Rapid detection of rotating stall in axial flow compressors on computer cluster

Abstract

Early detection of rotating stall is of great significance to improve the performance of the compressor. In this paper, based on the fault detection decision scheme, early detection of rotating stall is realized via dynamical pattern recognition on computer cluster. Firstly, based on deterministic learning theory, stall detection scheme used for early detection of rotating stall is introduced. Secondly, a PC cluster treated as the computing platform is constructed in detail. Finally, an experiment in an axial compressor through dynamic recognition is included to demonstrate the effectiveness of the detection scheme.