Research On Surface/subsurface Damage In The Optical Element Polishing Process

With the rapid development of space remote sensing observation technology,optical reflectors,which are one of the key components of space remote sensing cameras,are continuously developing toward large diameter,off-axis and highly lightweight.Therefore,the higher requirements have been put forward for the selection and processing of mirror materials.Silicon carbide represents a kind of ideal space mirror material by virtue of its high hardness and strength,excellent chemical stability,and good wear resistance.However,silicon carbide has high brittleness and low fracture toughness,which makes it easy to produce surface/subsurface micro-cracks,dislocations,and phase transitions during precision machining,thereby affecting the surface integrity and fatigue properties of the material.This may heavily degrade the mechanical strength and fatigue properties,as well as its adoption of surface integrity.Computer-controlled optical surfacing is a processing method that can effectively reduce the surface/subsurface damage caused by other processing operations.The common polishing methods are mainly based on contact type.Therefore,the surface/subsurface damage such as micro-cracks,brittle fracture and residual stress concentration still remains on the surface of the workpieces.In order to obtain better surface/subsurface quality during the polishing process of silicon carbide,it is important to study the mechanism of surface/subsurface damage and the depth of crack in silicon carbide materials.In this paper,on the basis of the removal mechanism and kinematic characteristics of the precision polishing of silicon carbide,a mathematical model was established between polishing process parameters(polishing depth,spindle speed,feed speed and abrasive grain size)and surface/subsurface damage.The finite element model of hard and brittle material was established by using the finite element simulation technology.By observing the dynamic polishing process,the change of the surface depth and the subsurface damage depth under different process parameters was analyzed.And the validity of theoretical model and simulation results was verified by a series of precision polishing experiments about silicon carbide.Finally,on the basis of experiments,a subsurface damage prediction model was established using BP neural network.And a subsurface damage suppression model was established by using an improved bacterial foraging algorithm(BFO)combined with a prediction model,and the process parameters were optimized.The optimum technological parameters and the minimum depth of subsurface damage of silicon carbide polishing are obtained.The study can be summarized as follows:(1)Based on the removal mechanism,fracture theory and kinematic characteristics of hard and brittle materials in the polishing process,the theoretical model of the surface/ subsurface damage depth considering the particle distribution and dynamic characteristic parameters is established.The mathematical relationship between damage depth and process parameters is obtained.The effect trend of different process parameters on the depth of surface/subsurface damage is preliminarily determined.(2)The two-dimensional and three-dimensional model of the single-abrasive grain polishing process is established by using ABAQUS finite element simulation software.The deformation and damage mechanism of material during the polishing process and the effect of different processing parameters on the depth of the surface crushing layer and the subsurface crack are analyzed.According to the analysis of the simulation data,the varieties of surface/subsurface damage with the parameters are discussed.(3)The effects of polishing parameters such as polishing depth,spindle speed,feed rate and abrasive grain size on the surface roughness and subsurface damage depth are revealed by the precision polishing experiment of silicon carbide ceramic.The validity of the theoretical model and simulation model is verified.(4)Subsurface damage is suppressed by optimizing process parameters,and a prediction model of subsurface damage depth was established by using BP neural network.The improved bacterial foraging algorithm was combined with the neural network prediction model.The subsurface damage depth was taken as the objective function.The polishing process parameters were used to construct the fitness function for the optimized parameters,and the subsurface damage suppression model was established.