Numerical modeling and active noise control of impact road noise inside a vehicle compartment

Even though active noise control (ANC) technique has been widely investigated and proven its efficiency for low-frequency vehicle interior noise control, current applications are mainly for the steady noise process such as powertrain noise and uniform road noise. When impulsive road disturbances due to road bumps or potholes are present, they may hinder the performance of the ANC system. One of the reasons is that the prevalent ANC algorithm, namely the filtered- x least mean square (FxLMS) algorithm, has been developed assuming that the signals follow the normal distribution. The impact road noise due to a sudden impact may exhibit a non-Gaussian characteristic. Hence, the FxLMS algorithm may not be appropriate for controlling this type of impulsive noise. In this study, a robust ANC system configured with a modified FxLMS (MFxLMS) algorithm by incorporating thresholds on reference and error signal paths is proposed for impact road noise control. To aid in the control system design, the spectral-based substructuring technique is implemented to develop a coupled vehicle system model to simulate the impulsive interior acoustic response due to road unevenness. The vehicle passenger compartment is simplified as a 3-dimensional flexible-panel backed cavity model, which is coupled with the tire-wheel system modeled as a flexible ring element and rigid wheel through the suspension system represented by a spring-damper model. The tire-road interaction is modeled as a set of flexible elements with certain stiffness over a half-cosine wave road bump. Numerical simulation results show that the proposed ANC system can effectively deal with the interior impact road noise without instability issue that may occur in the existing FxLMS algorithm. In fact, results show that approximately 8 dB reductions are achieved at the driver's ear position.