基于遗传算法的动态优化波叠加噪声源识别方法

Sound Source Identification Method with Dynamic Optimal Wave Superposition Algorithm Based on Genetic Algorithm

针对现有声全息以及波束形成等方法中重建声场的虚假声源问题,提出一种利用遗传算法搜索声源位置的波叠加噪声源识别方法。该方法通过传声器阵列测量声音信号,基于时间延迟算法进行声源面的声压预估;选取预估声压峰值点作为等效源的初始位置;根据初始识别结果确定声源位置搜索的三维空间范围,以重建传声器声压误差函数作为位置评价指标,利用遗传算法动态地优化等效声源的空间位置,并实现声场的波叠加重构。对该方法进行仿真试验,得到的识别结果中旁瓣引起的虚假声源强度下降到真实声源的10%以下。试验结果表明利用该技术重建声场时,与传统的全息和阵列技术相比,可以有效消除虚假声源以及旁瓣效应,与静态波叠加方法相比,可以取得准确的声源位置和重建声压值。

In order to solve the false source problem in the noise source identification with the existing methods such as beamforming and acoustic holography,a wave-superposition method based on the genetic algorithm is proposed.The sound pressure signal data are measured with a microphone array,the sound pressure of the source plane is pre-estimated with a time-delay based method;the positions of the equivalent sound sources are set at the peaks of pre-estimated pressure;the 3-dimensional search space is determined according to the pre-estimated results,and the evaluation index is the error function value of the reconstructed sound pressure of the microphones,then,the positions of the equivalent sources are optimized with the genetic algorithm,and the sound field is reconstructed with the wave-superposition method.In the simulation results with this method,the relative strengths of the false sources produced by the side lobe effect are reduced to less than 10% of the real source.The simulations and experiments indicate that the false sources and the side lobe effect in the traditional methods are eliminated effectively;the identified location and the reconstructed sound pressure are more accurate than those acquired by the static wave superposition algorithm.