[Enrique Lin]

Hi everyone,

This problem has been troubling me for some weeks. I hope to find some suggestions or solutions. Thanks!

I am trying to simulate the noise generated by an NREL Phase VI wind turbine with FLUENT (ANSYS ver. 19R1).

Problem description:
The sound pressure level (SPL in dB) values over a wide range of frequencies are negative (please refer to the image). I am comparing my acoustic results to this paper:
A. Tadamasa and M. Zangeneh, 2011, "Numerical prediction of wind turbine noise," Renewable Energy, 36(7), 1902-1912.
DOI: https://doi.org/10.1016/j.renene.2010.11.036

Mesh Comparison v230606-1812 for CFDOnline.jpg


My settings:
The flow fluid geometry is similar to this paper but without the turbine tower and nacelle:
J.-O. Mo and Y.-H. Lee, 2011, “Numerical simulation for prediction of aerodynamic noise characteristics on a HAWT of NREL phase VI,” Journal of Mechanical Science and Technology, 25(5), 1341-1349.
DOI: https://doi.org/10.1007/s12206-011-0234-1

Turbulence model: k-omega SST
Acoustic model: FW-H
Noise source: turbine blade surface
Acoustic receiver: located downstream of the turbine at a distance of H+D/2 on the ground. Where H is the tower height, and D is the blade diameter.
Methods: SIMPLE. Other spatial discretization settings are all set to 2nd-order.
Transient Formulation: Bounded 2nd-order
Initialization: Hybrid
Time step size: 0.000333 sec
Number of time steps: 5000
The simulation was initially calculated in steady state. After the results converged, the simulation was converted to transient state. Then continue to calculate the aerodynamic performance and acoustic results.


What actions have I tried to resolve the issue:
1. Use finer mesh: the SPL values crept toward positive, but most are still negative. Overall, no significant changes to the results.
2. Used Polyhedra-Hex mesh (in FLUENT 20R2). Can only use "Frame Motion" in steady state. And the results are even worst.
3. smaller time step size: only changes the range of the calculated frequency (as mentioned by online sources and the FLUENT user manual.)
4. larger number of time steps: only increase the acoustic resolution.
5. Google for solutions and suggestions: the following are my findings (or lack thereof)
5-1. I checked a thread titled "Why negative decibels" on this website. No replies.
5-2. A post on ANSYS Official Forum. A reply mentioned "reference pressure" of 2e-5 However, this value cannot be changed, as this IS the reference value of my case.
https://forum.ansys.com/forums/topic...stic-function/


Thank you all for your time.