Improving the prediction ability of reference correlation for viscosity of carbon dioxide

Abdolbaghi, Sahar; Barati-Harooni, Ali; Najafi‐Marghmaleki, Adel

doi:10.1016/j.jcou.2019.02.022

Cited by 18 publications

(6 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It cannot be compared with semi‐empirical regression models, e.g., the equation of Jarrahian and Heidaryan 52. In comparison, the NN and machine learning models 10, 58–61 show high accuracy in all phase regions including the RACP. However, they are highly dependent on the characteristics of both the training dataset and the optimization algorithm itself.…”

Section: Resultsmentioning

confidence: 99%

“…However, it is a locally optimized, stochastic algorithm and has the risk of failure for training. An improved radial basis function (RBF) network based on the particle swarm optimization (PSO) algorithm [58] shows higher training accuracy in almost all networks, but its generalization capability is usually very limited. Another new algorithm, the leastsquares support vector regression (LSSVR) [59], models the thermal conductivity using a hyperplane to fit the data and using least squares to simultaneously reduce the solution difficulty and speed up the convergence.…”

Section: Comparison Of Ai-based Modeling Approachesmentioning

confidence: 99%

“…NN and machine learning models [10,[58][59][60][61] show high accuracy in all phase regions including the RACP. However, they are highly dependent on the characteristics of both the training dataset and the optimization algorithm itself.…”

Section: Techno-economic Assessmentmentioning

confidence: 99%

See 2 more Smart Citations

Modeling Thermophysical Properties of Carbon Dioxide: Performance Comparison and Assessment

2022

View full text Add to dashboard Cite

Although many mathematical models have been developed for predicting the density, viscosity, and thermal conductivity of CO2, their accuracy and applicability have not been sufficiently established in the gas/liquid and supercritical regions and, in particular, in the region around the critical point. To understand their performances, the models are compared and analyzed, including (i) the cubic equations of state, multi‐parameter equations, and the Helmholtz free energy‐based model for density, (ii) residual theory‐ and regression‐based models for viscosity and thermal conductivity, and (iii) artificial intelligence‐based models (neural networks/machine learning) for these properties. A technical assessment is then performed. The results may provide a positive reference for determining the thermophysical properties of CO2.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Comparison Of Ai-based Modeling Approachesmentioning

confidence: 99%

See 1 more Smart Citation

Modeling Thermophysical Properties of Carbon Dioxide: Performance Comparison and Assessment

2022

View full text Add to dashboard Cite

show abstract

“…FL uses an inference mechanism during its performance. This usually leads to various cognitive uncertainties and brings complexities for designing the structure of FL because of insufficient information about the problem and different biases of human experts (Abdolbaghi et al, 2019). Hence, in order to control and process systems with cognitive uncertainty, the abilities of ANN and FL were used in ANFIS framework.…”

Section: Multilayer Perceptron Neural Networkmentioning

confidence: 99%

“…The convergence criterion of a network during training process is controlled by minimizing the Average Absolute Relative Deviation (AARD%) between experimental data and predictions of network. The workflow of multilayer perceptron neural network (MLP‐NN) model is available in literature works (Abdolbaghi et al, 2019; Al‐Musawi and Al‐Rubaie, 2017).…”

Section: Detail Of the Modelsmentioning

confidence: 99%

Accurate Prediction of Kinematic Viscosity of Biodiesels and their Blends with Diesel Fuels

Mehrizadeh

Nikbin‐Fashkacheh

Zand

et al. 2020

J Americ Oil Chem Soc

View full text Add to dashboard Cite

Viscosity of mixtures of biodiesels (admixtures) and mixtures of biodiesel/diesel (blends) is a important parameter for determining their combustion behavior. There is no universal and general model for prediction of viscosity of these systems at different conditions. Hence, developing simple, accurate, and general models for prediction of viscosity of these systems is of great importance. In this work, three computer-based models named multilayer perceptron neural network (MLP-NN), radial basis function optimized by particle swarm optimization (PSO-RBF), and adaptive neuro fuzzy inference system optimized by hybrid approach (Hybrid-ANFIS) were developed for prediction of viscosity of blends and admixtures. A number of 966 experimental data covering wide ranges of influencing parameters were utilized to develop the models. The accuracy of predictions of the developed models was examined by using different statistical quality measure approaches as well as comparing their results with the predictions of literature models. Results showed that the developed models present accurate predictions and are superior to the literature models. The predictions of PSO-RBF model were also better than Hybrid-ANFIS and MLP-NN models.

show abstract

Experimental Study and Artificial Intelligence Modeling of Dye Removal in Microfluidic Systems

Hosseini

Rahimi²

2023

Chem Eng & Technol

View full text Add to dashboard Cite

A modeling approach was utilized to achieve efficient operational conditions for Alizarin removal from synthetic wastewater in a T‐type micromixer. Besides experimental work, the neuro‐fuzzy system and artificial neural network techniques were utilized for this purpose. Input parameters were the pH, the initial Alizarin concentration in the feed, the extractant volume percentage in the organic phase, and the fluid flow rate. Based on the obtained results, both models have high precision. However, the accuracy of the neural network for estimating the extraction percentage is higher compared to that of the neuro‐fuzzy model. The optimal values of the operating parameters were determined by the genetic algorithm technique and the extraction percentage value was obtained as about 99.4 %.

show abstract

Improving the prediction ability of reference correlation for viscosity of carbon dioxide

Cited by 18 publications

References 39 publications

Modeling Thermophysical Properties of Carbon Dioxide: Performance Comparison and Assessment

Modeling Thermophysical Properties of Carbon Dioxide: Performance Comparison and Assessment

Accurate Prediction of Kinematic Viscosity of Biodiesels and their Blends with Diesel Fuels

Experimental Study and Artificial Intelligence Modeling of Dye Removal in Microfluidic Systems

Contact Info

Product

Resources

About