Al₂O₃-H₂O Nanofluid Viscosity Prediction Model And Experimental Study Of Surface Tension

Due to its excellent thermal conductivity,nanotechnology is innovatively combined with traditional thermal engineering.However,due to the influence of preparation technology,experimental conditions and measurement technology,the stable preparation method of nanofluids is still immature,which restricts the application and development of nanofluids in practical engineering.The surface tension and viscosity of nanofluids have a certain impact on the liquid surface transmission,resulting in energy consumption,equipment wear and other problems.Therefore,further improving the preparation stability of nanofluids,establishing the viscosity prediction model of nanofluids and revealing the variation of surface tension of nanofluids have practical engineering application value and certain theoretical research significance.Aiming at the problems of the preparation stability of Nanofluids and the limited prediction accuracy of viscosity prediction model,the alumina water-based nanofluids with different volume fractions were prepared by"two-step"method.The stability preparation of water-based alumina nanofluids,the establishment of viscosity prediction model and the experimental exploration of surface tension were studied.The main research contents are as follows:0.1 vol%Al₂O₃-H₂O nanofluids were prepared by"two-step"method,and characterized by TEM and XRD.The effects of time and different surfactants on the stability of water-based nanofluids were investigated.It is found that the actual particle size of nanofluids is larger than that of solids(20nm),and the stability of nanofluids decreases with the standing time.The addition of surfactant PVP and SDS can significantly improve the suspension stability of nanoparticles in the base solution,while the particle size in the nanofluid with CTAB can reach 856 nm,which indicates that a variety of agglomeration has occurred.The results show that the stability of nanofluids prepared by adding surfactant SDS,magnetic stirring for 15 minutes and ultrasonic oscillation for one hour is good,which can be used in the next step of physical property measurement experiment.The influence of the content of alumina nanoparticles(0.1-1vol%)and temperature on the viscosity of nanofluids was studied.The results show that the viscosity of nanofluids increases with the increase of volume fraction and decreases with the increase of temperature.In addition,the data-driven method of artificial neural network is used to predict the viscosity of nanofluids.The prediction accuracy of BP neural network,RBF neural network and polynomial regression model for viscosity of nanofluids was compared.The results show that,compared with the polynomial regression model,the artificial neural network has higher accuracy in the prediction modeling of nanofluids,and the BP neural network with the optimal structure can be determined as 2-5-1structure.The BP neural network with improved L-M algorithm has the highest prediction accuracy,R2,RMSE and SSE are 0.9996,6.3e-3 and 4.764e-4 respectively.In this paper,the change of surface tension of nano fluid with the addition of nano particles was studied experimentally.The change of surface tension of nano fluid with the addition of nano particles and temperature was mainly explored.The results show that within a certain concentration(0.1%-1 vol%),the surface tension of nanofluids increases with the addition of nanoparticles,and decreases with the increase of temperature.