The buzz, squeak, and rattle (BSR) problem is one of the most important issues affecting vehicle satisfaction related to unexpected noise and vibration and has been attracting increased attention because of the rise of electric and hybrid vehicles. However, until recently, the vehicle underbody BSR has not been studied as extensively as the upper-body or interior BSR owing to a number of complex issues, such as the lack of suitable methods for measuring or visualizing the underbody BSR. This study proposes a method for localizing the rattle noise source in the vehicle underbody using the accelerations measured during a vehicle driving test. The acceleration signals are denoised by considering the characteristics of rattle; then, a series of rattle-induced impulse signal units included in the denoised acceleration signals are identified via machine learning at the level of BSR experts. Finally, the accelerometer closer to the rattle noise source is quantitatively determined by analyzing the correlation indexes in the frequency domain based on the properties of the rattle-induced impulse signal units, which depend on their relative position from the rattle noise source. The proposed method is based only on the accelerations measured from the driving test and the rattle signal transmission characteristics, and does not require complex vehicle dynamics analysis; therefore, it can be widely used regardless of the vehicle type at low cost.

嗡嗡声、吱吱声和嘎嘎声 (BSR) 问题是影响与意外噪音和振动相关的车辆满意度的最重要问题之一，并且由于电动和混合动力汽车的兴起而引起越来越多的关注。然而，直到最近，由于许多复杂的问题，例如缺乏合适的方法来测量或可视化车身底部 BSR，车辆底部 BSR 的研究还没有像上部或内部 BSR 那样广泛。本研究提出了一种使用车辆驾驶测试期间测量的加速度来定位车辆底部的嘎嘎声源的方法。考虑嘎嘎声特性对加速度信号进行去噪；然后，在 BSR 专家级别通过机器学习识别包含在降噪加速度信号中的一系列嘎嘎声引起的脉冲信号单元。最后，通过分析频域中的相关指标，基于嘎嘎声感应脉冲信号单元的特性，定量确定更接近嘎嘎声噪声源的加速度计，这取决于它们与嘎嘎声噪声源的相对位置。所提出的方法仅基于驾驶测试中测得的加速度和嘎嘎声信号传输特性，不需要复杂的车辆动力学分析；因此，无论何种车型，它都可以以低成本广泛使用。通过分析频域中的相关指标，基于噪声引起的脉冲信号单元的特性，定量确定更靠近噪声源的加速度计，该特性取决于它们与噪声源的相对位置。所提出的方法仅基于驾驶测试中测得的加速度和嘎嘎声信号传输特性，不需要复杂的车辆动力学分析；因此，无论何种车型，它都可以以低成本广泛使用。通过分析频域中的相关指标，基于噪声引起的脉冲信号单元的特性，定量确定更靠近噪声源的加速度计，该特性取决于它们与噪声源的相对位置。所提出的方法仅基于驾驶测试中测得的加速度和嘎嘎声信号传输特性，不需要复杂的车辆动力学分析；因此，无论何种车型，它都可以以低成本广泛使用。所提出的方法仅基于驾驶测试中测得的加速度和嘎嘎声信号传输特性，不需要复杂的车辆动力学分析；因此，无论何种车型，它都可以以低成本广泛使用。所提出的方法仅基于驾驶测试中测得的加速度和嘎嘎声信号传输特性，不需要复杂的车辆动力学分析；因此，无论何种车型，它都可以以低成本广泛使用。