Development of a predictive model for estimating the specific heat capacity of metallic oxides/ethylene glycol-based nanofluids using support vector regression

Abstract

The specific heat capacity of nanofluids is a fundamental thermophysical property that measures the heat storage capacity of the nanofluids. is usually determined through experimental measurement. As it is known, experimental procedures are characterised with some complexities, which include, the challenge of preparing stable nanofluids and relatively long periods to conduct experiments. So far, two correlations have been developed to estimate the The accuracies of these models are still subject to further improvement for many nanofluid compositions. This study presents a four-input support vector regression (SVR) model hybridized with a Bayesian algorithm to predict the specific heat capacity of metallic oxides/ethylene glycol-based nanofluids. The bayesian algorithm was used to obtain the optimum SVR hyperparameters. 189 experimental data collected from published literature was used for the model development. The proposed model exhibits low average absolute relative deviation (AARD) and a high correlation coefficient (r) of 0.40 and 99.53 %, respectively. In addition, we analysed the accuracies of the existing analytical models on the considered nanofluid compositions. The model based on the thermal equilibrium between the nanoparticles and base fluid (model II) show good agreement with experimental results while the model based on simple mixing rule (model I) overestimated the specific heat capacity of the nanofluids. To further validate the superiority of the proposed technique over the existing analytical models, we compared various statistical errors for the three models. The AARD for the BSVR, model II, and model I are 0.40, 0.82 and 4.97, respectively. This clearly shows that the model developed has much better prediction accuracy than existing models in predicting the specific heat capacity of metallic oxides/ethylene glycol-based nanofluids. We believe the presented model will be important in the design of nanofluid-based applications due to its improved accuracy.

Investigation of effect of process variable on dimensional accuracy of FDM component using response surface methodology

“…Equations (4) and (5) show the developed quadratic models for the linear and radial dimension in terms of process variables. The value of R 2 for regression model of linear dimension is 0.979 and for radial dimension is 0.928, which is approximately close to 1, which is desirable and shows that the models are statistically correct (Alade et al , 2019a, 2019b, 2019c; Alade et al , 2018). It is also observed the minor difference between the predicted and adjusted R 2 – value which leads to the adequate relation between the input and output parameters (Al-Jamimi and Saleh, 2019): …”

“…Residual versus observation order plot of the responses indicates both positive and negative residuals runs. The occurrence of both positive and negative values shows the existence of the correlation of the parameters (Alade et al , 2019a, 2019b, 2019c). The statistical frequencies of the residuals were shown by histogram plots.…”

Utilization of Machine Learning Methods in Modeling Specific Heat Capacity of Nanofluids

“…There are two general methods that are applicable for rough estimation of specific heat capacity [46]. The first one is based on the idea of mixing theory for ideal gases (model I) which is defined as follows [47]:…”

“…where subscripts nf, n and bf refer to nanofluid, nanoparticle, and base fluid, respectively. Another correlation is proposed based on the thermal equilibrium of nanoparticles and base fluid, which is defined as follows (model II) [47]:…”

“…5, based on the values of root mean squared error (RMSE), using support vector regression led to the highest accuracy which was followed by ANN. In another work, Alade et al [47] applied support vector regression (SVR) for modeling the specific heat capacity of EG-based nanofluids with different metal oxide particles including CuO and Al 2 O 3 particles. Inputs of their model were the specific heat capacity values of the base fluid and nanoparticles in addition to temperature and volume fraction of solid phase.…”