Journal of Computational and Applied Research in Mechanical Engineering
Volume:12 Issue: 2, Winter-Spring 2023

In a turning process, it is essential to predict and choose appropriate process parameters to get a component’s proper surface roughness (Ra). In this paper, the prediction of Ra through the artificial neural network (ANN), multiple regression analysis (MRA), and random forest method (machine learning) are made and compared. Using the process variables such as feed rate, spindle speed, and depth of cut, the turning process of glass fiber-reinforced plastic (GFRP) composite specimens is conducted on a conventional lathe with the help of a single-point HSS turning tool brazed with a carbide tip. The surface roughness of turned GFRP components is measured experimentally using the Talysurf method.  By utilizing Taguchi's L27 array, the experiments are carried out and the experimental results are utilized in the development of MRA, ANN, and random forest method models for predicting the Ra. It is observed that the mean absolute error (MAE) of MRA, ANN and random forest for the training cases are found to be 39.33%, 0.56%, and 24.88%, respectively whereas for the test cases MAE is 54.34%, 2.59%, and 24.88% for MRA, ANN, and random forest, respectively.

In this research, the thermodynamic analysis of a two-stage absorption compression refrigeration system employing a flash tank with indirect subcooler is presented. The absorption cycle uses LiBr-water solution as the working fluid and prepares the high temperature medium for the bottoming cycle, which is a two-stage compression refrigeration system with R744 refrigerant. The thermodynamic analysis indicates that the proposed system decreases the required electrical work and the total exergy destruction rate results in the improvement of the overall COP and total exergy efficiency. The results are compared with the same system without the subcooler and a simple cascade absorption compression refrigeration system. It was found that the overall COP and the total exergy efficiency of the proposed system are 7.86% and 11.21% higher than the system without the subcooler. These enhancements are 11.42% and 16.48% in comparison with the simple cascade absorption compression refrigeration system. Moreover, the effect of the generator temperature, condenser temperature, cascade condenser temperature, evaporator temperature, and the intermediate pressure of the compression section on the system electrical work, overall COP, total exergy destruction rate, and the total exergy efficiency of the proposed system are discussed.

Neonatal incubators provide an artificial thermal environment to maintain the thermoregulation of premature babies. Several studies revealed the          dry and latent heat exchange estimation between the newborn's body and the surrounding environment. Heat transfer due to convection is leading over the thermal radiation in incubators. The aim of this article is to study the airflow modeling and heat transfer coefficient over an infant’s body inside the incubator. For this purpose, an experiment and a numerical simulation are carried out to develop the methodology, and subsequently computational fluid dynamics (CFD) analysis is accomplished to evaluate the heat transfer coefficient of a preterm infant. By means of the shear stress transport (SST K-ω) turbulence model, 3-D computational, models are numerically studied using the commercial CFD tool Star CCM+. Flow visualization reveals that a large-scale flow circulation pattern is produced in the mean region of the enclosed chamber, and small-scale eddies are generated at corners and close to the walls. The numerical results obtained for heat transfer assessment in the present study are validated with experimental and numerical results available in biomedical open literature.

Finite-element modeling of structures using elements without rotational degrees of freedom (DOFs) is usually stiffer than their physical behavior. Therefore, the stiffness of a structural system will be smoothed by adding rotational DOFs in the numerical model. In the traditional displacement-based finite-element method, adding drilling rotations is not easy. The main contribution of this paper is performing dynamic analyses using the finite strip element with added drilling rotations to the elements. For this purpose, any quadrilateral area is divided into two independent sets of orthogonal strips comprising truss and Bernoulli-Euler beam elements. Then by using new shape functions, mass, damping, stiffness matrices, and equivalent nodal forces are derived. Finally, time history analysis for plane stress or strain type problems for direct earthquake records is performed using the developed formulations. The numerical studies show that the results of the finite strip method using coarse meshes are competitive with the results of the finite-element method using fine meshes. This advantage is valuable in time-consuming computational problems, e.g., dynamic or nonlinear analyses.

In this paper, by considering the processing parameters, including blank holder force, blank holder gap, and cavity pressure as the most important input factors in the hydroforming process, an experimental design is performed, and an adaptive neural-fuzzy inference system (ANFIS) is applied to model and predict the behavior of aluminum thinning rate (upper layer and lower layer), the height of wrinkles and achieved depths that are extracted in hydroforming process. Also, the optimal constraints of the network structure are obtained by the gray wolf optimization algorithm. Accordingly, the results of experimental tests are utilized for training and testing of the ANFIS. The accurateness of the attained network is examined using graphs and also based on the statistical criteria of root mean square error, mean absolute error, and correlation coefficient. The results show that the attained model is very effective in approximating the aluminum thinning rate (upper layer and lower layer), the height of wrinkles, and achieved depth in the hydroforming process. Finally, the results also show that the root means of the square error of aluminum thinning rate (upper layer and lower layer), the height of wrinkles, and achieved depth of the test section are 1.67, 2.25, 0.05, and 2.67, respectively. It is also observed that the correlation coefficient for the test data is very close to 1, which demonstrates the high precision of the ANFIS in predicting the outputs of the hydroforming procedure.

