THE BALANCE OF THE AIR FLOW NOISE IN EXTRACTING ACOUSTIC SIGNALS OF THE CRACKED BLADE

The blade fault is a very serious problem; recently, the acoustic method has achieved a great success in exploring the crack developing. Synchronization time averaging is a key technique, which was used to eliminate white noise; in the de-noise the effects of the load change is still a problem to be solved. The effect of the unstable flow will have some bad influence on the results averaged. In order to eliminate the bad influence, a symmetrical decomposition method has been discussed for de-noising in a symmetrical container in this paper. The main goal is trying to convert an unstable problem into a relative stable problem. In this method the noise signal of the symmetrical section has been decomposed at the symmetrical base, then the noise signal measured will be balanced with the symmetrical terms; the symmetrical terms are relatively stable, the unbalanced blade signal will be enhanced. The sensor position has been determined to get an optimized result, in this way the signal-to-noise ratio has been improved and the effects of the flow disturbance have been avoided.