Stall-Warning Approach Based on Aeroacoustic Principle

Abstract

In a fan/compressor, the pass of the blade is a periodic event and this periodicity deteriorates when rotating stall is gradually approached. The present work studies this periodicity of the pressure signal and explains when and why this periodicity deterioration happens from the perspective of the aeroacoustic principle. A theoretical analysis shows that the change of blade circulation has a close association with rotating stall and can be reflected on the periodicity deterioration of pressure signals sampled by a near-field dynamic pressure sensor located on the casing wall. A defined parameter , which can revalue this periodicity, has the potential to be a stall-warning criterion. Then, a stall-warning approach is developed in terms of this feature. The values of are calculated by the pressure signals of several blade pitches based on the current and the previous shaft periods. Statistical estimations are conducted on within statistical windows by calculating the probabilities of less than a threshold. The relevant experiments are carried out on a low-speed compressor, a transonic compressor, and a low-speed fan. The results show that the proposed approach has the capability of generating a stall-warning signal sufficiently in advance of the modal wave and spike in turbomachinery.