Font Size: a A A

Study On The Key Technologies Of Slow Tool Servo Turning For Machining Complex Optical Surfaces

Posted on:2019-04-05Degree:DoctorType:Dissertation
Country:ChinaCandidate:J YangFull Text:PDF
GTID:1362330602968611Subject:Agricultural mechanization project
Abstract/Summary:PDF Full Text Request
Compared with traditional optical parts,the optical parts with complex surfaces can not only reduce the number of components in the optical path,simplify the optical path structure,but also can improve the imaging quality and reduce the size and weight of the optical system.With the rapid development of science,technology and social economy,the optical parts with complex surfaces now are widely used in many different fields,and the demand for them is growing.Moreover,it is one of the trends guiding the development of optical parts.Due to the high machining accuracy and processing efficiency,the slow tool servo(STS)machining technology has become one of the main methods for machining optical parts with complex surface.Compared with the large number of mature industrial applications abroad,the development in this field in China is relatively backward,and the demand for machine tools from abroad is still large.With the purpose of STS machining of optical part with complex surface,the key technologies including tool path planning,finite element simulation and surface quality prediction are studied.1.Tool path planning for slow tool servo turning of optical parts with complex surfaceAs the key technology of slow tool servo turning,the performance of tool path directly affects the surface quality of machined workpiece.In this thesis,the tool path planning for slow tool servo turning is optimized and then used for the machining of parts with complex surface.Firstly,the relative coordinate systems used in tool path planning are defined and the relationships among them are given.Then,the tool types and tool interference are analyzed.Tool path planning includes discretization of cutting contact points,tool geometry compensation and interpolation of cutting location points.In the discretization of cutting contact points,an integrated discretization method is given.By setting the maximum distance in Z direction between tool path and the surface,the radius for the transition of equal angle method and equal arc method is determined.In the compensation of tool geometry,a method for rake angle compensation is proposed,and the compensation of tool nose radius in normal direction and Z direction are analyzed.In the interpolation of cutting location points,the Hermite interpolation is transferred into spline interpolation due to the disadvantage that the second order derivative of former is not continuous,and the three bending monment method is proposed to calculate the entrance paprameters for interpolation.Finally,the above researches are verified by simulation analysis of toric surface and sin array surface.The simulation results show that the integrated discretization method has smaller discrete error than equal angle method and equal arc method;By using the Z direction compensation for tool nose radius,the movement in X axis becomes more stable than that by using the compensation in normal direction.With the use of three bending moment method for the interpolation of cutting location points,the velocity curve is more smooth and the fluctuation degree decreases resulting in the significantly improvement in interpolation precision.2.Simulation analysis of slow tool servo turning of complex optical surfacesBased on the characteristics of slow tool servo turning,the Johnson-Cook constitutive equation with consideration of size effect is developed based on the strain gradient plasticity theory,and the strain gradient and the length in primary deformation zone are solved.Secondly,a two-dimensional model is established in the Abaqus software to analyze the mechanism of the slow tool servo turning.The existence of the size effect and its effect on the maximum equivalent stress and cutting forces are verified.Then,the influences of the cutting edge radius,rake angle,clearance angle,depth of cutting and the cutting speed on the cutting forces are studied.According to the results,the influences of cutting edge radius,rake angle,clearance angle and depth of cutting on cutting forces are obvious,and cutting speed has the minimal influence on cutting forces.Finally,in Deform-3D software,the user subroutine is used to complete the establishment of the constitutive equation,and the simulation analysis of spherical surface and toric surface are completed by developing user motion control subroutine,and the influences of feed rate,depth of cutting and tool nose radius on surface roughness are studied.The conclusion is that the tool nose radius has the greatest influence on the surface roughness,followed by feed rate,and depth of cutting has the least influence.3.Predictive modeling of surface roughness in STS turning of complex surfacesThe orthogonal experiments are completed with 5 factors,tool nose radius,feed rate,rake angle,clearance angle and depth of cutting.According to the results of range analysis,the main influence factors of surface roughness are tool nose radius,following by feed rate,rake angle and depth of cutting,and the influence of clearance angle on surface roughness is not obvious.At the same time,variance analysis shows that tool radius is a significant factor affecting surface roughness.Based on the characteristics of different kernel functions,a hybrid kernel function is developed and used in the least squares support vector regression(LS-SVR)model.Meanwhile,the optimized quantum particle swarm optimization(WQPSO)algorithm is used to optimize the penalty factor,the kernel function parameter and the weight of hybird kernel function.In the algorithm,the wavelet function is used for mutation processing.Finally,the LS-SVR prediction model based on WQPSO method is developed.The results show that the prediction accuracy of the LS-SVR model based on WQPSO is obviously better than the LS-SVR model with grid search or basic PSO algorithm.4.Design of progressive addition lens surface and the C#based user programming in Solidworks softwareFirstly,an indirect method is used to design progressive addition lens(PAL)surface,and a variational difference method is used to solve the minimum value of the functional for evaluating the design quality.As result,the discrete point coordinates on the PAL surface are obtained.In the simulation analysis,the fitting accuracy of the Zernike polynomial method and the B spline method is compared and analyzed.The results show that the fitting accuracy of B-spline method is better than the Zernike polynomial with same degree.Then,the influences of the radius of spherical base surface on the design quality is studied,and the optimal radius is determined.Finally,the solving precision,convergence speed and solving speed of the variation difference method and B-Spline method for solving the minimization problem of the functional are discussed.The simulation results indicate that the distribution of the optical power and astigmatism is agrees well with ideal distribution.Due to the disadvantage of involving a large number of mathematical calculations and analysis,the Solidworks software is used to build the PAL surface directly,and the tool path generation plug-in is programmed with the use of application programming interface.Therefore,the surface construction and tool path generation could be completed in the same CAD software.5.Slow tool servo turning of typical complex surfacesFirstly,a NC machining software based on C#language is developed,which is used to generate machining codes for IMAC CNC system.The experiments of spherical surface,toric surface and PAL surface are carried out to verify the effects of different discretization algorithms of cutting contact points,tool geometry compensation and different interpolation algorithms of cutting location points on surface quality respectively.Moreover,the LS-SVR model built based on the experimental results of toric surface is used to predict the surface roughness of spherical surface and PAL surface.The prediction error is less than 10%,which shows the validity and generality of the prediction model.
Keywords/Search Tags:slow tool servo, complex optical surface, tool path generation, finite element simulation, surface roughness
PDF Full Text Request
Related items