Characterization And Evaluation Of Micro-structured Freeform Surface Topography

This paper summarizes the results of previous studies, On the basis of Point Signature-a representation for free-form surfaces. A generalized approach for the measurement andcharacterization of freeform surfaces, named the Three-points Point Signature (TPS), ispresented in this paper. The TPS make use of intrinsic surface properties to conduct datamatching and use some algorithms to search for correspondences such as correlationfunctions. Based on the basis of the above constraints, searching for the correspondingmatch point group. Completed the matching process used the optimization function.Analyzed how to eliminate free-form surface measurement positioning error, a noveland effective way to accomplish freeform surface feature matching; by establishing pointfeature constraints, distance constraint, angle constraint, etc. With these three constraints,searched the three model point group in the measured surface which corresponded to thethree seed points in the design surface. Then established optimization equation to single outthe optimal transformation and determine if it is good enough to be accepted solution forthe registration task.The method is experimentally verified through a series of simulation experiments,Accomplished the registration experiments by using TPS method. The experiments forerror budgeting are used to determine the systematic error of the ISFM through a series ofsimulation experiments. The artificial error evaluation and noise simulated error aims toevaluate the robustness of the surface characterization method in detecting the errors.Experimental results shows that the error obtained with the TPS method, there is no loss ofany error information we added before. The simulation results proved the effectiveness ofthe method.Designed and machined the test surface, The measurement data were analyzed on thecomputer. The power spectral density method straightforward shows the frequencydistribution of the measured surface. Spectral moment method by calculating the surfaceand contour of each order spectral moments, get both surface orientation and uniformity ofinformation. Characteristic of this method is that all parameters are accurate numericaloutput, in each case under different processing workpieces do horizontal comparison, thismethod can exhibit a huge advantage. Fractal characterization of contour lines is no longera single frequency decomposition, but more attention to every tiny part of the overall similarity, and as a basis for characterization, surface morphology is proposed to have anew knowledge and understanding.