Simulation And Analysis Of Small Tool Polishing Process On Optical Lens

The application of aspherical optical lens in optical systems can effectively improve the imaging quality and reduce the number of optical lens in the optical system,making the system smaller in size,simplified in structure,and lighter in weight.Therefore,it has been widely used in aerospace,optical instruments and other fields.With respect to the processing of aspherical lens,the most common method at present is the computer-controlled polishing technology of small polishing tools.Among them,the wheel-shaped polishing tools technology has the characteristics of smaller beam diameter,higher Gaussian distribution and higher stability of the material removal function,which can better correct the optical surface shape error and reduce the influence of edge effects.However,this technology has problems such as long processing time and processing accuracy to be improved.For this reason,this paper optimizes the dwell time algorithm and polishing path planning in this polishing technology.First,the polishing tools material removal function was modeled and analyzed.Based on the Preston equation and vector analysis,the material removal function of the wheel-shaped polishing tools was modeled.It is concluded that the removal function shape satisfies Gaussian-like distribution when the rotation speed ratio is greater than 0.1.Because of its structural advantages,the wheel-shaped polishing tools has the advantages of smaller beam diameter of the removal function(effectively reducing the edge effect)and stable machining process compared with the traditional disc-type polishing tools.Then,based on the removal function of the wheel-shaped polishing tools,the algorithm for solving the dwell time was optimized.The solution of dwell time is one of the key issues in correcting the surface shape error of the optical lens during polishing.According to the actual physical meaning of the polishing process of the small polishing tools,the solution process of the dwell time was transformed from deconvolution to matrix equation solving.At the same time,the regularization factor a was introduced to overcome the morbidity of solving matrix equations caused by the excessive number of conditions.Convolution iteration method and regularization method based on matrix equations are used to simulate and process the same structured initial surface,under the same conditions,iterated 35 times.Convergence rate of the surface error using the convolution iterative method was 91.57%,while the convergence rate of the surface error based on the matrix equation solving method was 95.21%.The latter shows better in surface error convergence.Finally,the polishing path planning method was improved based on the former research.The Archimedes spiral is often used as a path planning method in the polishing of rotationally symmetric surfaces.In order to against the disadvantage of too fast rotation speed at the center of the optical lens when the Archimedes spiral path is used for polishing path planning of small polishing tools,introducing ? factor to deform the Archimedes spiral path,generate a non-uniform spiral path,and use the equal arc length method to plan the dwell points so as to overcome the disadvantage of the tool rotating too fast at the machining center.Simulation processing was performed on a rough surface with a diameter of 200 mm.When the convergence accuracy was equivalent,the non-uniform spiral length path of equal arc length was 88.1 min less than the Archimedes spiral,indicating that the optimized path effectively improved the machining effectiveness.