Active control for vehicle interior noise based on DWT-FxLMS algorithm using a piezoelectric feedback system

摘要

A piezoelectric feedback system for active noise control (ANC) of vehicle interior noise is built in this paper. Firstly, a mathematical vibration model of a typical vehicle-used plate with piezoelectric patch is derived for revealing piezoelectric mechanism. An improved filtered-x least mean square (FxLMS) algorithm based on the discrete wavelet transform (DWT), so-called DWT-FxLMS, is presented. Comparisons with the common-used algorithms suggest that, the DWT-FxLMS is superior to the time-domain FxLMS (TD-FxLMS) algorithm in terms of the computational complexity and the convergence speed. The DWT-FxLMS can cope with both the stationary and nonstationary noises, but the frequency-domain FxLMS (FD-FxLMS) algorithm is failed to eliminate the nonstationary noise. Secondly, based on a simplified vehicle cavity model, some numerical simulations are performed for its structural vibration and acoustic characteristics. The top plate with maximum acoustic contribution is selected as a target for active control. The arrangement of piezoelectric sensor and actuator on the top plate is determined following the D-optimization criterion. Finally, the experimental verifications are conducted by using the established piezoelectric feedback system. The results suggest that, after applying ANC, the vibration acceleration of the top plate and the sound pressure level (SPL) at the passenger's ear sides are attenuated to a certain extent. The proposed method based on the piezoelectric feedback DWT-FxLMS algorithm is effective and feasible for active control of vehicle interior noise, and can be regarded as a promising technique for developing the active noise and/or vibration control systems of other products with enclosed spaces in engineering. (C) 2020 Elsevier Ltd. All rights reserved.