator. These improvements enable a more balanced compromise between global exploration and local exploitation, which is essential for optimization in complex, coupled design spaces. The optimized algorithm successfully generates a well-distributed Pareto front, offering a set of high-quality design alternatives. The preferred solution from the Pareto set achieves an 8.75% improvement in sound insulation performance with negligible changes in weight and cost, demonstrating the effectiveness of the multi-objective optimization approach.