基于延时陷波算法的车内主动噪声控制

传统车内主动噪声控制(Active Noise Control,ANC)中使用的FxLMS(Filtered-x Least Mean Square)算法，计算量较大，所需硬件成本相对较高，不利于批量生产。因此文章提出一种使用参考信号与滤波参考信号数字合成的改进自适应陷波方案，弥补了次级通路延时模型在相位误差方面的缺陷，同时具有结构简单、计算量小的优点。最后，基于CompactRIO建立了车内双通道延时陷波窄带ANC系统，进行车辆定置发动机定转速与行驶过程急加速两种工况下的实车道路试验。试验结果表明，所提方案的主动噪声控制效果良好，急加速工况下，车内噪声总声压级最高仍可降低8.46 dB(A)，为主动噪声控制的工程应用提供了参考。

Active vehicle interior noise control based on delayed notch algorithm

The filtered-x least mean square (FxLMS) algorithm used in the traditional active noise control (ANC) for vehicle interior noise requires relative high-cost hardware due to large calculation amount, which is impractical to mass production. Therefore an improvement scheme using direct digital synthesizer to generate reference signal and filtered reference signal is proposed, which makes up for the deficiency in phase errors of a secondary path delay model, and has the advantages of simple structure and less calculation amount. Finally, based on CompactRIO, a narrowband ANC system with two-channel delayed notch in vehicle is established and the road tests are carried out in two conditions of vehicle stationary with a constant engine speed and rapidly accelerating in vehicle driving. The experiment shows that vehicle interior noise is reduced effectively by the scheme proposed, the A-weighted sound pressure level (SPL) of the overall noise is reduced by 8.46 dB(A) in rapidly accelerating condition, which provides a reference of engineering applications for active noise control.