The hemispherical resonator gyro (HRG) is a type of precision inertial sensor that has the advantages of direct angle measurement and unlimited dynamic range. The overall accuracy of the HRG is due to the quality of its resonator shell, and improving the performance of resonators requires a proper understanding of the processes of energy damping in each resonance cycle, which has a significant impact on sensor performance. In this paper, in order to investigate the losses in the hemisphere shell resonator, first, the equations governing the shell are studied, and three-dimensional modeling is performed in COMSOL software. By performing mechanical simulations, the resonance modes and the natural frequency of the shell are investigated, and finally, the second and third resonance modes are selected as the optimal operating mode of the gyroscope. Also, by performing thermal simulations, the dominant energy damping processes, such as thermo-elastic damping and anchor loss were analyzed and simulated, and the effect of shell material on damping was investigated. Then the quality factor of the resonator was evaluated based on its geometry and material. In this way, according to the scope of work of the gyroscope, this process can be used to design the specifications of the shell to achieve a resonator with the desired quality factor.

In the present study, the pressure drop of the nanofluid flow of carbon-water nanotubes (CNT/water) in a helical three-tube heat exchanger with constant fluid physical properties has been experimentally evaluated. For this purpose, first, the experimental device was designed and manufactured and then the carbon-water nanotube nanofluid with volume percentages of 0.01%, 0.1%, and 0.5% was prepared and stabilized. For the experiment, two triple-tube helical heat exchangers with different geometries are considered, in which the diameter of the middle pipe varies in two geometries. The pitch of the helical coil is 100mm and the helix radius is 9.235mm. The experiment was performed on Dean numbers between 1000 and 5000. The measured and calculated data were according to the available correlation in the literature with an error of less than 4%. It is found that at low volumetric percentages of CNT, the pressure drop is almost equal to that of the base fluid, and with increasing volumetric percentage of nanoparticles, the pressure drop also increases. By changing the geometry of the tube (decreasing the middle diameter of the tube), the pressure drop decreases.

In this study, comprehensive numerical simulations were conducted to examine laminar pulsatile developing flows through flat channels. The developing velocity fields and the hydrodynamic entry length were explored for the Reynolds numbers from 20 to 200, and the low and intermediate non-dimensional pulsation frequency or the Womersley number (1.08 ≤Wo≤ 8.86). For all simulations, the pulsating amplification factor was considered from zero to one, (0 ≤A≤ 1), and to achieve more practical and relevant outcomes, time-dependent parabolic inlet velocity profiles were applied. The outcomes reveal that for the higher values of the pulsation frequency or the Womersley number (6 ≤ Wo ≤ 8.66), the maximum pulsatile entranced length during a cycle is close to the inlet length of the mean component of the flow. On the other hand, for the rest of the Womersley number range (1.08 ≤ Wo < 6), and high amplification factor (0.5 ≤ A), the value of the entrance length increases and is significantly different from the development length of the steady component. Moreover, the results demonstrate that the entry length correlates with the Womersley number through a power-law function, whilst it has linear correlations with the Reynolds number and the amplification factor. Further, using the result of the accomplished numerical study, a practical correlation of the entrance length is offered to be used in the design phase for any type of pulsatile flow through the flat channels.

Carbon deposition has a serious effect on the failure mechanism of solid oxide fuel cells. A comprehensive investigation based on a two-dimensional model of a solid oxide fuel cell with the detailed electrochemical model is presented to study the mechanism and effects of carbon deposition and unsteady state porosity variation.  Studies of this kind can be an aid to identify the SOFC optimal working conditions and provide an approximate fuel cell lifetime. It has been revealed that, due to carbon deposition, the porosity coefficient of the fuel cell decreases. Consequently, a reduction in the amount of fuel consumption along the fuel cell and the chemical and electrochemical reaction rates are resulted which can be clearly seen in the off-gases molar ratio. The percentage of output fuel changes in the timeframe is useful information for optimizing CHP systems including fuel cells. The percentage of the output water vapor, which usually increases compared to the input, decreases by 17% at the end of the working period. Also, unreacted methane in the output of the fuel cell increased by 12%; in other words, it is wasted. The other consequence of carbon deposition reduced electrochemical and chemical reaction rates and the reduction of temperature difference along the cell. The study shows that after 145 working days, the temperature difference along the cell varies from 117 °C for the starting time to 7 °C. Also, by reducing the current density, the cell output power density decreases by 72% after 145 working days.

The purpose of this paper is to investigate the effect of aspect ratio on vortex shedding, and transient flow-induced noise over a rectangular cylinder is presented. The freestream velocity is assumed 50 m/s. URANS equations with turbulence model  are employed to flow analysis. Aerodynamic noise calculations are performed using the FW-H analogy. The rectangular cross-section with various lengths and widths is considered. A comparison of the results extracted in the present study with the experimental results of other references indicates the accuracy of the present research. The aspect ratios from 0.6 to 6 (equivalent to Reynolds numbers from 2.5 × 104 to 5.6 × 104) are studied. The simulations can be divided into two categories. In the first category, the ratio of length to width (R = B/H) is less than one, and in the second one, this ratio is greater than one. In the first case, noise is reduced by a relatively low slope. But in the second condition, the behavior of noise is different in various ratios and the slope of noise variations is high. The flow structure is also discussed in this paper. It is founded that for the first category, by increasing the aspect ratio, both the fluctuations and aerodynamic forces are reduced, and the longitudinal wake zone is increased. But in the second category, fluctuations of flow may be increased or decreased in various aspect ratios.