A novel nonlinear regression model of SVR as a substitute for ANN to predict conductivity of MWCNT-CuO/water hybrid nanofluid based on empirical data

Karimipour, Arash; Bagherzadeh, Seyed Amin; Taghipour, Abdolmajid; Abdollahi, Ali; Safaei, Mohammad Reza

doi:10.1016/j.physa.2019.01.055

Cited by 129 publications

(39 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sharma and Singh [46] investigated stagnation point flow past a linearly stretching sheet with additional effects of variable thermal conductivity, heat source/sink, and MHD. Some other important studies with some interesting experimental findings are given in the works of [47][48][49][50][51][52][53][54][55][56][57][58][59][60] and cross references cited therein.…”

Section: Introductionmentioning

confidence: 99%

Influence of Single- and Multi-Wall Carbon Nanotubes on Magnetohydrodynamic Stagnation Point Nanofluid Flow over Variable Thicker Surface with Concave and Convex Effects

et al. 2020

View full text Add to dashboard Cite

This paper reports a theoretical study on the magnetohydrodynamic flow and heat exchange of carbon nanotubes (CNTs)-based nanoliquid over a variable thicker surface. Two types of carbon nanotubes (CNTs) are accounted for saturation in base fluid. Particularly, the single-walled and multi-walled carbon nanotubes, best known as SWCNTs and MWCNTs, are used. Kerosene oil is taken as the base fluid for the suspension of nanoparticles. The model involves the impact of the thermal radiation and induced magnetic field. However, a tiny Reynolds number is assumed to ignore the magnetic induction. The system of nonlinear equations is obtained by reasonably adjusted transformations. The analytic solution is obtained by utilizing a notable procedure called optimal homotopy analysis technique (O-HAM). The impact of prominent parameters, such as the magnetic field parameter, Brownian diffusion, Thermophoresis, and others, on the dimensionless velocity field and thermal distribution is reported graphically. A comprehensive discussion is given after each graph that summarizes the influence of the respective parameters on the flow profiles. The behavior of the friction coefficient and the rate of heat transfer (Nusselt number) at the surface (y = 0) are given at the end of the text in tabular form. Some existing solutions of the specific cases have been checked as the special case of the solution acquired here. The results indicate that MWCNTs cause enhancement in the velocity field compared with SWCNTs when there is an increment in nanoparticle volume fraction. Furthermore, the temperature profile rises with an increment in radiation estimator for both SWCNT and MWCNT and, finally, the heat transfer rate lessens for increments in the magnetic parameter for both types of nanotubes.

show abstract

Section: Introductionmentioning

confidence: 99%

Influence of Single- and Multi-Wall Carbon Nanotubes on Magnetohydrodynamic Stagnation Point Nanofluid Flow over Variable Thicker Surface with Concave and Convex Effects

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Support Vector Regression (SVR) belongs to a classic application in Support Vector Machine (SVM) [1]. It has been widely used in the fields of sample regression [2], data prediction [3], fault diagnosis [4] and so on. The existing libraries for implementation of SVR, such as SVM Light and LibSVM, need to be run based on personal computers.…”

Section: Introductionmentioning

confidence: 99%

“…Performance with different platforms. The precision type of floating point IP core is single 2. The cache of kernel is valid.…”

mentioning

confidence: 99%

Research and Implementation of ε-SVR Training Method Based on FPGA

Liu

et al. 2019

Electronics

View full text Add to dashboard Cite

Online training of Support Vector Regression (SVR) in the field of machine learning is a computationally complex algorithm. Due to the need for multiple iterative processing in training, SVR training is usually implemented on computer, and the existing training methods cannot be directly implemented on Field-Programmable Gate Array (FPGA), which restricts the application range. This paper reconstructs the training framework and implementation without precision loss to reduce the total latency required for matrix update, reducing time consumption by 90%. A general ε-SVR training system with low latency is implemented on Zynq platform. Taking the regression of samples in two-dimensional as an example, the maximum acceleration ratio is 27.014× compared with microcontroller platform and the energy consumption is 12.449% of microcontroller. From the experiments for the University of California, Riverside (UCR) time series data set. The regression results obtain excellent regression effects. The minimum coefficient of determination is 0.996, and running time is less than 30 ms, which can meet the requirements of different applications for real-time regression.

show abstract

“…One of the recent technologies of improving heat transfer is suspending nano‐sized solid particles of metallic or oxide‐metallic with high intrinsic thermal conductivity in a base fluid. This mixture is called a nanofluid …”

Section: Introductionmentioning

confidence: 99%

“…This mixture is called a nanofluid. [1][2][3][4][5] Nanofluids thermophysical properties can be controlled by chemical and physical specifications of the base fluid and nanoparticles. Nanofluids thermal conductivity is one of the factors that play an important role in heat transfer performance.…”

Section: Introductionmentioning

confidence: 99%

Modeling the effective thermal conductivity of nanofluids using full factorial design analysis

Grine

Benhamza

2019

Heat Trans. Asian Res.

View full text Add to dashboard Cite

In this paper, a full factorial design analysis is proposed for predicting nanofluid thermal conductivity ratio (TCR) as well as determining the effects of critical factors and their interactions. A statistical design of experiment approach with three variables (volume fraction, temperature, and nanoparticle diameter) at two levels is carried out. Three types of oxide‐water nanofluids (Al2O3‐water, CuO‐water, and TiO2‐water) are used to evaluate the effectiveness of the proposed mathematical model. The significance and adequacy of the regression model were evaluated by the analysis of variance. The predicted model has a root mean square error equals to 0.0074, R2 = 0.99, and P < .0013, thus showing good results compared to a set of experimental data as well as other mathematical model results. The results illustrate that the TCR of metallic oxide nanoﬂuids increases with temperature and nanoparticles volume fraction but decreases when nanoparticle size intensiﬁes. Furthermore, it is found that the nanoparticles volume fraction has a great impact on the nanoﬂuids thermophysical properties. Finally, the obtained results confirm that the proposed model is considerably accurate and capable of predicting nanoﬂuids thermal conductivity and that it can be used with ease as an alternative to many other models.

show abstract

A novel nonlinear regression model of SVR as a substitute for ANN to predict conductivity of MWCNT-CuO/water hybrid nanofluid based on empirical data

Cited by 129 publications

References 66 publications

Influence of Single- and Multi-Wall Carbon Nanotubes on Magnetohydrodynamic Stagnation Point Nanofluid Flow over Variable Thicker Surface with Concave and Convex Effects

Influence of Single- and Multi-Wall Carbon Nanotubes on Magnetohydrodynamic Stagnation Point Nanofluid Flow over Variable Thicker Surface with Concave and Convex Effects

Research and Implementation of ε-SVR Training Method Based on FPGA

Modeling the effective thermal conductivity of nanofluids using full factorial design analysis

Contact Info

Product

Resources

About