Study On Gear Fault Diagnosis Technology Based On The Method Of Transmission Error

As an important part in mechanical equipments,the gears are easy to break down in the gearbox because they often work under complex working conditions.Therefore,the fault diagnosis of gears is extremely significant to the operation of equipment.Aiming at the problems which includes low signal-to-noise ratio of traditional vibration signals and inconspicuous fault characteristics,a fault diagnosis method based on gear transmission error is proposed in the paper by referring to the error detection method of machine tool transmission chain.In order to verify the validity of this method,the analysis of transmission error theory as well as experiment are both carried on.The relationship between gear transmission error theory and fault,early fault characteristics analysis,signal detection and processing are studied in the paper.Through the influence of tooth surface fault on gear meshing form,the mathematical model of transmission error signal is established,and the features of gear early fault are analyzed.According to the requirement of transmission error signal acquisition,a single-chip microcomputer-based acquisition system is built through the existing fault simulation platform,and then the transmission error signal is obtained by measuring the angular displacement signal of two axles.The improved LMD decomposition method is used to process the collected signals.As a consequence,a series of PF components are obtained.Combined with the order spectrum analysis,the gear fault characteristics are extracted.Finally,PSO-BP neural network,as the identification system of different fault types,would divide the signal into different order segments,and select its energy value as the input vector to identify the gear fault types.Experiments show that the improved LMD decomposition method combined with order spectrum analysis is effective in fault feature extraction.PSO-BP neural network is more accurate for identifying different fault types of gears.The above methods provide a new thought of further research on gear fault diagnosis.