An Investigation Into Nanometric Surface Roughness Characterization In Ultra-Precision Machining

Ultra-precision machining is able to achieve the manufacturing accuracy with submicron form error and nanometric surface roughness,it has been widely used for the fabrication of free-form surfaces,micro-/nano-structured surfaces,and complex optical functional surfaces.Nanometric surface roughness is an important factor to evaluate the surface quality,it generally contains high frequency,medium frequency,and low frequency signals which shows multi-scale characteristics in ultra-precision machining.However,the current ISO standard is not fully suitable for nanometric surface roughness evaluation,which not only reflects with boundary effect during the microstructured surfaces filtering,but also lacks of multi-scale analysis and characterization methods to relate machining process and evaluation results.Motivated by this,main contents of this research are as follows:1)Discrete wavelet analysis was proposed for the nanometric surface roughness extraction,and the relationship between bior 6.8 wavelet function and decomposition layers was established under a given cut-off wavelength.In order to reduce the boundary effect,different bilateral quantiles were selected for height parameters and further parameterized;2)Multi-scale analysis and characterization was carried out for nanometric surface roughness evaluation,the surface topographies characteristics under different decomposition layers were investigated and further quantified by surface roughness parameters,the intrinsic parameters,namely difference and surface curvature were proposed and parameterized to realize anisotropic features characterization.The influence of sampling scale on nanometric surface roughness evaluation was analyzed in frequency domain for a reasonable sampling condition.3)Ultra-precision machining experiments revealed that: wavelet analysis a promising approach for the extraction of the nanometric surface roughness from micro-structured surface,and the probability statistics method is capable of reducing the influence of boundary effect.Multi-scale analysis well reflects the machining process,the intrinsic parameters effectively characterize the anisotropic characteristics,and sampling scale is a key factor during nanometric surface roughness evaluation.In summary,this thesis focuses on the nanometric surface roughness evaluation in ultra-precision machining.An accurate and reasonable nanometric surface roughness evaluation has been realized for micro-structured surface,methods have been proposed to enrich the multi-scale analysis and characterization of nanometric surface roughness.The research is beneficial for promoting the development of ultra-precision machining technology.