A method for the reconstruction of buzz-saw noise source and pressure field in transonic fan with multi-types of blade variations

Abstract

Buzz-saw noise is a main component of the noise measured in the forward-arc of transonic fans during the take-off and climbing of aircraft. Full resolving of this noise component is computationally expensive because it is very sensitive to random blade variations, such as manufacturing tolerance, deformations or defects. To avoid this, artificially-synthesized irregular buzz-saw noise source is widely used for the prediction and further study. However, the irregularities of the synthesized source are hardly linked to the blade geometry and the so-induced flow structures. In this study, it is found that blade stagger, axial and azimuth variations maintain a quasi-linear relation with the so-induced static pressure perturbations. Based on this quasi-linear relation, the disturbances induced by blade variations are represented by a set of acoustic modal perturbations generated by several pairs of “basic” rotors. By superposing such modal perturbations, the buzz-saw noise source and the static pressure field generated by a fan with multiple arbitrary blade variations can be achieved instantly. This method is programmed to develop a code package named the Mode-based Fast Reconstruction (MFR). It is validated from simple to complex application scenarios based NASA rotor 67. An accuracy level of 0.9 dB is achieved for the multiple pure tones generated by rotors with multiple blade variations that satisfy the quasi-linear assumptions. This enables high-accuracy fast reconstruction of buzz-saw noise source and pressure field in more real transonic fans. Additionally, the impact of blade variations on buzz-saw noise is much stronger than the influence on rotor aerodynamic performances. So, the effects of small blade variations should be considered during the design and optimization of transonic fan rotors, even though the aerodynamic performances may not benefit much from this.