路面识别技术在路噪主动控制降噪性能优化中的应用

Application of road surface recognition technology in the optimization of noise reduction performance of active road noise control

为了提高路噪主动控制系统对不同类型路面激励的适应性,采用卷积神经网络方法分类不同路面,根据不同路面输入下路噪主动控制系统的输出响应来调节收敛因子,优化路噪主动控制系统的降噪性能,使车辆在不同路面下达到最优降噪效果。通过采集3种典型路面图像,采用一种迁移学习的方法训练和测试网络,根据路面识别结果,调整路噪主动控制系统中LMS算法的收敛因子。仿真结果表明:利用VGG16对路面进行识别,可以准确识别不同路面类型。基于路面识别和LMS算法的路噪主动控制系统可以根据不同的路面类型调节收敛因子,提高车辆的降噪效果。

In order to improve the adaptability of the road noise active control system to different types of road excitations,the method of convolutional neural network is used to classify different roads,and the convergence factor is adjusted according to the output response of the road noise active control system under different road input.The noise reduction performance of the active noise control system enables the vehicle to achieve the optimal noise reduction effect on different roads.By collecting three typical road images,a transfer learning method is used to train and test the network,according to the road surface recognition results,the convergence factor of the LMS algorithm in the road noise active control system is adjusted.The simulation results show that the use of VGG16 to identify road surfaces can accurately identify different road surface types.The road noise active control system based on road recognition and LMS algorithm can adjust the convergence factor according to different road types,thereby effectively improving the noise reduction effect of the vehicle.