Computational strategy for predicting CROR noise at low-speed Part I: review of the numerical methods

Abstract

This paper is the first part of a series of three papers dedicated to the prediction of lowspeed interaction noise of an isolated contra-rotating open rotor (CROR). The objective of this study is to assess available analytical and numerical methodologies in terms of robustness, performance and accuracy. The analytical procedure follows the classical Sears' theory, while the numerical procedure follows a two-step approach based on CFD coupled with the analogy of Ffowcs Williams-Hawkings (FWH). Concerning the numerical tools, both chorochronic (single-passage) and Chimera (full-annulus) approaches are considered. The prediction of interaction noise requires fulfilling three modelling steps: the aerodynamic perturbations, the rotor unsteady blade response and the resulting noise radiation. Comparisons with noise measurements are provided. It is shown that the results are highly sensitive to the description of the incoming wake deficit. The chorochronic approach, as a single blade passage is computed, allows the use of much finer meshes than the Chimera approach and requires a lower computational cost. This approach is therefore highly attractive for computing isolated axisymmetric configurations. Nomenclatureunsteady lift k X = chord-wise wave number M ri = relative Mach number across blade i mR = Rear rotor tones frequency, m×BPF (rear rotor) nF = Front rotor tones frequency, n×BPF (front rotor) nF+mR = Rotor-to-rotor tones frequency, n×BPF (front rotor) + m×BPF (rear rotor) RPM = Revolutions Per Minute R i = radius of rotor i S = Incompressible Sears function S c,hf = High-frequency compressible Sears function S c,lf = Low-frequency compressible Sears function SPL = Sound Pressure Level 2 U ri = relative velocity across blade i (X i , Y i ) = coordinate system attached to the blade row i α i = local stagger angle of the section i ß r = relative Mach number function Ω i = rotational speed of blade row i σ 2 = reduced frequency φ = reception angle

Pylon trailing edge blowing effects on the performance and noise production of a pusher propeller

“…The harmonics of the unsteady blade lift coefficient due to the change in angle of attack in the pylon wake region ∆c ∆α linst k are computed using the Sears function (Refs. 26,27):…”

Computational strategy for predicting CROR noise at low-speed Part II: investigation of the noise sources computation with the chorochronic method

“…The design, the experimental facility and the computational domain were described in the first part of this study 5 . Simulations are performed for a Mach number of 0.23 and no angle of attack.…”

Prediction of Contra-Rotating Open Rotor broadband noise in isolated and installed configurations

“…7dB is equivalent to an error of 125% on or 400% on the turbulent kinetic energy in the CFD. Considering the good results obtained in the tonal noise prediction 25,26 with those CFD computations, it is unlikely that such a large error exists, so CFD uncertainty could only explain a small part of the mismatch. A second significant input parameter is the wake half width, but for the same reason the error on this parameter is likely to be small.…”

“…This analysis has showed the difficulty associated with predicting the trailing edge noise. While there is a fairly good confidence in the CFD input data, as their use in predicting tonal noise showed excellent results 25,26 , the main uncertainties remain in the empirical models of the surface pressure fluctuation spectrum, convection velocity and spanwise integral length scales. It is intended to undertake more work on these topics to improve the prediction accuracy.…”

“…The CFD simulation shall be accurate enough to propagate the wakes from front rotor trailing edge down to the rear rotor leading edge, accounting for potential effects from the rear rotor. The solution retained here is the unsteady RANS phase-lagged approach (chorochronic) used by Colin 25,26 .…”