Data assimilation of flow-acoustic resonance

Area of Science:

• Aeroacoustics
• Computational Fluid Dynamics (CFD)
• Data Assimilation (DA)

Background:

• Internal aeroacoustic systems present simulation challenges, particularly in quantifying acoustic wave and shear layer vortex interactions.
• Accurate prediction of flow-acoustic resonant fields is crucial for understanding and mitigating noise in such systems.

Purpose of the Study:

• To develop and validate a data assimilation strategy for accurate prediction of flow-acoustic resonant fields.
• To improve the simulation of transfer loss between acoustic waves and shear layer vortices.
• To enhance the prediction of spatiotemporal evolution of shear layer vortices.

Main Methods:

• A data-assimilated momentum loss model, incorporating viscous and inertial loss terms, was embedded into Navier-Stokes equations.
• The ensemble Kalman filter was employed as the optimization algorithm, using acoustic pressure pulsations as observational data.
• A three-dimensional transient computational fluid dynamics method with an explicit algebraic Reynolds stress model (EARSM) served as the predictive system.

Main Results:

• The data assimilation strategy significantly reduced numerical errors in acoustic pressure pulsation frequencies and amplitudes.
• Optimal simulations achieved better agreement between time-averaged flow distributions and fluctuations.
• The numerical simulation successfully reproduced the complete spatiotemporal evolution of shear layer vortices.

Conclusions:

• Data assimilation provides an effective approach for accurate simulation of flow-acoustic resonant fields in channel-branch systems.
• The developed method enhances the predictive capability for complex aeroacoustic phenomena.
• Accurate simulation of vortex dynamics is crucial for understanding and controlling internal flow-acoustic resonances.