Wear Ttrace Micro/nano Detection Using White-light Phase-shifting Interfermotry And It's 3D Rrepresentation

Because the data of microscopic appearance can provide abundant valuableinformation for the analysis of failure mechanism, the measurement andrepresentation of failure surface morphology of workpiece gradually trends to lowerdimensions with the development of micro/nano technology.Among all methods forsurface profile measurement, white-light phase-shiftinginterference technique can notonlyrealizefastandnon-contactmeasurementwithmicro-nano-levelprecisionin3D,butalsohavesimplesystemstructurewithlowcost.Ontheotherhand,inthemethodsof representation surface topography, the frequency domain method based on thepower spectrum density can convert the time and spatial information to frequencydomain, where not only a single frequency or wavelength of information can behighlighted, but the absolute or relative influence of different wavelength ingredientson the mean square root of surface roughness can be revealed by simple calculationwithout cycle count. Based on this, the paper adopted white-light phase-shiftinginterferencetechniquetorealizethemicro/nanodetectionofweartraces,andexploreda new method of wear failure analysis by processing and analyzing the data with thepowerspectrumrepresentation.Firstly, in establishing apparatus for microscopic three-dimension profiledetection, we designed a kind of closed-loop PZT micro/nano positioning scanningsystem: The main core of the system is 80C196KC, using a bipolar capacitancedisplacement sensor (ZNXsensor) to detect the displacement of micro displacement.AdoptingFuzzy-PID control algorithm, the algorithm not onlyinherited the qualityofroutine PID, but also possessed the advantage of fuzzy control. Secondly, in order toimprove the speed of 3D reconstruction, this paper proposed a partition algorithm ofFFT, and carried out mathematical derivation and the corresponding application, thisalgorithm can improve the speed of transformation of FFTthrough the original matrixpartitioning and diagonalization processing. Then, interference images of particleattrition sample(mild wear, serious wear, heavier wear or out of order) were obtainedwith a micro/nano positioning scanning system, fast space frequency domainalgorithm was used to process the interference images and reconstruct the 3D surface topography. Finally, 2D FFT was adopted to process the reconstructed 3D surfaceimage, through which the power spectral, angular spectrum, radius spectrum of wearsample were obtained so as to represent and analyze various property of the wearsample.Experimental result shows that: the designed micro/nano positioning scanningsystem has the output voltage ranging form 0 to 200V and the maximum ripplevoltage within 1.06mv, the control accuracy is 5nm and relative error of system is0.15%. This paper presented a partition algorithm of FFT, the algorithm greatlyimproves both the speed and the quality of 3D reconstruction. Based on therepresentation of surface's characteristics, the reasons of wear were analysised andcorrespondingpreventivemeasureswereinvestigated.