Simulation Of Airbag Polishing Process And Surface Roughness Prediction For Curved Optical Elements

Airbag polishing,as a new polishing method for processing high precision curved optical components,plays an important role in the field of precision manufacturing,laying the foundation for the wide application of curved optical components.This article mainly studies the simulation of the airbag polishing process,as well as the material removal and surface quality prediction after polishing.The specific content is as follows:(1)Establish a model structure for simulation of airbag polishing machining of planar,concave convex spherical surfaces,and free form surfaces,and select the material property parameters of flexible airbag and K9 optical glass;Preparation work for material removal and surface roughness measurement of the workpiece surface before and after airbag polishing experiments.(2)The shape and area of the polishing contact zone were studied.Through polishing experiments on planar and spherical surfaces with different polishing parameters,the results of the elliptical shape of the polishing contact area and the variation of the contact area with the parameters were obtained;The finite element simulation of fixed-point polishing in airbag polishing was conducted,and the differences in pressure distribution and temperature distribution in the polishing contact area between four types of optical elements were compared.The influence rules were analyzed and compared with experimental results;After that,a simple theoretical formula for calculating the velocity of the polishing contact area was derived.Through MATLAB analysis,the velocity distribution and its influence rules in the polishing contact area of four optical elements were obtained,and the results were mutually verified with the experimental results.(3)Establish a material removal model for airbag polishing.With Preston equation as the core,the theoretical results of pressure and velocity distribution are obtained by using Hertz contact theory and rotation matrix respectively;Afterwards,the polishing trajectory calculation of straight lines and curves is introduced to achieve continuous precession.Secondly,based on the interaction between the polishing head and the workpiece at the mesoscopic scale,combined with the 3σ principles,the effective number of abrasive particles is calculated and the material removal cross-sectional area under different single abrasive particle shapes is analyzed to obtain the total volume of material removal.By using the polishing trajectory removal analysis per unit length,the Preston coefficient can be simplified to obtain a mathematical model for material removal in airbag polishing.Finally,through orthogonal experiments on polishing removal rate,the degree of influence and optimal parameters of various influencing factors on material removal rate were explored;And the theoretical material removal model calculation results obtained using MATLAB were compared with the measured results of the Taylor profilometer to verify the feasibility and authenticity of the model.(4)Conduct orthogonal experiments on surface roughness to explore the degree of influence of various influencing factors on surface roughness and the optimal parameters;Constructing a surface roughness prediction model,using the first 12 sets of parameter results from orthogonal experiments as the original dataset and the last 4 sets as the prediction set,constructing a BP neural network model for training,resulting in poor accuracy of the results;Subsequently,genetic algorithm optimization was used to iteratively solve for the optimal threshold and weight values,resulting in excellent prediction results and significant improvement compared to the relevant error situation.