Research On Gear Fault Diagnosis And Method Based On Fluid-Thermal-Solid Coupling And Dynamics Analysis

Among traditional mechanical transmission mechanisms,the gear transmission mechanism occupies an important position and is widely used in various fields and various types of equipment.The efficient and stable operation of gear components is the basis for the regular operation of equipment,and the study of the transmission characteristics of gears in case of failure is of great significance for their fault diagnosis and early warning.This paper takes cylindrical spur gears as the research object,simulates different faults such as active wheel wear,driven wheel tooth breakage and pitting,and studies gears’ mechanical characteristics and vibration characteristics under single and mixed faults,respectively.The main research contents are as follows:(1)To address the problem of lubrication and thermal coupling during gear transmission that cannot be accurately calculated by traditional CFD technology,a fluid-thermal-solid coupling method based on the combination of coupled Euler-Lagrange(CEL)and thermal coupling method is proposed.The three-dimensional finite element model of gearbox was established,the mechanical properties of different faulty gears in the transmission process were analyzed,the variation law of the stress field under each fault was obtained,and the influence of stress field change under different faults on the stability of gear transmission was compared and analyzed.The research results show that:the wear fault has the most significant influence on the gear,the broken tooth fault has a relatively small influence on the gear,and the pitting fault has a minor influence on the gear;the effect of the mixed fault on the gear is much more significant than that of the single fault.(2)The fatigue crack expansion characteristics of gears were studied based on the extended finite element method(XFEM),and the crack expansion laws under different crack initial positions and different loading positions were investigated,respectively.The results show that the cracks produced by the gears are mainly of type KⅠ.There are large differences in the crack expansion path and the stress intensity factor distribution trend under different conditions.(3)A rigid-flexible coupling dynamics model was established to extract the gear vibration signal with acceleration as the carrier.The gear transmission system experimental bench was built,and the gear vibration signals were collected.The time domain and frequency domain characteristics of the simulated and experimental signals of the dynamics model were compared and analyzed to verify the accuracy of the dynamics model.(4)The simulated signal is subjected to singular value decomposition and generalized linear FM wavelet transform to extract signal features with high time-frequency aggregation,reduce noise’s influence and improve the signal source’s quality.Then the unbiased autocorrelation analysis is performed on the noise reduction signal based on Autogram theory to extract the characteristic frequencies and frequencies at multiples.The smoothed support vector machine model is used to classify the frequency features to achieve gear fault diagnosis and identify gear fault states.