Viscosity of carbon nanotube suspension using artificial neural networks with principal component analysis (original) (raw)
Related papers
International Communications in Heat and Mass Transfer, 2016
Regarding the viscosity of the fluids which is an imperative parameter for calculating the required pumping 18 power and convective heat transfer, based on experimental data, an optimal artificial neural network was 19 designed to predict the relative viscosity of multi-walled carbon nanotubes/water nanofluid. Solid volume 20 fraction and temperature were used as input variables and relative viscosity was employed as output variable. 21 Accurate and efficient artificial neural network was obtained by changing the number of neurons in the hidden 22 layer. The dataset was divided into training and test sets which contained 80 and 20% of data points respectively. 23 The results obtained from the optimal artificial neural network exhibited a maximum deviation margin of 0.28%. 24 Eventually, the ANN outputs were compared with results obtained from the previous empirical correlation and 25 experimental data. It was found that the optimal artificial neural network model is more accurate compared to 26 the previous empirical correlation.
International Communications in Heat and Mass Transfer, 2020
The emerging applications of nanofluids in heat transfer makes it imperative to study their viscous properties. The knowledge and assessment of physical properties with changes in concentration and temperature are essential for the practical applications of nanofluids. The first part of the current study is the synthesis of Titania (TiO 2) nanotubes via a conventional method. The experimental investigation of viscosity behavior of TiO 2 nanotubes dispersed in ethylene glycol/water-based nanofluid by different process parameters such as the mass concentration of nanotubes (0 to 1%), temperature (25-65 °C) and shear rate (150-500 s −1). The results showed a 30% increase in viscosity at 55 °C by increasing the mass concentration of nanotubes from 0 to 1%, while 22% increase was observed at 25 °C. In the second part, a multivariable correlation, and Artificial Neural Network (ANN) have been used to predict the viscosity at varying temperatures and shear rates based on the experimental data. Statistical analyses were done to investigate the accuracy of both empirical correlation and ANN modeling. It was observed from the results that ANN prediction is highly accurate, with 0.1981 AAD% and 0.999 R 2 as compared to empirical correlations (2.68 AAD%, 0.9872 R 2).
Prediction of graphite nanofluids' dynamic viscosity by means of artificial neural networks
International Communications in Heat and Mass Transfer, 2016
Recently, nanofluids have been studied extensively by the researchers as a result of the developments in nanotechnology. It is essential for researchers to know nanofluids' physical properties in order to make calculations regarding their specific research topics. Determination of viscosity issue is an actual one due to its common usage in heat transfer and thermodynamics. In this study, graphite particles are selected to have nanofluid mixture with its base fluid of pure water. Their volumetric concentrations are varied from 0 to 2% in pure water. Once the stabilized nanofluid is prepared by a sonicator and ultrasonic bath, viscosity is measured by a viscosity meter for the temperatures ranging from 20°C to 60°C. Validation of the experiments have been done by means of the comparison of them with the 32 empirical correlations in the literature. Then, Artificial Neural Network (ANN) analyses have been performed in order to have better empirical correlation than those in the literature. Furthermore, detailed information on the preparation nanofluids, measurement of viscosity, a list of measured data, numerical model by Matlab software, and alteration of viscosity with temperature and concentration have been given in the paper. It was concluded that viscosity correlations in the literature can predict different types of nanofluids' viscosity although they have been derived using specific type and diameter of nano particles and their base fluids.
Heat and Mass Transfer, 2017
The main objectives of this study have been measurement of the dynamic viscosity of CuO-MWCNT s /SAE 5w-50 hybrid nanofluid, utilization of artificial neural networks (ANN) and development of a new viscosity model. The new nanofluid has been prepared by a two-stage procedure with volume fractions of 0.05, 0.1, 0.25, 0.5, 0.75 and 1%. Then, utilizing a Brookfield viscometer, its dynamic viscosity has been measured for temperatures of 5, 15, 25, 35, 45, 55°C. The experimental results demonstrate that the viscosity increases by increasing the nanoparticles volume fraction and decreases by increasing temperature. Based on the experimental data the maximum and minimum nanofluid viscosity enhancements, when the volume fraction increases from 0.05 to 1, are 35.52% and 12.92% for constant temperatures of 55 and 15°C, respectively. The higher viscosity of oil engine in higher temperatures is an advantage, thus this result is important. The measured nanofluid viscosity magnitudes in various shear rates show that this hybrid nanofluid is Newtonian. An ANN model has been employed to predict the viscosity of the CuO-MWCNTs/SAE 5w-50 hybrid nanofluid and the results showed that the ANN can estimate the viscosity efficiently and accurately. Eventually, for viscosity estimation a new temperature and volume fraction based third-degree polynomial empirical model has been developed. The comparison shows that this model is in good agreement with the experimental data. Keywords Viscosity. Hybrid nanofluid. Newtonian fluid. Empirical correlation. Ann Highlights • Preparing the CuO-MWCNTs/SAE 5w-50 hybrid nanofluid. • Measuring the dynamic viscosity of CuO-MWCNTs/SAE 5w-50 hybrid nanofluid. • Comparison with predictions of existing models and presenting a new correlation. • The CuO-MWCNTs/SAE 5w-50 nanofluid shows Newtonian behavior. • An ANN method is employed to predict the viscosity of the hybrid nanofluid.
