The Design Of Rotating Machinery Fault Diagnosis System Based On Ultrasonic

Rotating machinery plays an important role in the industrial production, detectionand fault diagnosis for the operating state of rotating machine can ensure the equipmentrunning safely,avoid suffered economic losses. Rotating machine’s fault often appearslightly friction or turbulence in the early stage, but ultrasonic sensor is the most sensitiveto the phenomenon.In view of this, this paper puts forward and designs a rotating machinefault detection and diagnosis system which was based on the ultrasonic.The ultrasonic fault diagnosis system is divided into fault diagnosis system based onanalog circuit and fault diagnosis system based on digital circuit. Firstly, according to thepositive relationship between the ultrasonic signal and the severity of friction, this paperdesign a LED display circuit, while this paper design an audio conversion circuit tomonitor the sound signal during operation of the device. And then combined with signalconditioning circuits, such as the pre-adjusted differential amplifier circuit, a high-passfilter circuit, a low-pass filter circuit and so on which constitute the fault diagnosis systembased on analog circuits. Then, combine with signal conditioning circuit, transistor circuitsand digital envelope detector system circuit to form a fault diagnosis system based ondigital circuit, which can display real-time waveform, histogram,calculate the timedomain parameters and feature extraction based on the histogram. Real-time waveformand histogram can intuitively monitor the machinery operational status and fault type, thetime domain parameters fault can provide a criterion for fault diagnosis. Finally, thedesign also add an SD card for storing real-time data, real-time waveform and histogram,it is easy to download to a PC to process the data analysis processing in order to furtheranalysis and study of the fault signal.At the end of this paper, the correctness of the time-domain waveform parametersdisplayed in this system is verified by using the signal generator and MATLAB software.And this paper use MATLAB to verify the fault data, the verification involves thatenvelope histogram can diagnose various types of faults effectively and intuitively,finally the circular statistical method is used for fault data analysis, it has proved the superiority on the fault diagnosis of rotating machinery by using circular statisticalmethod.