Array acoustic detection method for train bearing defect online monitoring

The invention discloses an array acoustic detection method for train bearing defect online monitoring, and is applied to detection of various train axle box bearings. According to the method, acoustic sensor arrays are arranged at the two sides of a track for acquiring bearing sound data, wheel sensors are installed on the track for determining speeds and positions of wheelsets, and according to the positions of the wheelsets, wave beam formation is performed on array sound data so as to improve the signal-to-noise ratio of the bearing sound data. Each acoustic sensor array comprises at least two omni-directional microphone sensors, the frequency response scope of each microphone sensor is 50 to15KHz, the center interval between adjacent microphones is smaller than one half of a received wavelength, and the installation heights of the microphones are consistent with bearing heights. The at least two wheel sensors are installed on the track all together, the sensors are symmetrically distributed within a +/-3 meter scope of the center line of each microphone array, the distance between each sensor and the center line of each microphone array can be changed, yet the distances have to be accurately measured so as to satisfy needs of the arrays for signal processing.