Research On Machining Aspheric Surface By Tangent Method

With the rapid development of science and technology in China,aspheric surfaces are widely used in practice,and even become one of the important optical components in some optical systems.Aspheric mirror can correct the aberration in the application system,improve the imaging quality of the equipment,simplify the system structure,so the user has higher and higher requirements for the accuracy of the workpiece surface.At present,how to ensure high efficiency and high quality of aspheric elements has become the focus of research.NC interpolation algorithm plays an important role in the aspheric ultra precision machining,which directly affects the machining efficiency and precision of machine tools to a great extent.In order to improve the machining efficiency and reduce the roughness,the genetic BP neural network track control algorithm is introduced on the basis of tangent NC machining of aspheric surface.（1）This paper introduces the characteristics of optical aspheric surface profile curve,analyzes the principle of the trajectory formation of aspheric optical elements machined by tangent method,and obtains the mathematical control model of the trajectory formation of aspheric optical elements machined by tangent method by analyzing the geometric relationship of motion.（2）By controlling the single variable method,this paper studies the influence of the process parameters on the surface roughness of the workpiece when the tangent method is used to process aspheric surface,so as to provide reliable processing parameters for the following confirmatory experiments.At the same time,the theoretical error of trajectory shaping in tangent method is analyzed,and the genetic BP neural network algorithm is proposed to improve the surface shape accuracy.（3）This paper describes the basic principles of genetic algorithm and BP neural network algorithm,illustrates BP neural network interpolation model based on the theory of tangent Machining Aspheric surface,uses genetic algorithm to optimize BP neural network for its own defects,such as local optimization,slow convergence and large workload of online training,so as to reduce the possibility of falling into local minimum in the operation process and speed up the convergence speed of the network.Through the network training of non-circular curve,the training error reaches9.995×10^-6,and the curve interpolation error is less than 0.4μm,which meets the interpolation requirements.Finally,based on the existing control system of AGPM-01,the genetic BP neural network algorithm is added to complete the trajectory optimization of aspheric surface machining by tangent method.（4）The genetic BP neural network algorithm is applied to the experiment of cutting aspheric surface to verify the feasibility and superiority of the interpolation control model.K9 glass lens with a diameter ofφ36.2mm is processed.After rough grinding,the Ra value of the workpiece is 4.1785μm,and there is no edge collapse problem;after fine grinding,the Ra value of the workpiece is 0.3821μm,and the surface of the workpiece is smooth and transparent,and there is no need for pre polishing,which saves the processing time and improves the processing efficiency and accuracy;after polishing,the Ra value of the workpiece reaches 0.0522μm,which meets the design requirements of the workpiece,and proves the feasibility of the method The algorithm is feasible.