| Nano-fluid, which is prepared by adding nano-particles into liquid, cansignificantly increase the thermal conductivity of the liquid. To study the nano-fluidheat transfer mechanism, the measurement on the flow field of this nano-fluid is oneof the important things.PIV(Particle Image Velocimetry)and LSV(Laser Speckle Velocimetry)iscommonly used to measure the instantaneous velocity of fluid flow field. And withthe modern computer technology and image processing technology, DPIV(digitalparticle image velocimetry)technology as succession of the former, giving moreconvenience in flow measurement, which is widely used in modern flowmeasurement.Besides the sample preparation, development of experimental setup, imageacquisition, this paper focuses on the study of image processing algorithm that isconcerned in the measurement of nano-fluid flow field. Based on a detailed discussionon the related theory and algorithms of DPIV, the algorithm speed in our task israpidly improved by using the cross-correlation method in space domain, thefrequency domain method FFT (mainly phase correlation method) and the blockmatching method that is maturely used in video compression to improve. Thedistribution map of the flow field and the precise velocity vector at any point isobtained by processing the speckle images recorded by our experiment setup. In orderto get more accurate results, furthermore, sub-pixel interpolation and fitting algorithmare used, which make the measurement accuracy of the displacement vector of imagesbelow 0.1 pixel.Research shows that the method in this paper can get effective results in theinstantaneous flow field of nano-fluid. Those results provide a theoretical referencefor the further study of nano-fluid heat transfer mechanism. |
PDF Full Text Request