13th European Conference on Turbomachinery Fluid dynamics & Thermodynamics
The timely detection of stall inception is of great significance for safe operation and stability control of axial compressor. In this paper, fast wavelet tool is selected to predict the stall precursor in the compressors with spike-type and modal-wave inception. Detailed measurements on the casing were conducted using a collection of time-resolved pressure transducers with circumferential and chord-wise spatial resolution. For modal-wave stall inception, low frequency reconstruction can detect the initial inception 100 rotor revolutions before stall through observing the increasing amplitude. For spike-type stall inception, low frequency reconstruction is unsuccessful to predict the inception. However, with the help of high frequency reconstruction, a specific frequency band of 0.2-0.8 blade passing frequency with increasing amplitude can be identified before spike-type inception. The fast wavelet method can predict two kinds of stall inceptions in advance and realize the early stall warning in axial compressor through a reasonable selection of the reconstructed frequency.