Maximization of heat transfer density from radially finned tubes in cross-flow using the constructal design method

The maximization of volumetric heat transfer density from radially finned tubes in cross-flow is investigated in this study based on the constructal design method. A row of radially finned tubes is placed in cross-air flow. The tubes and the radial fins are heated at uniform temperatures and cooled by the air cross-flow. The cross-air flow is generated by a finite pressure difference. Two dimensionless pressure differences (Bejan number) are considered (Be = 10³ and Be = 10⁵). The objective function, the degrees of freedom, and the constraints in the constructal design method should be identified. The objective function is the maximization of the heat transfer density from the finned tubes. The degrees of freedom are; the fin tip-to-fin tip spacing, the number of fins, the tube diameter, the fin thickness, and the angle between the fins. The constraints are the length and height of the space occupied by the finned tubes. The pressure-driven flow and energy equations (steady, two-dimensional, and incompressible) are solved by means of the finite volume method. The ranges of the dimensionless fin tip-to-fin tip spacing are (0.2 ≤ S ≤ 1 for Be = 10³ and 0.05≤ S ≤ 0.3 for Be = 10⁵). The number of fins is changed as (N = 2, 4, 6, 8, 10, and 12). The dimensionless tube diameter is changed as (D = 0.25, 0.5, and 0.75). The dimensionless fin thickness is changed as (T = 0.001, 0.01, and 0.05). The results showed that for both (Be = 10³) and (Be = 10⁵), the highest value of the maximum volumetric heat transfer density is for (N = 2) and decreases as the number of fins increases. In addition, the minimum values of the maximum volumetric heat transfer density occur when the vertical fins exist at (N = 4, 8, and 12).     

Geotechnical properties of problematic expansive subgrade stabilized with guar gum biopolymer

Clean Technologies and Environmental Policy - This article presents a comprehensive study on the geotechnical behavior of problematic expansive subgrade stabilized by guar gum (GG) biopolymer. In...     

Challenges of oxidative/extractive desulphurization of heavy fuel oil

Oxidative desulphurization (ODS) of heavy fuel oil (HFO) has some challenges such as gum formation and a high level of waste hydrocarbons. Simple calculations show that, assuming dibenzothiophene (DBT) as a representative component of sulphur-containing components in the cut, about 20% of hydrocarbons are lost within the extraction process. An experimental investigation has been conducted in a three-neck glass flask with a mechanical stirrer to obtain more insight about the gum formation during ODS of HFO. The gum formation process was investigated in diluted forms of fuel oil. It was observed that the fuel oil converted to the gum increases by decreasing the hydrogen/carbon (H/C) content of the diluting solvents. The main reason for gum formation during ODS is some polymer formation reactions induced by peroxide radicals. At an oxidant to sulphur ratio (O/S) ratio of 0.25, no gum was formed, while at O/S = 5.0, all of the fuel oil was converted to gum. Finally, a simple and efficient extractive desulphurization procedure has been proposed as an alternative method. At the best condition, about 84% desulphurization was obtained with dimethylformamide (DMF) solvent extraction of HFO. It can be assumed as a potential method to use mild mixing conditions with low contact time for extractive desulphurization of HFO on an industrial scale.     

Enhancing CNN for Forensics Age Estimation Using CGAN and Pseudo-Labelling

Age estimation using forensics odontology is an important process in identifying victims in criminal or mass disaster cases. Traditionally, this process is done manually by human expert. However, the speed and accuracy may vary depending on the expertise level of the human expert and other human factors such as level of fatigue and attentiveness. To improve the recognition speed and consistency, researchers have proposed automated age estimation using deep learning techniques such as Convolutional Neural Network (CNN). CNN requires many training images to obtain high percentage of recognition accuracy. Unfortunately, it is very difficult to get large number of samples of dental images for training the CNN due to the need to comply to privacy acts. A promising solution to this problem is a technique called Generative Adversarial Network (GAN). GAN is a technique that can generate synthetic images that has similar statistics as the training set. A variation of GAN called Conditional GAN (CGAN) enables the generation of the synthetic images to be controlled more precisely such that only the specified type of images will be generated. This paper proposes a CGAN for generating new dental images to increase the number of images available for training a CNN model to perform age estimation. We also propose a pseudo-labelling technique to label the generated images with proper age and gender. We used the combination of real and generated images to train Dental Age and Sex Net (DASNET), which is a CNN model for dental age estimation. Based on the experiment conducted, the accuracy, coefficient of determination (R2) and Absolute Error (AE) of DASNET have improved to 87%, 0.85 and 1.18 years respectively as opposed to 74%, 0.72 and 3.45 years when DASNET is trained using real, but smaller number of images.