AIUB Journal of Science and Engineering (AJSE)

Influence of the ratio of nozzle inlet to nozzle throat areas on the performance of a jet pump for vacuum applications using computational fluid dynamics

Jose Alfredo Palacio, Ivan Patino, William Orozco — Wed, 20 Dec 2023 00:00:00 +0600

The nozzle design is one of the most important issues because it determines the pressure range and the other dimensions to guarantee an adequate performance of a jet_pump. An incorrect design of this part can cause shock waves and unnecessary overexpansion of the power fluid. The nozzle’s main purpose is to allow the high-pressure, low-velocity primary fluid to be accelerated in such a way as to substantially decrease the fluid pressure while increasing its velocity. This is achieved because the subsonic flow accelerates when entering the convergent part of the nozzle, obtaining a sonic or supersonic flow at the nozzle throat that accelerates even more when entering the divergent part of the nozzle. Therefore, to achieve the highest possible nozzle discharge velocity, the nozzle must be able to change the flow conditions from subsonic to supersonic. Considering the high importance of the nozzle design in the jet_pump performance, five cases are simulated in the present work, where the ratio of nozzle inlet to nozzle throat areas is modified (10,15,20,25 y 30), to study the behavior of three performance parameters, namely, drag coefficient (Cd), pressure ratio (PR) and Energy Efficiency (η), as well as the Mach number (Ma) and velocity fields.

The Thermal Absorption/Generation on Ferro-fluid Combined Convective Flow Over Curvilinear Porous Surfaces

Maleque Kh. Abdul — Wed, 27 Dec 2023 10:09:58 +0600

In the influence of fluid buoyancy forces, the ferrofluid combined convective flow in porous curvilinear surfaces is studied with thermal generation/absorption effect. In the ambient flow conditions, the pressure gradient terms and ferrofluid buoyancy forces are replaced by the free steam velocity . The governing equations of the present problem are converted to ODEs by introducing non-dimensional functions and similarity variable. Boundary conditions of first derivative of velocities and temperature of our problem were constructed by the initial value problem, also the unknown initial conditions are found by shooting methods, and then a set of ODEs is solved numerically by the integration scheme of the six-order Range-Kutta method. The results of the solutions are presented graphically of velocity and thermal profiles with the help of MATLAB for different values of suction parameter and heat absorption parameter . Finally, the comparisons of the results highlight the justification of the numerical calculation accepted in the presence study. The problems in curvilinear surface study of boundary layer flow are complicated in fluid mechanics with applications of natural science and engineering.

Modeling and performance analysis of a transparent multilayer solar cell

Maria Sultana Rupa, M. Tanseer Ali, Mohammad Abdul Mannan, Mehedi Hasan — Wed, 20 Dec 2023 00:00:00 +0600

Transparent solar cells have emerged as a promising frontier in renewable energy research, offering the dual functionality of generating electricity while maintaining transparency. Five layers of InAs/InSb/AlGaAs/GaN/Si are incorporated into the proposed model structure which takes the properties of the source materials into consideration. Simulations of electromagnetic waves are used to evaluate optical and electrical properties. The cell is assumed to function at room temperature in the simulated settings. The I-V curve's fill factor (F.F.) of 0.6531 corresponds to a maximum conversion efficiency of 15.2655%, according to the data. Furthermore, at 530 nm, this combination and device configuration show a very good transparency of up to 60%.

Calibration and Validation of INA219 as Sensor Power Monitoring System using Linear Regression

Wed, 27 Dec 2023 11:43:55 +0600

Electricity demand which increases up to 2.7%, needs to be evaluated to prevent power wastage. This paper proposes an INA219 sensor and a power monitoring solution based on the ESP8266. Power Monitoring stores and displays real-time data in Google Sheets via Blynk version 1.0.1. The system has been calibrated with a fixed LED and resistor as a voltage calibration load. Meanwhile, the lamp and shunt resistors calibrate the shunt voltage. The measuring tools for comparison in calibration are digital multimeters, oscilloscopes, and power data loggers. Calibration using the linear regression technique with accuracy, precision, and uncertainty analysis are determined by Mean Absolute Percent Error (MAPE), Relative Standard Deviation (RSD), and Gaussian distribution. Successively, the sensor coefficient of determination (R2), accuracy, precision, and uncertainty of the load voltage and shunt voltage are 0.999 and 0.997, 99.27% and 93.71%, 99.82% and 99.55%, 0.37 V and 0.89 mV.