2022
Hybrid nano uids have great potential for use in thermal systems due to their improved thermal properties. In this paper, the rheological behavior of oil (5w30)-10% multi walled carbon nanotubes (MWCNT)-90% silicon dioxide (SiO 2) is experimentally examined in the temperature range of 5°C to 65°C. The volume fractions (VFs) are in the range of 0.05 to 1 vol.% and the shear rate (SR) range is 665.5-13330 1/s. Measuring viscosity at different SRs indicated a pseudoplastic rheological behavior of the nano uids in all VFs and temperatures. Measurement results show that the dynamic viscosity in different volume fractions is reduced when the temperature is increased from 5°C to 65°C. In addition, when the VF is increased from zero to 1%, the dynamic viscosity is augmented between 30.43% and 70.55%. Based on obtained data, a novel three-variable correlation for relative viscosity is proposed which estimates experimental results with a good accuracy. Then, the correlation results are compared to available correlations for hybrid nano uids in the literature. Finally, a GMDH-type neural network model based on experimental data is developed to predict the relative viscosity of oil (5W30)/SiO 2-MWCNT hybrid nano uids which reveals the predictability of studied hybrid nano uid using GMDH-type neural network. Nomenclature μ Dynamic Viscosity (kg/m.s) γ Shear rate (SR)(1/s) τ Shear stress (SS)(Pa) ϕ Volume fraction (VF) M Mass(kg) m Consistency index n Power law index ρ Density(kg/m 3)
Powder Technology, 2018
Nanofluids have recently been considered as one of the most popular working fluid in heat transfer and fluid mechanics. Viscosity is one of the most important thermo-physical properties that influence both momentum and heat transported by the nanofluids. Accurate estimation of this parameter is required for investigation the heat transfer performance of nanofluids. Therefore, in this study 1-the most influential variables on viscosity of the nanofluids are determined 2-various artificial intelligence (AI) models are developed for prediction of viscosity of alumina nanoparticle in various base fluids, 3-by comparing predictive accuracy of the developed models and available empirical correlations, the best one is selected. Correlation matrix analyses confirmed that the reduced pressure, invers of reduced temperature, acentric factor of the base fluids, and diameter and volume concentration of the nano particles in base fluids are the most influential independent variables on viscosity of nanofluids. Various statistical indices including mean square errors
Begell House, Inc. www.begellhouse.com, 2020
In this paper, an artifi cial neural network (ANN) has been studied for the viscosity of MWCNTs–ZnO/water–ethylene glycol (80:20 vol.%) non-Newtonian nanofl uid. To evaluate the rheological behavior of the nanocoolants, for each solid volume fraction and temperature, all experiments were repeated at diff erent shear rates. Aft er generating the experimental data, an ANN method is applied. The ANN is selected based on the diff erent generating architectures (neuron numbers). The algorithm for choosing the best ANN is presented. Also, using the correlation method, the viscosity of nanofl uid is predicted. Finally, ANN and correlation results are compared with the obtained data from the correlation method. It was found that the ANN had a bett er ability in predicting the viscosity of nanofl uid compared with the correlation method because the (MSE) of ANN was 0.0885, and the MSE of the correlation method was 0.9531. However, both approaches are useful, but ANN had a bett er ability to model the viscosity of nanofl uid based on the input values.
Correlation of viscosity in nanofluids using genetic algorithm-neural network (GA-NN)
Heat and Mass Transfer, 2011
An accurate and proficient artificial neural network (ANN) based genetic algorithm (GA) is developed for predicting of nanofluids viscosity. A genetic algorithm (GA) is used to optimize the neural network parameters for minimizing the error between the predictive viscosity and the experimental one. The experimental viscosity in two nanofluids Al2O3-H2O and CuO-H2O from 278.15 to 343.15 K and volume fraction up to 15% were used from literature. The result of this study reveals that GA-NN model is outperform to the conventional neural nets in predicting the viscosity of nanofluids with mean absolute relative error of 1.22% and 1.77% for Al2O3-H2O and CuO-H2O, respectively. Furthermore, the results of this work have also been compared with others models. The findings of this work demonstrate that the GA-NN model is an effective method for prediction viscosity of nanofluids and have better accuracy and simplicity compared with the others models.
on prediction of viscosity of nano fluids for low volume fraction of nano particles
Nanofluid which is consisting of nanoparticles in base fluid has high performance of heating and cooling in an industrial process and may create a saving in energy. The flow behavior of nanofluid plays a vital role in designing of Heat transfer equipment. Therefore, the prediction of viscosity of nanofluid which depends on base fluid properties, type of nanoparticles, temperature and particle volume fraction is now a challenging task.
Nanofluids are new heat transfer fluids which aimed to improve the poor heat removal efficiency of conventional heat transfer fluids. The dispersion of nanoparticles into traditional heat transfer fluids such as ethylene glycol, glycerol, engine oil, gear oil and water has become widely applicable in engineering systems because of their superior heat transfer properties. However, viscosity increase due to nanoparticle dispersion is an issue which needs attention and proper experimental investigation. Therefore, in this study, it is experimentally optimized the two-step preparation procedure for Al 2 O 3 -glycerol nanofluids consisting of 19, 139 and 160 nm particle sizes, and then studied the effective viscosity between 20 and 70°C for the range of 0 to 5% volume fractions. The nanofluids' viscosity showed a characteristic increase as volume fraction increases; decrease as the working temperature increases; and the smallest nanoparticles showed the highest shear resistance. Based on the available experimental data, an empirical correlation has been offered using dimensional analysis. Thereafter, a hybrid neural network based on the group method of data handling (GMDH-NN) has been employed for modeling the effective viscosity of Al 2 O 3 -glycerol nanofluid. The correlations obtained from both modeling procedures showed higher accuracy in the prediction of the present experimental data when compared to most cited models from the open literature.