Crack identification of compressor blades based on directivity factor decomposition model

The research on the acoustic characteristics is extremely important for the detection of blade crack faults in compressors. In this paper, aiming at the blade crack of compressor, a directivity factor decomposition method is established, and the influence of blade cracks on the noise characteristics is investigated. The noise propagation laws at different monitoring points under the monopole and dipole sound source mode were explored. It was discovered that the noise directivity characteristics of compressor possess stronger anti-interference ability, higher signal-to-noise ratio, and greater sensitivity to minor disturbances compared to the single-point amplitude. A directivity factor decomposition method targeting blade crack of rotating machinery was established. Three parameters were defined to assess the directivity characteristics of compressor noise, among which the performance of the directivity intensity factor was better. Employing the established method, the sensitivity of noise characteristics of the compressor to the size, shape, and position of blade cracks under different rotational speeds was explored. The conclusion is obtained that after the blade cracks occur, the directivity of the compressor inlet noise is stronger, and under the high and low rotational speeds, the influence law of the crack size on noise directivity is opposite. Under low speed, the noise of rectangular cracks possesses a stronger directivity. Under high speed, triangular cracks at the blade leading edge have a stronger directivity. The established method can excellently identify the crack of compressor blades and possesses a favorable anti-interference capability, offering a solid basis for the crack identification of blades in rotating machinery.