Research On The Location Of Gearbox Fault Diagnosis Measuring Point

Gear failure is an important cause of machine failure.Therefore,fault diagnosis of gears is of great significance for reducing equipment maintenance costs and preventing sudden accidents.Gear fault prediction and fault diagnosis based on vibration signals is the most widely used method.The vibration acceleration sensor in the gearbox installation position will directly affect the sensitivity of the vibration signal to the gear failure.This paper studies the impact of vibration sensor installation location on fault diagnosis.The main contents are as follows:(1)According to the acquisition of the original vibration information in the gearbox fault diagnosis,it is proposed that the vibration signal acquired on the gearbox using the acceleration sensor should be placed at the position of the measuring point that is optimal for the vibration signal.And by accumulating the sensor in three-dimensional arrangement(normal position,axial position and bearing position),the acceleration sensor can collect the vibration data in three directions of X,Y and Z,and carry out the data collected by the sensors of three different positions.Analyze the pair to get the location that is most sensitive to the fault information.(2)Using the Time synchronous averaged signal method to preprocess the vibration signals collected by the three measurement points,and then select the appropriate fault characteristic parameter indicators to perform the entire life cycle of the vibration information after the noise is completed.Comparative analysis,by comparing and comparing the timeliness and accuracy of different measurement points in different periods,the optimal measurement point position of the acceleration sensor is obtained.After comparative analysis,it was found that the position most sensitive to the vibration signal on the gearbox should be the position of the bearing measurement point.(3)For the obtained sensitive measurement point position,supplemental verification is performed by using a fully connected multilayer sensor.The fault vibration signal and the normal bearing signal are collected at the bearing position and directly input into the network model under the pre-processing of any different signals.Observe the accuracy of the fault classification of the multilayer sensor output to verify the position of the bearing.The vibration data signal does have a better signal.