Hob wear state condition monitoring based on statistical distribution law

The evaluation of the wear state of the hob during gear machining can effectively optimize the tool change strategy, which is of great significance to the improvement of machining efficiency. Aiming at the wet ma-chining conditions for large gears, a method of hob wear monitoring called 'Drift Prediction of Gamma Distribution Parameter (DP-GDP)' is proposed based on time domain impulses statistical distribution. Firstly, a vibration signal model of the hob is established based on the combined action of periodic cutting force ex-citation and random impact excitation. The vibration parameters such as damping and stiffness are discussed. Then, the Poisson process mechanism of random collision between hob and material particles during hob cutting is taken into consideration. The relationship between hob wear stage and occurrence of impulses excited by material particles collision is modeled under over-damping condition. The gamma distribution mechanism of the time required for impulses generation is elucidated. Furthermore, a wear feature extraction method based on the statistical parameters of random impulses is proposed. The method focuses on single impulse identi-fication and description of impulses occurrence probability distribution in time domain signal. Based on this, a prediction curve that reflects the wear state can be got for hob life degeneration assessment. Finally, a full-life test of hobs during the hob cutting process in the industrial field is completed. The data of hob whole life in different wear states are acquired. The simulation and experimental results show that the proposed method can effectively predict the whole life wear process of the hob. The extracted features can accurately map the hob wear state. Compared with traditional methods such as RMS and wavelet, the proposed method has better validity and accuracy under specific processing conditions.(c) 2023 CIRP.

Automated summary

The evaluation of hob wear during gear machining is crucial for optimizing tool change strategy and improving machining efficiency. This study proposes a monitoring method, called 'Drift Prediction of Gamma Distribution Parameter (DP-GDP)', based on time domain impulses statistical distribution. The method establishes a vibration signal model of the hob considering periodic cutting force and random impact excitation, discusses vibration parameters, and models the relationship between hob wear stage and occurrence of impulses. It also proposes a wear feature extraction method based on statistical parameters of random impulses, and successfully predicts the hob's wear process in industrial tests.