Performance Characterization System And Prediction Of Surface Layer Of Wind Turbine Bearing Material By Ultrasonic Rolling Extrusion

Wind power bearings play an important role in wind power generation equipment.The surface performance of a wind power bearing determines its reliability,and a better surface performance can significantly increase its service life.Ultrasonic roll extrusion strengthening process is a surface strengthening treatment technology based on plastic deformation.This process can effectively reduce the surface roughness of the wind power bearing,improve the degree of surface work hardening,and preset a certain residual compressive stress on the surface of the wind power bearing.So that the fatigue strength and wear resistance of wind power bearings can be improved.Therefore,it is of great practical application value to study the effect of ultrasonic roll extrusion technology on the surface properties of 42 Cr Mo which used in wind power bearing.During the ultrasonic roll extrusion experiment,the rotation speed,feed rate,amplitude and static pressure were picked as the process parameters.In addition,the surface roughness,residual stress and surface work hardening degree were used as the surface performance characterization parameters.The influence of each process parameter on the characterization parameters was explored through single-factor analysis.The results show that the surface roughness is positively correlated with the rotation speed and feed rate,and the surface roughness decreases first and then increases with the increase of static pressure.The effect of amplitude on the surface roughness is similar to the effect of static pressure.The residual stress decreases with the increase of the rotation speed and the feed rate,respectively.Also,the magnitude of residual stress is positively correlated with static pressure and amplitude.The degree of surface work hardening increases first and then decreases as the rotation speed increases,and decreases with the increase of feed rate.In addition,the degree of work hardening is positively correlated with ultrasonic amplitude and static pressure.Based on the test data of the orthogonal test and the qualitative analysis of single factor analysis,the AHP and GRA are used to establish the ultrasonic roll extrusion surface performance characterization system.Firstly,the AHP is used to obtain the subjective weight of each process parameter on each surface performance parameter.Then,the GRA method is used to obtain objective weights.Based on the principle of minimum information entropy,the comprehensive weight of process parameters on surface performance is obtained.Through calculation,the influence degree of each process parameter on each characterization parameter is obtained.Furthermore,the degree of influence of various process parameters on the surface properties is obtained.The BP neural network prediction model for the surface performance of ultrasonic roll extrusion is established based on orthogonal test data and surface performance characterization system.After calculating the error between the predicted value and the measured value,the average prediction error of each performance parameter is less than 5%.Therefore,the results show that the built prediction model has high prediction accuracy and can predict the surface integrity of ultrasonic roll extrusion.