This study uses a statistical analysis on given pressure signals to identify the surge condition in radial compressors, providing insight into their dynamics and performance. Specifically, the investigation focuses on a two-stage centrifugal compressor used in refrigeration applications. Through Computational Fluid Dynamics (CFD) Unsteady Reynolds-Averaged Navier-Stokes (URANS) simulations, static pressure time-series are collected to cover four distinct operational points of the machine, ranging from a stable condition to an unstable one. To assess the stability of the compressor operation, the analysis introduces the Approximate Entropy (ApEn) as a statistical parameter to measure the regularity and predictability of the time-series signals. The results show that the proposed methodology is effective in pattern recognition, able to identify sources of instabilities, and to distinguish between different operating conditions. This approach enhances the comprehension of the operational characteristics of centrifugal compressors and sets the stage for improved monitoring and diagnostic methodologies. ApEn can therefore be a useful index to analyse time-series from sensors easy to install in real compressor configurations. The proposed methodology is included in a set of models, based on machine learning techniques, previously developed by the authors, to identify compressor instability during operation.

The use of Approximate Entropy analysis for flow pattern identification in radial compressors to detect instable operating conditions

Carrattieri, Lorenzo;Cravero, Carlo;Marsano, Davide;Valenti, Emiliano
2024-01-01

Abstract

This study uses a statistical analysis on given pressure signals to identify the surge condition in radial compressors, providing insight into their dynamics and performance. Specifically, the investigation focuses on a two-stage centrifugal compressor used in refrigeration applications. Through Computational Fluid Dynamics (CFD) Unsteady Reynolds-Averaged Navier-Stokes (URANS) simulations, static pressure time-series are collected to cover four distinct operational points of the machine, ranging from a stable condition to an unstable one. To assess the stability of the compressor operation, the analysis introduces the Approximate Entropy (ApEn) as a statistical parameter to measure the regularity and predictability of the time-series signals. The results show that the proposed methodology is effective in pattern recognition, able to identify sources of instabilities, and to distinguish between different operating conditions. This approach enhances the comprehension of the operational characteristics of centrifugal compressors and sets the stage for improved monitoring and diagnostic methodologies. ApEn can therefore be a useful index to analyse time-series from sensors easy to install in real compressor configurations. The proposed methodology is included in a set of models, based on machine learning techniques, previously developed by the authors, to identify compressor instability during operation.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11567/1225356
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact