Dynamic Frequency Reliability-based Sensitivity Analysis And Robust Optimization Design Of High Speed Motorized Spindle

The motorized spindle is one of the crucial units of the high-speed CNC machine tool.The performance of the motorized spindle not only directly determines the precision of machining and the quality of processing surface,but also has a significant effect on the reliability,stability and production efficiency of the whole machine tools to some extent.Generally,the working speed of the motorized spindle is extremely high.When the speed gets close to the critical speed of the motorized spindle system,the resonance phenomenon will happen.For one high-speed motorized spindle equipped with angular contact ball bearings,the dynamic performance of high-speed angular contact ball bearings has a great influence on the performance of the motorized spindle.The dynamic performance of one high-speed angular contact ball bearing will vary with the speed of the spindle.Furthermore,there exits some certain randomness associated with geometrical dimensions and material attributes of the spindle.These factors could increase the possibility of motorized spindle resonance.Then,based on the dynamic performance research of high-speed angular contact ball bearing,it is necessary to analysis the natural frequency of the motorized spindle,study the frequency reliability and sensitivity,and carry on reliability-based robust design.In this paper,details of studies are as follows:(1)Based on the Hertz contact theory and Ferrule-Control theory,the de Mul's quasi-static analytical model of the angular contact ball bearings is established and developed to calculate the bearing stiffness.The static and dynamic performance parameters including the bearing stiffness are obtained through programming in MATLAB.Then the factors which affect the bearing stiffness are studied.(2)The structure of motorized spindle is simplified,and then the spindle model combined with the bearing model is confirmed as the research object.Afterwards,the parameterized model is built in ANSYS software.Meantime,the natural frequency of the motorized spindle is studied through modal analysis.(3)With the iSIGHT platform integrated into ANSYS,several parameters of the motorized spindle system are selected out as design variables by means of DOE(Design Of Experiments).Taking these design variables into consideration,sufficient samples are obtained to establish one database.On basis of the database,the BP neural networks are constructed to fit the function between the minmum natural frequency and the design variables.The reliability function between the natural frequency and critical frequency is built.Considering the influence of bearing stiffness changing along with the spindle rotating,the frequency reliability and sensitivity of the motorized spindle are calculated with first order second moment method.The results are verified by Monte Carlo method.(4)Based on the theory of reliability-based robust design,the spindle-bearing system under one certain speed is regarded as the research object.By taking the quality of spindle and the frequency sensitivity of the motorized spindle as the objective function,the design variables which obviously affect the natural frequency are optimized.Based on the optimal structure parameters,the frequency sensitivity and quality of spindle are both minimum.Finally,the goal that the motorized spindle system becomes more robust is achieved through the robust design.