In this paper, planar and the cylindrical broadband non-uniform acoustic absorbers were constructed, both of which use broadband absorption units (BAUs) as their building blocks. The impedance boundary Navier–Stokes equation (IBNSE) method was developed to predict the absorption characteristics of the lined duct with non-uniform acoustic absorbers, in which each small piece of perforated plate is acoustically equivalent to a semi-empirical impedance model through the boundary condition. A total of four semi-empirical impedance models were compared under different control parameters. The full Navier–Stokes equation (FNSE) method was used to verify the accuracy of these impedance models. It was found that the IBNSE method with the Goodrich model had the highest prediction accuracy. Finally, the planar and the cylindrical non-uniform acoustic absorbers were constructed through spatial extensions of the BAU. The transmission losses and the absorption coefficients of the rectangular duct–planar acoustic absorber (RDPAA) and annular duct–cylindrical acoustic absorber (ADCAA) systems under grazing flow were predicted, respectively. The results demonstrated that the broadband absorption of the designed non-uniform acoustic absorbers was achieved. The developed IBNSE method with Goodrich model was accurate and computationally efficient, and can be used to predict the absorption characteristics of an acoustically treated duct in the presence of grazing flow.

In this paper, a new fractal-based broadband metasurface absorber is developed, which is based on the excellent space-filling ability of the Hilbert fractal. Each unit cell in absorber consists of a micro perforated panel (MPP) and a coplanar coiled cavity. In order to gain a deep insight into the sound absorption mechanism and perform a rapid design, a theoretical model for analyzing the sound absorption characteristics of the unit cell is established, in which the Fok function is used to account for the coupling effect between holes. Afterward, the absorption mechanism as well as the effects of parameter variations on absorption characteristics are investigated. To increase the space utilization, each unit cell is arranged in space according to the Hilbert fractal curve. Consequently, a metasurface absorber with 6 detuned unit cells is constructed. The multiple resonant cavities with dissimilar lengths can provide peak absorptions at multiple frequencies, thereby broadening the attenuation frequency range. Finally, the absorption performance of the designed absorber is obtained by theoretical calculations, finite element (FEM) simulations and experimental measurements, respectively. The experimental results show that a continuous absorption spectrum is achieved in the range of 855–1359 Hz with absorption coefficient above 0.8 under a deeply sub-wavelength thickness (22.2 mm). This study provides an effective way for the design of a space-limited broadband absorber. With the advantages of ultrathin thickness, broadband, and compactness, the developed fractal-based metasurface absorber has great potential in the field of noise reduction.

This work proposes a noise-reduction structure that integrate phase-modulating metasurface (PMM) with acoustic liners (ALs) to enhance the absorption performance of a duct with relatively small length-diameter ratio. The PMM manipulate the wavefront by introducing different transmission phase shifts, so that the spinning wave within the duct is generated. Compared with the plane wave, the generated spinning wave has a lower group velocity, which results in a greater traveling distance over the ALs in the duct. The optimization design is performed to determine the final structural parameters of the PMM, which is based on the predictions of the amplitude and phase shift of the acoustic wave at the outlet of the PMM using the theory of metascreen-based acoustic passive phased array. With the manipulation of the PMM, the incident plane wave is modulated into a spinning wave, and then enters into the ALD, whose structural parameters are optimized by maximizing the transmission loss using the mode-matching technique. Finally, the noise-reduction performance of this combined structure is evaluated by numerical simulations. The results demonstrate that, compared with the traditional ALD, the proposed structure exhibits a 83.8% increase in transmission loss, and hence the noise-reduction performance is significantly improved.