基于NSGA-Ⅱ算法对发动机噪声激励下的整车声学包优化

Vehicle Acoustic Package Optimization Under Engine Noise Excitation Based on NSGA-Ⅱ Algorithm

通过整车声学包合理设计能有效的改善发动机传至车内的噪声。基于统计能量法,利用Hypermesh和VA One软件搭建整车SEA模型,对比测试与仿真分析的PBNR值验证模型的精度。前围板作为发动机到驾驶室的主要传递路径,以前围内外侧声学包材料的厚度、属性、堵孔件厚度及声学材料覆盖率为自变量,利用创建的SEA模型得到主驾右耳的PBNR值、声学材料的总质量及总价格并作为响应。在MATLAB中建立自变量与响应间的数学模型,对各响应赋予一定权重,利用非支配排序遗传算法(NSGA-Ⅱ)获得Pareto最优解。通过SEA模型分析最优解及任意三组Pareto解的响应,验证最优解的可信性;对比原声学包响应:PBNR值提高2.0dB、总重量降低8.5%、价格降低3.6%。分析表明,通过对SEA模型进行PBNR分析结合NSGA-Ⅱ算法能快速得到整车的最优声学包设计方案。

The vehicle interior noise transmitted from engine can be effectively optimized via the rational design of vehicle sound package.The SEA model of a car is built by using HyperMesh and VA One softwares based on the statistical energy analysis.The accuracy of the model is verified by comparing simulation PBNR with test results.As the main transfer path from the engine to the vehicle cabin,the properties and thickness of the bulkhead acoustic materials,as well as the thickness of the plugging parts will be taken as independent variables.The power based noise reduction(PBNR)of driver’s right ear,total weight and the total price of sound package are obtained in response by the SEA model.The mathematical model between independent variable and response is established in MATLAB,and each response is given to a certain weight,and the Pareto optimal solution is gained by using the non dominated sorting genetic algorithm(NSGA-Ⅱ).By analyzing the response of the optimal solution and stochastic three Pareto solutions with SEA model,the credibility of the optimal solution is verified as well.Compared with the original sound package,it is found that the PBNR increased by 2.0dB,the total weight decreased by 8.5%,and the price decreased by 3.6%.The analysis shows that the PBNR analysis of SEA model combined with NSGA-Ⅱ algorithm can quickly get the optimal sound package design scheme of the whole vehicle.