With the development of new energy and intelligent vehicles, aerodynamic noise problem of pure electric vehicles at high speed has become increasingly prominent. The characteristics of the flow field and aerodynamic noise of the rearview mirror region were investigated by large eddy simulation, acoustic perturbation equations and reduction order analysis. By comparing the pressure coefficients of the coarse, medium and dense grids with wind tunnel test results, the pressure distribution, and numerical accuracy of the medium grid on the body are clarified. It is shown from the flow field proper orthogonal decomposition of the mid-section that the sum of the energy of the first three modes accounts for more than 16%. Based on spectral proper orthogonal decomposition, the peak frequencies of the first-order mode are 19 and 97 Hz. As for the turbulent pressure of side window, the first mode accounts for approximately 11.3% of the total energy, and its peak appears at 39 and 117 Hz. While the first mode of sound pressure accounts for about 41.7%, and the energy peaks occur at 410 and 546 Hz. Compared with traditional vehicle, less total turbulent pressure level and total sound pressure level are found at current electric vehicle because of the limited interaction between the rearview mirror and A-pillar.

随着新能源和智能汽车的发展，纯电动汽车高速行驶的空气动力噪声问题日益突出。通过大涡模拟、声扰动方程和降阶分析，研究了后视镜区域的流场和气动噪声特性。通过将粗、中、密网格的压力系数与风洞试验结果进行比较，明确了中网格在车身上的压力分布和数值精度。从中段流场适当正交分解可知，前三种模式的能量之和占16%以上。基于频谱适当的正交分解，一阶模式的峰值频率为 19 和 97 Hz。对于侧窗湍流压力，第一模态约占总能量的 11.3%，其峰值出现在 39 和 117 Hz。而第一模态声压约占 41.7%，能量峰值出现在 410 和 546 Hz。与传统车辆相比，由于后视镜和A柱之间的相互作用有限，当前电动汽车的总湍流压力水平和总声压水平较低。