TL;DR
This paper investigates the aerodynamic noise generated by an axial fan as it transitions from stable operation to rotating stall. Using advanced numerical simulations, the researchers reveal how the fan's acoustic signature changes dramatically during stall inception and development, identifying specific low-frequency noise components that emerge. This work is crucial for predicting and mitigating noise issues in critical industrial applications like thermal power plants.
What Problem Does This Paper Address?
Axial fans are vital components in thermal power plants, responsible for supplying combustion air. However, they are susceptible to flow instabilities like rotating stall, which can lead to significant operational issues, including increased vibration and noise. While the aerodynamic noise characteristics of fans under stable operating conditions are relatively well-understood, the acoustic behavior during the complex and dynamic process of stall inception and development has been less explored. This paper aims to bridge this gap by numerically simulating and analyzing the aerodynamic noise of an axial fan as it enters and progresses through rotating stall, providing insights into the underlying mechanisms and identifying key acoustic features.
Methods
The researchers employed a sophisticated numerical approach combining Large Eddy Simulation (LES) for fluid dynamics and the Ffowcs Williams-Hawkings (FW-H) acoustic analogy for noise prediction. LES is a powerful technique for capturing turbulent flow structures, which are critical for generating aerodynamic noise, especially under unsteady conditions like stall. The FW-H model allows for the calculation of sound generated by turbulent flows based on the simulated flow field.
To induce and study rotating stall, a throttle valve model was incorporated into the simulation, allowing for controlled reduction of the outlet pressure. The study analyzed the fan's performance under four distinct conditions: design condition, near stall inception, stall inception, and mature rotating stall. Time-domain analysis of acoustic pressure and its peak values, along with frequency-domain analysis using Fast Fourier Transform (FFT), were used to characterize the noise.
Datasets
Not discussed in the paper. The study relies on numerical simulations rather than experimental datasets for training or evaluation.
Key Insights and Ideas
- Stall Induces Distinct Acoustic Signatures: The transition to rotating stall dramatically alters the fan's acoustic output, moving from a predictable, periodic noise to a more complex and fluctuating soundscape.
- Emergence of Low-Frequency Noise: During stall inception, specific, discrete low-frequency noise components appear, which are absent during stable operation. These are identified as key indicators of the onset of stall.
- Correlation Between Flow and Acoustics: The study clearly links changes in the flow field (e.g., vortex formation, relative velocity increases) directly to observable changes in acoustic pressure and its peak values.
- Stall Evolution Mirrored in Noise: The progression from stall inception to a mature stall cell is accompanied by increasingly severe fluctuations in acoustic pressure and amplitude, indicating a deteriorating flow condition.
Key Results
- Design Condition: Under stable operation, the acoustic pressure exhibits regular periodicity, with noise energy concentrated at the fundamental frequency and its harmonics.
- Stall Inception: As the fan approaches stall, acoustic pressure amplitude begins to fluctuate, and discrete low-frequency components emerge. Specifically, three prominent frequencies were identified: 27.66 Hz, 46.10 Hz, and 64.55 Hz.
- Rotating Stall: In the mature rotating stall condition, acoustic pressure fluctuations become significantly more pronounced. The acoustic pressure peak becomes unstable, with a notable sudden increase observed at the 34.5th revolution, coinciding with the first sudden increase in relative velocity.
- Flow Field Dynamics: The simulation revealed the formation and development of stall cells, starting with a single vortex and evolving into four distinct, circumferentially uniform stall cells over seven revolutions.
Limitations and Future Work
- Numerical Model Limitations: While LES and FW-H are powerful, they are still models and may have inherent limitations in fully capturing all complex aeroacoustic phenomena. The paper does not explicitly detail these limitations.
- Validation: The paper presents numerical results. Experimental validation of these findings would strengthen the conclusions.
- Generalizability: The study focuses on a single axial fan design. Further research could explore how these acoustic characteristics vary across different fan geometries and operating parameters.
- Noise Mitigation: While the paper identifies stall-related noise characteristics, it does not propose specific noise reduction strategies. Future work could focus on developing active or passive control methods to mitigate this noise.
- A-Weighted Spectrum: The paper utilizes A-weighted sound pressure levels to better reflect human perception of noise, highlighting that the fluctuation of acoustic pressure is more intense in the low-frequency range after stall inception. This suggests that stall-induced noise could be particularly bothersome to humans.
Summary updated