基于响应面法的乘用车消声器声学性能优化

Acoustic Performance Optimization of the Exhaust Muffler for a Car Based on Response Surface Method

针对某乘用车车内、外消声器中频段(0.5~1.5kHz)降噪能力较差的问题,利用响应面法对消声器在中频段的传递损失进行优化.首先,采用Latin超立方法对插入管长度、穿孔板位置和穿孔率等设计变量进行采样,采用声学有限元法计算各样本点在中频段的传递损失;然后,利用Kriging模型建立中频段传递损失的响应面模型,并验证其计算精度;最后,利用遗传算法对该响应面模型的计算结果进行优化.结果表明,与优化前消声器在中频段的传递损失(27.05dB)相比,优化后消声器在中频段的传递损失(41.38dB)增加了14.33dB,其消声性能得到了明显改善.

The noises in the interior and exterior of a vehicle are measured in a stationary test,and the results reveal that noise reduction capability of its exhaust muffler is quite weak in the mid-frequency.In order to improve the mid-frequency transmission loss（TL）of the muffler,optimization is conducted based on the response surface method（RSM）.Firstly,the length of insert tubes,position of perforated plates,perforation rate,etc.are selected as design variables and are sampled by the Latin hypercube method,and the frequency-averaged TL of each sampled point is calculated by acoustic finite element method.Secondly,an RSM model of the mid-frequency TL is developed by the Kriging surrogate technique,and its accuracy is validated by re-analysis.Finally,based on this model,the mid-frequency TL is optimized by genetic algorithm.After optimization the mid-frequency TL is 41.38 dB increasing 14.33 dB than the original value 27.05 dB,and also the overall noise reduction performance of this muffler is improved significantly.The RSM-based optimization procedure in this paper provides a guideline for the acoustic performance design of the mufflers.