Robust and Adaptive Feedback Noise Attenuation in Ducts

Abstract

Abstract-In this paper the attenuation of sound propagation in an air-handling duct using robust and adaptive feedback active noise control strategies is investigated. The case of multiple narrow band disturbances located in distinct frequency regions and the interference occurring in the presence of disturbances with very close frequencies are considered. The active control uses a loudspeaker as compensatory system. The objective is to minimize the residual noise at the end of the duct segment considered. The system does not use any additional sensor for getting in real time information upon the disturbances. A hierarchical feedback approach will be used for the control of the system. At the first level a robust linear controller will be designed taking advantage of the knowledge of the domains of variation of the frequencies of the noise disturbances. To further improve the performance, a direct adaptive control algorithm will be added. Its design is based on the use of the internal model principle combined with the Youla-Kučera parametrization of the controller. Both robust and adaptive control require the knowledge of the discrete-time model of the secondary path (the transfer function between the control loudspeaker and the residual noise measurement) which is obtained by identification from experimental data. Experimental results on a relevant duct active noise control test bench will illustrate the performance of the proposed methodology.

Algorithms for Adaptive Feedforward Noise Attenuation—A Unified Approach and Experimental Evaluation

“…The identification procedure follows closely the procedure described in [27], [28] The PRBS characteristics used in the identification process as excitation signal was: magnitude = 0.14 V, register length = 15, frequency divider of 1, sequence length: 2 15 − 1 = 32 767 samples, guaranteeing a uniform power spectrum up to 1250 Hz.…”

Adaptive rejection of narrow-band disturbances in the presence of plant uncertainties—A dual Youla–Kucera approach

“…Most of the proposed algorithms are based on the Youla-Kucera (YK) parametrization (or Q-parametrization) [2,20], that al-Email addresses: bernard.vau@ixblue.com (Bernard Vau), ioan-dore.landau@gipsa-lab.grenoble-inp.fr (Ioan Doré Landau). 1 For known disturbances subject to small variations of their characteristics a robust approach can be considered [1], [16].…”

“…The Q filter will be adapted in real time in order to cancel asymptotically the disturbance d. A successful adaptive rejection of disturbances resulting from an unstable exogenous system is mentioned in [22]. In the field of active noise control, first references seem to be [3], [4] and a more recent reference is [16]. In the field of active vibration control a first reference seems to be [11].…”

“…For all the experiments, a constant trace adaptation gain has been used. The magnitude of the trace was 30 x nb.of parameters 16. PSD evaluated on an horizon of 3s…”

Youla-Kucera adaptive feedback disturbance rejection in the presence of plant uncertainties

“…The first contribution suggesting to blend a robust controller with an adaptive structure established on the Q-filter seems to have been reported in [9], and a successful adaptive rejection of disturbances resulting from an unstable exogenous system is mentioned in [12]. In the field of active noise control, a first reference seems to be [2], [3] and a more recent reference is [8]. In the field of vibration control a first reference seems to be [6].…”