Compressing units are subject to a potentially damaging phenomenon known as surge at low flow rates. This effect may be preceded by an effect called inlet recirculation – a flow reversal upstream of the impeller. An experimental compressor rig was the basis of this investigation. High frequency pressure measurements were taken at a number of flow conditions and locations within the compressor. The measured signals were processed using Singular Spectrum Analysis (SSA), a nonparametric technique. SSA decomposes a signal into a number of Reconstructed Components (RCs), from which trends and oscillatory components may be extracted. The frequency spectra of each RC and their relative contributions to signal reconstruction were examined and comparison was made with spectral maps in existing literature. This study aimed to investigate SSA’s ability to detect the presence of recirculation through analysis of a decomposed portion of a signal. The results demonstrated the potential of SSA to identify and extract oscillatory components with information about the local effect of inlet recirculation.
|Number of pages||10|
|Publication status||Published - 18 Oct 2018|
|Event||13th International Symposium On Compressor & Turbine Flow Systems - Theory & Application Areas - Lodz, Poland|
Duration: 18 Oct 2018 → 20 Oct 2018
Conference number: 13
|Conference||13th International Symposium On Compressor & Turbine Flow Systems - Theory & Application Areas|
|Abbreviated title||SYMKOM 2018|
|Period||18/10/18 → 20/10/18|
- singular spectrum analysis
- inlet recirculation
Logan, A., Garcia Cava, D., & Liskiewicz, G. (2018). Detection of inlet recirculation as an early indicator of proximity to instability in centrifugal compressors. Paper presented at 13th International Symposium On Compressor & Turbine Flow Systems - Theory & Application Areas, Lodz, Poland.