Aviation Engines Noise
Development of Physical and Mathematical Bases and Computer Programs for Simulation and Calculation of Aviation Engines Noise Generation, Propagation and Attenuation and Decrease of its Influence on Passengers, the Population and an Environment
Tech Area / Field
- SAT-AER/Aeronautics/Space, Aircraft and Surface Transportation
3 Approved without Funding
TsIAM (Aviation Motors), Russia, Moscow
- The Boeing Company, USA, WA, Seattle\nSnecma, France, Moissy Cramayel
Project summaryThe objectives of this project are to identify and to increase our understanding of the principal turbojet aerodynamic noise generation mechanisms and to seek practical ways to reduce this noise. The project will involve a series of coordinated experimental and theoretical studies of the fundamental interrelations between unsteady aerodynamics and the radiated acoustic field that is the main source of aerodynamic noise.
These investigations will be conducted in the following two general directions:
1. The first direction concerns a study of turbojet fan or compressor tone noise mechanisms and the spatial and temporal characteristics of inlet or exhaust acoustic radiation. In this part of the project, the main effort will be the development of two numerical simulations of unsteady rotor-stator-interactions. This involves, primarily, the development of effective numerical procedures that are based on a direct numerical integrating of the unsteady turbomachinery stage flow equations, together with an adequate description of the role of viscous vortical structures. An economical approach will first be developed that is based on a simplified simulation of turbulent rotor wakes using known semi-empirical relations obtained from steady state viscous calculations. This wake model will be coupled to an inviscid stage simulation based on the Euler equations. A more general calculation procedure will also be developed based on a numerical integration of the Reynolds-averaged Navier-Stokes equations and modern turbulent models. Numerical models will be developed for 2D and 3D problems and a comparative analysis of the different models will be conducted with respect to their capabilities for studying noise generation and reduction in turbomachinery stages.
Numerical procedures will be also developed for modeling the so-called fan multiple pure tones noise, which corresponds to high-speed flow between the blades and the appearance of shock-waves. Using the analytic methods of nonlinear acoustics and direct numerical integrating of the Euler equations, the 2D and 3D characteristics of this noise component will be studied, with particular emphasis on its dependence on casual blade row irregularities.
Within the limits of linear acoustics, an effective method will be developed for the numerical simulation of sound radiation from engine inlets and exhausts. High efficiency will be achieved using high order numerical approximations, integral relations for far field and effective solution procedure for the linear algebraic equation set for nodal acoustic potential values. This calculation method is intended for the calculation of the far field noise for different inlet or exhaust flow regimes and for the evaluation of the impact of duct acoustic treatment.
2. The second direction is an effort to improve the theory of noise radiation from three-dimensional turbulent jets and to develop new methods for jet noise suppression with an associated decrease in its negative impact on the environment.
The main efforts will be an experimental study of coaxial jet from by-pass nozzles. Four different experimental test facilities (4T-11, U-389, C17-A4, C5/6) will be used for this work. These facilities allow the testing of nozzles with diameters of from 30mm to 200mm, using full scale total pressures and total temperatures.. The main goals of our experiments work are:
- to study the vortex structure near chevrons in order to obtain a better understanding of noise suppression mechanism;
- to test several new jet noise suppressor concepts, namely: “aerodynamics” chevrons; asymmetric nozzles and combinations of different known designs.
- to investigate why small changes in chevron shapes, or small changes in the turbulence structure of the initial boundary layer near the nozzle exit, can produce noticeable changes in jet noise, these possible effects can lead to some uncertainty when we try to use model-scale nozzle noise measurements and to use it in order to predict real engine jet noise.
- to identify the noise generation mechanisms in hot jet and the influence of external streams (modeling take-off and angular cruise flight conditions, in particular, at flight Mach number up to 0.8 and at high-speed jet velocities).
The first goal involves finding out why strongly immersed chevrons generate excessive high-frequency noise and how this noise increase depends on the shape of the chevrons.
The second goal is the optimization of “aerodynamics” chevrons nozzle designs and other new noise suppressors design. In “aerodynamics” chevrons nozzles, air from the fan nozzle is injected into external stream, using inclines slots in the nozzle wall to form a series of micro-jets. These micro-jets interact with each other to produce an effect similar to that of metal chevrons. There are many variables involved in the design of such nozzles (relative thickness of slots, distance between its, angle relative stream wise and so on). It is believed that an optimum design of such “chevrons” will not lead to an increase in high-frequency noise. In our previous work, it was found that a small angular rotation of the fan nozzle can reduce jet noise. There are also several publications which report noise reductions for larger nozzle deformations (Papamoschou, Viswanathan and other). Our goal is to study different combinations of these concepts in order to find an optimum design.
The third goal is the most unclear and complex of these goals. These effects, however, are very important for correct modeling of jet noise and we are forced to study then.
Finally the fourth goal is more traditional, with many scientists drawing attention to the scatter in the noise data for hot jets and for the effects of external streams on the jet noise levels and spectra shapes.
In addition to the four experimental tasks mentioned above, the following problems in jet noise theory will also be studied:
5. the development and improvement of direct numerical simulation methods (LES/DES) of turbulence in coaxial nozzle and jets, and far jet noise prediction using Kirchhoff’s integral method;
6. the investigation of acoustic ray tracing through the large-scale non-uniformities inside turbulent coaxial jets, the clarification of refraction an convection effects and wave length changes due to these interaction;
7. the development of new simplified linear aeroacoustic wave equation, based on short-wave approximation, and with usage of the refraction and convection acoustic wave effects on turbulent velocity and temperature large-scale heterogeneities, the development of simple numerical method of its solution;
8. the development of a simplified engineering calculation method for jet noise prediction, based on approximate analytical relations for the spectral density of noise radiation per unit volume, combined with the numerical solution of the flow equations using an improved turbulence model;
9. validation of the resulting prediction methods using know and recently obtained experimental data for round and coaxial jets.
Although the LES/DES methods mentioned in the point 5) have the best long term potentially for the prediction of jet noise, accurate noise modeling of the flow close to the nozzle exit requires the direct numerical simulation of not only the jet itself, but also of the flow inside and around the nozzle. Therefore, such an approach requires very powerful computers (much more powerful than modern computers). In addition, this method has some fundamental limitations for the prediction of high frequency noise. Therefore, we plan to use such methods mainly to obtain a clearer understanding of the physical features of complex turbulent jet flows.
As it was shown earlier (including our own work) short-wave length acoustic waves are very strongly refracted by turbulence non-uniformities (see point 6). As a consequence, real high velocity, heated, turbulent jets can be, in a certain sense, “non-transparent” for short-wave acoustic waves. This effect complicates noise prediction, in particular it can induce azimuthal noise asymmetry. This and similar effects will also be study in our project.
In this project we propose to pursue a “classical” aeroacoustic approach (point 7) using wave equations of the type proposed by Lighthill or Lilley, but using a short-wave approximation and including refraction effects due to turbulence. The main efforts will be directed to obtaining a rather simple numerical (or approximate analytical) solution for the Green function of the wave equation.
Finally (point 8), a simple engineering method for the prediction of jet noise for real coaxial jet will be developed. This method is based on RANS calculations and on relatively “simple” relations for the spectral density of noise radiation from unit of jet volume. This method will allow the prediction of noise spectra and directivity diagrams.