Research On Gear Grinding Burn Detection Based On Barkhausen Noise Signal

Gears are a key component of mechanical transmission.When the tooth surface of the gear is ground and burned,its microstructure changes accordingly.Macroscopically,burns,cracks and breaks will occur.Therefore,it is necessary to carry out safety inspection on basic components such as gears.The research on gear grinding burn has important theoretical significance and engineering application value for improving production safety,ensuring the quality of processed products and improving transmission efficiency.This thesis proposes a new gear grinding burn detection method,which can avoid the influence of excitation frequency on Barkhausen noise signal output,and has the ability to detect and characterize the hardness and microscopic metallographic structure of specimens.It can be applied to gear grinding.Nondestructive testing of burns.The main research contents include:(1)A burn specimen was prepared by a laser machining method for simulating a burn.Due to the laser processing method,the temperature of the gear tooth surface burn can be precisely controlled.By correlating the laser processing parameters with the degree of thermal damage,burn gears with different burn levels can be prepared.(2)Feature extraction of the Barkhausen noise signal.Starting from the domain theory,the mechanism of Barkhausen noise is described.Two different eigenvalue extraction methods of the butterfly curve and the hysteresis-like return line of the Barkhausen noise signal and the measurement method of the surface tangential magnetic field are studied.Combined with the hardware system,the corresponding signal system is developed.Automatic processing of data can be achieved.(3)Barkhausen noise detection method for burn gear hardness.The microhardness tester was used to detect the hardness value of the burned test piece,and the change law of the hardness increase of the Barkhausen noise signal was calculated one by one.The BP neural network and adaptive neural network were used to establish the prediction model of the hardness value of burn specimens.Finally,the qualitative detection of the burn degree of burn specimens was realized.(4)Experimental study on Barkhausen noise characterization of metallographic structure of burn specimens.Aiming at the metallographic structure change of the test piece caused by laser processing,the metallographic sample of the burned gear was prepared to carry out the Barkhausen noise detection experiment,and the relationship between the characteristic parameters of Barkhausen noise and the metallographic structure was quantitatively analyzed.The microstructure of the material was observed and analyzed by metallographic microscope,and the mechanism of the experimental phenomenon was tried.