Surge Limit Prediction of Centrifugal Compressor Using Semi Classical Signal Analysis

This paper presents the use of the recently developed semi classical signal analysis (SCSA) method for surge limit detection of a centrifugal compressor. The SCSA is based on the fact that the studied temporal signal is considered as a trap-potential for a semi-classical particle and is represented by discrete energy levels given by the discrete spectrum of a Schrödinger operator. This approach provides new spectral parameters whom variability might be interesting for signals analysis. Inlet and outlet pressure signals recorded on a turbocharger test bench while reducing the compressor mass flow rate are analysed. The spectral parameters provided by the SCSA show interesting behaviour especially around the transition zone between stable operation and surge. Following these observations, a new criterion for surge limit detection is proposed. The sensitivity of the detection may be adjusted by the threshold value. The implementation of this new method combined with active regulation may extend the usable zone of the compressor map.