Prediction of supersonic fan noise generated by turbofan aircraft engines

Abstract

Prediction of Supersonic Fan Noise Generated by Turbofan Aircraft Engines was focussed on improving the capability of predicting supersonic fan noise from modern high-bypass-ratio turbofan aero-engines. The shift from single core jet engines to highbypass-ratio turbofan engines brought about a reduction in the overall aircraft engine noise principally by reducing the jet-broadband noise. However, this new design meant the size of the fan of a high-bypass-ratio turbofan engine, over subsequent years, has increased in diameter. This increase allowed for the speed of the tips of the fan blades to reach and exceed the speed of sound.

At high power engine operation conditions, especially at take-off conditions, the noise levels observed from such engines is very high. A major component of this noise is the supersonic fan noise which is also referred to as buzz-saw noise. Shocks are produced at the fan blade tips at this high power engine operation condition. These shocks propagate upstream, against the inflow, following a helical path dictated by the rotation of the fan. The pressure field produced at the tip of the fan is represented as a series of shock waves and expansion waves. As this pressure field advances, it interacts with the incoming flow and acoustic treatment in the intake duct.

The shocks in the pressure field are all unique and are of different amplitudes. This is because the fan blades, although manufactured to tight tolerances, are not perfectly alike. Also, the arrangement of these fan blades on the fan hub will also lead to unavoidable differences among the fan blades. These minute differences are reflected in the amplitudes of the shocks, making each shock slightly different from the others. Shocks in the pressure field propagate with respect to the magnitude of their pressure amplitude. Therefore, the shocks travel at different speeds. In the course of propagation, faster shocks catch up with slower ones, and they merge into a single shock, even as the shocks’ amplitudes are attenuated. The difference in speeds and the interactions among the shocks ensure a transfer of energy among the harmonics of the pressure field. This process is nonlinear; the work in this thesis is focussed on modelling the nonlinear propagation of the shocks pressure pattern.

These interactions greatly enhance the lower frequency harmonics of the pressure field shifting the dominance from the blade passage frequency and its harmonics. Further upstream, the dominance of the low frequency harmonics is unmistakable. Subsequently the pressure field is radiated from the aircraft intake duct. The resultant radiated pressure field is that which is perceived by an observer in the far-field. The models presented in this thesis capture the main features of this nonlinear propagation and radiation of the pressure field generated at the fan blade tips, and generates predictions for supersonic fan noise levels in the intake duct and in the far-field.

A time domain model named SPRID (Sawtooth Propagation in Rigid Intake Ducts) developed is presented. This model predicts the supersonic fan noise levels in ducts without any acoustic treatment, and has been validated against a benchmark frequency domain nonlinear propagation model (FDNS), and also measured data from a modelscale fan rig test provided by Rolls-Royce PLC. The need to incorporate the effect of acoustic liners in the modelling led to the development of a new model which employs the combined time-frequency domain approach. In this model, the nonlinear propagation of the pressure field is simulated in the time domain, while the acoustic liner effects are implemented in the frequency domain. This model also has been validated with measured data.

The combined time-frequency domain prediction method was improved to incorporate more complex features of supersonic fan noise propagation. Features such as the change in duct radius along the duct axis and the consequent change in mean flow speeds, and boundary layer effects on the liner absorption have been included in a more advanced model. The advanced nonlinear model is a more representative model of real aircraft intake duct. Also, a theoretical radiation model (GX-Munt) was utilized to predict supersonic fan noise in the far-field. In this thesis, a whole study of supersonic fan noise, starting from source generation at the fan plane up to the radiation to the farfield is presented. The thesis includes an extensive literature review, research on the generation of a source sawtooth for propagation utilizing measured data, and development of equations for nonlinear propagation in axisymmetric intake ducts. Results of the parametric studies using the advanced nonlinear propagation model reliably show all the effects of nonlinear distortion of the shock waves, variation in intake geometry, flow speeds, and variations in the acoustic liner absorption as a consequence of changes in boundary-layer thickness. Comparisons made against measured data, from modelscale fan rig tests conducted by Rolls-Royce PLC, show good and reasonable agreement. The advanced nonlinear propagation model achieves improved prediction capability for supersonic fan noise.

More information

Identifiers

ORCID for Alan McAlpine: orcid.org/0000-0003-4189-2167

Catalogue record

Export record