基于声阵列技术的噪声源识别.pptVIP

中国科学院声学研究所北京声望声电技术有限公司 噪声源识别方法 声强法 稳态噪声源 长平均时间 测量面上点点测量 近场声全息： 采用多路传感器在 波长的距离测量 通过测量近场的声学特性，如声压，速度，运用声辐射计算出远场的噪声分布。 噪声源识别方法 声阵列法 利用传声器阵列，线阵，面阵，十字，T字 测量复 FFT, 计算相位特性 声阵列与声全息技术比较 阵列的构成 （29个传声器） 阵列的硬件结构 Hardware: PHASED ARRAY with 29 microphones line-configuration with microphone distance 0.3?m, 0.15?m and 0.075?m cross-configuration with microphone distance 0.3? or 0.08?m or 0.04?m electret microphones with 1/4” capsule, 11 mV/ Pa, frequency range up to 5000?Hz Multichannel power supply for microphones 阵列的算法 Software for the analysing and beamforming by phase shift 阵列的指向性 Directivity Directivity Directivity Directivity Directivity 非稳态发动机噪声 非稳态声源 阵列的现场实验 汽车通过噪声测量 Example Wind Tunnel Trans-Rapid, 350 km/h 结论 声阵列法是目前最强声源识别系统 阵列法可运用于，稳态声源，非稳态声源，和移动声源 BBM 的阵列系统可以是一字，十字阵列。 吴群力 博士 基于声阵列技术的 噪声源识别 阵列的应用 稳态声源 stationary sound sourcese.g. engine test bench, video tape recorder 非稳态声源 instationary sound sourcese.g. engine run up test with increasing motor speed or sorter machine 运动声源 moving sound sourcese.g. railway, high speed train 1 15 16 29 (8) x y z directivity index 10 lg (?(?)2) dB for an array configuration with N = 15 microphones, d = 0,08 m ;?? 500 Hz ???????? 1000 Hz -- -- -- 2000 Hz d/? = 0,25 d/? = 0,50 d/? = 1 gleichgewichtet cos2-Wichtung optimiert 500 Hz 1000 Hz 2000 Hz Example Loudspeaker upper: woofer lower: tweeter left: Order spectra of SPL at the array center microphone, distance: 1?m right: Ordercut of SPL over engine speed at 19.5th and 33rd order left: contour plot of SPL in the 19.5th order band, engine speed 3250 rpm right: contour plot of SPL in the 33rd order band, engine speed 3500 rpm 1250-Hz-1/3 octave, v = 80 km/h side mirror normal side mirror optimized Müller-BBM VibroAkustik Systeme * * */32 ? CRITERIA PHASED ARRAY NEARFIELD ACOUSTIC HOLOGRAPHY principle beamforming by adding microphone signals soundfield transformation using cross power spectra number