The Principle And Applications Of Noise Source Identification Based On Beamforming

When People increase continuously the request of quality of acoustical environment, the noise pollution presents worsening trend. The traffic noise caused by automobile is one of creating noise pollution of the uppermost factors. Controlled of car-radiation-noise is the problem which production business must consider. But the car-radiation-noise effective controlled must be with the precondition of accurate identification of noise source.First, this article introduces all main methods of noise source identification, and puts forward simply comparing and analyzing on it. It reviews developing process of noise identification methods based on beamforming, and puts in some problems which will be solved by this article.In the discussion of one-dimension array beamforming method, it introduces the basic principle of method, along with the major performance of system response, array pattern, effective bandwidth and so on. Applying spherical wave theory of near field condition modify on traditional beamforming arithmetic, making it to the car noise source identification. Additionally, it introduces cross-spectral beamforming arithmetic, analyzes predominance of suppress incoherent noise on the method. By numerical simulation, it discusses some factors of influencing beamforming capability which are analytical frequency, measuring distance, microphone separation, microphone number and array dimension and so on. In using of one-dimension beamforming, the effective bandwidth of system must cover frequency range of sound source, must avoid appearing the condition of ghost image and lack of resolution. Balance the relationship between resolution and effective bandwidth, choice suitable measuring distance, using microphone number as small as possible completes sound source identification measuring. The comparing of delay-and-sum arithmetic and cross-spectrum arithmetic incarnates obvious predominance of latter in sidelobe level and resolution.To obtain distribution map of sound field, beamforming method is extended two-dimension condition. Introducing basic principle of two-dimension delay-and-sum beamforming, it analyzes array pattern, maximun sidelobe level (MSL), resolution, cut-off frequency and so on. Array pattern and maximum sidelobe level is the parameter which only relates with microphone position distributing of the array, by the way of evaluation criterion of array capability quality, measuring array of capability superiority should have lower maximum sidelobe level. In the near field condition of finite measuring distance, delay time of the microphone should make corresponding correction, at the same time, target area of measuring should exist in 30 degree of the array angle range. The spatial sampling theorem restricts application frequency range of regular array, optimized array have the sampling of various different separation, breaking through the previous limitation, beamforming method may use to the identification of high frequency sound source. In the numerical simulation of delay-and-sum method, four arrays be compared, predominance is showed by the optimized array. In this base, it derives cross-spectrum beamforming method which is in two-dimension condition, by numerical simulation, it proofs comparative advantage of that method in sidelobe level and resolution.For further understanding the character of two-dimension beamforming, by numerical simulation it analyzes chief parameters which are array diameter, measuring distance, microphone numbers, analysis frequency and so on. Regular array and optimized array carry on multi-angle correlation and discussion. In regard to regular array, influence of the parameters are basic identical with one-dimension condition. In regard to optimized array, maximum sidelobe level can not rapidly amplify with array diameter amplification. The increase of microphone numbers is really not to bring the promotion of performance, in the suitable microphone numbers, the optimization of microphone position have the better effect. For general measuring, microphone numbers about 60 is more practicably. The Superiority of Beamforming method is the identification in high frequency source (above 1200Hz), that may combine with the superiority of near field acoustical holography method in low-frequency (below 1200Hz), to reach the best effect of noise source identification.Two applications have been outlined. the one is the linear array measurement in semi-anechoic chamber, the other is the cross array measurement for the moving vehicle.In the application of one-dimension homogeneous linear array of beamforming of semi-anechoic chamber, making use of two-rows linear array of pass noise measuring system in semi-anechoic chamber, to realize one-dimension identification of sound source position. By spectrum analysis of noise source signal, finding main component of noise source comes from using rotational speed of drum for various order harmonic wave of fundamental frequency. The second order harmonic noise both side of automobile exist engine nearby. On the left side, the third order harmonic noise and the fourth order harmonic noise are in car axle of front wheel nearby; On the other side, cooling system of automobile engine radiated main noise on this orders. The abnormality of the fourth order harmonic noise on the left side make for SPL of left side greater than normal. the position of this noise source is 1 .24±0.08m, near the front axle. By testing it dues to the malfunction from left side front drum of semi-anechoic chamber. Analytical result is accurate and reliable and gets the approval of foreign equipment supplier.the pass noise measurement mothod of accelerated moving vehicle based on Nearfield Acoustic Hohography. these parameters doesn't adapt to Beamforming. But the results of Beamforming fit the factual position of source really . this application verified Beamforming method on the other hand.Theoretical analysis, numerical simulation and practical application even verity the availability and reliability of beamforming noise source identification method. Previous work supplies technical support for the development of beamforming measurement instrument, the widespread application accumulates experience for the method.