Method for fault diagnosis of large rotating machinery

The method of fault diagnosis for large rotating machinery and the design of its fault diagnosis system will use the spindle device of the mine hoist as the monitoring object, and use the vibration signal sensor to monitor the working state of the spindle device and collect the corresponding vibration signal data. The vibration signal collected by the frequency domain analysis and the RBF neural network are combined to identify the vibration signal of the hoist and determine the current working state of the hoist. The online monitoring and fault diagnosis system of the basic mine hoist spindle device. Simulation signal analysis The fault simulation test of the mine hoist spindle device is difficult to implement, and the fault signal of the hoist spindle device cannot be collected. Therefore, the corresponding fault simulation test is completed on the rotating machinery fault diagnosis simulation test rig.

The motor drives the main shaft through a coupling. The two ends of the main shaft are equipped with rolling bearings. The middle part of the main shaft is equipped with a rotor, and the main shaft drives the gear box through the V belt. The structure of the test rig is similar to that of the mine hoist spindle, so its vibration signal can be used to study the fault diagnosis of the hoist spindle device.

During the test, the vibration signal of the bearing housing is collected by the vibration sensor mounted on the bearing housing; the different bearing faults are simulated by replacing the bearing, and the other parts are unchanged to ensure the same working condition. The replaced bearings have been pre-machined with corresponding faults, including outer ring faults, inner ring faults, rolling element faults and hybrid faults. The training accuracy error curve and the actual output test result of the constructed RBF neural network are compared with the expected output result. The actual output value of the established RBF neural network classifier has a certain error with respect to the theoretical value, but the error is within the acceptable range. . (Finish)

Sisal & Jute Twine

Sisal & Jute Twine