中文摘要：

为提高波束形成识别汽车前围板隔声薄弱部位的精度,开发了CLEAN-SC清晰化波束形成声源识别软件。对多种已知模拟声源的识别结果表明：该方法能够显著提高分辨率、衰减旁瓣,更准确地识别单声源及不相干声源,且随迭代次数的增加收敛快、受传声器及通道频响失配等因素的干扰小。某汽车前围板的隔声薄弱部位识别试验结果表明：空调进气口左上角位置是主要薄弱部位,空调进气口内外循环转换阀与阀口贴合不紧密是根本原因。为改善其隔声性能指明了方向,验证了CLEAN-SC清晰化波束形成方法在汽车前围板隔声薄弱部位识别中的有效性及所开发软件的正确性。

英文摘要：

In order to improve the positioning accuracy of sound insulation weak position identification for car dash panel with beamforming method, the sound source identification software is developed based on CLEAN-SC clearness beamforming principle. Simulations and calculations for various kinds of given sources are conducted. The results show that this method can identify single source and incoherent sources more precisely, it can improve resolution and suppress sidelobes dramatically; besides, it not only converges fast with the increase of iterative times, but also is robust to some disturbing factors such as frequency response mismatch of microphones and measurement channels. Experimental results of the sound insulation weak position identification for car dash panel indicate that the top left corner of the air conditioner＇s air inlet is the dominant weak position and the root cause is that the change-over valve for the inside cycle and the outside cycle is poorly fitted on the valve port. All the results provide a guide for the improvement of the dash panel＇s sound insulation performance, and validate the effectiveness of the CLEAN-SC clearness beamforming method in the sound insulation weak position identification of car dash panel and the correctness of the developed software.