Modeling and Economic Assessment of an Agricultural Microgrid: A Comparative Analysis

Sayeed Hasan, Md. Rifat Hazari, Effat Jahan, Mohammad Abdul Mannan — Wed, 20 Dec 2023 00:00:00 +0600

Sustainable energy resources are essential to meet the world's growing population and extending energy demands. Among the potential solutions, incorporating renewable energy sources into hybrid energy systems holds a lot of opportunities. This paper presents a design and economic analysis for an off-grid microgrid intending to power agricultural loads. Solar resources and PV-inverter system were modeled using pvlib

-python, while the remainder of the microgrid, including the battery energy storage system (BESS) and biogas-based generator (BGG), was modeled and simulated using a custom dispatch method. The same system was modeled in Homer as well, and the outcomes of the designed microgrid were compared. When compared to Homer, the proposed approach reduced life cycle cost (LCC), and leveliz ed cost of energy (LCOE) by 25%, and 20%, and emissions by 85%. In terms of generation, the proposed strategy reduced PV production by 20%, BGG output by 85%, and unmet load and surplus energy by 14% and 65%, respectively. The study additionally addressed an in-depth approach to modeling PV using various data sources and the associated modules and

functionalities.

Employee Promotion Prediction Using Improved AdaBoost Machine Learning Approach

Md. Abu Jafor, Md. Anwar Hussen Wadud, Kamruddin Nur, Mohammad Motiur Rahman — Wed, 20 Dec 2023 00:00:00 +0600

Employee promotion is an important aspect of the human resource management process. Due to different factors, it refers to the automatic improvement among the employees in an organization. Promoting employees from the lower level to the higher level brings a feeling of satisfaction among the employees. It improves their job satisfaction and motivation by providing more significant income, status, and responsibilities. By building up loyalty, promotion reduces employee attrition. Thus, it is difficult to accurately decide, whether an employee should or should not be promoted based on their current and past performance. So, human resource management does research about promotion, because there are a limited number of research about the finding of employee promotion prediction in the existing studies. First, to find the reasons for employee promotion, we need to analyze the research study for finding the factors which are related to the promotion. The aim of this research study is to implement an employee promotion prediction framework using machine learning. A modified AdaBoost classifier is used for automatic promotion prediction, and six machine learning techniques for instance, Support Vector Machine (SVM), Logistic Regression (LR), Artificial Neural Network (ANN), Random Forest (RF), XGBoost (XGB), and AdaBoost are applied in performance comparison. Through a complex assessment process, the performance of these supervised machine learning algorithms for predicting employee advancement is analyzed using assessment metrics on the employees' evaluation dataset for promotion prediction. The Artificial Neural Network (ANN) and AdaBoost model provide better results on this dataset than all traditional machine learning techniques. Finally, Our proposed modified AdaBoost approach outperformed all other methods evaluated with an accuracy of 95.30%.

Unraveling the Burden of T2D among the Adolescents in Bangladesh: A Statistical Exploration of Prevalence and Influencing Factors

Md. Mortuza Ahmmed, M. Mostafizur Rahman, Mahfuz Khatun — Fri, 22 Dec 2023 00:00:00 +0600

This study aims to investigate the prevalence and determining factors of Type 2 Diabetes (T2D) among youths in Bangladesh using a statistical approach. The research objectives were to determine the prevalence of T2D in this population and identify the factors associated with its occurrence. A survey questionnaire was formed encompassing certain relevant variables. A sample of youths was selected through cluster sampling strategy. By collecting relevant data and employing appropriate statistical analyses, the study provided insights into the prevalence and associated factors of T2D among the youths, which can contribute to the development of effective prevention and management strategies. Statistical analyses were performed using chi-square tests and logistic regression, to explore the relationships between T2D prevalence and the determining factors identified in the study. Lifestyle factors played a significant role in the development of T2D among youths. Besides, certain socio-demographic factors like occupation, education, income, age, marital status, and residential origin were found to be associated with an increased risk of T2D among youths in Bangladesh. These findings highlight the multifactorial nature of T2D among youths in Bangladesh. Addressing these factors through targeted interventions and public health policies can play a crucial role in preventing and managing T2D in this population. The study emphasized the importance of health awareness and education programs targeting youths in Bangladesh. The findings from this study can contribute to the development of evidence-based strategies to prevent and manage T2D in this population, ultimately reducing the burden of T2D in Bangladesh

Probabilistic Modeling for Conditional Statements

Alaa Ghazi, Yasir Hashim — Fri, 22 Dec 2023 00:00:00 +0600

A new mathematical framework is proposed in this study to comprehend the impact of program architecture on input random variables, the IF statement was the main topic. The primary idea that is theoretically and experimentally supported in this study is that the part of the joint pmf of a collection of random variables that represents the condition will be shifted to the part that represents the action. After sorting two random variables, the framework is used with four random variables, and the theoretically produced results were realistically validated. The study's equations can be applied to assessing probabilistic models of various sorting algorithms or other intricate program structures. This may also result in future investigations formalizing more precise execution time expectations.

DSC Index: Measuring the Digital Supply Chain Practice among the Higher Education Institutions Community in Least Developed Countries

Nurakmal Ahmad Mustaffa, Malina Zulkifli, S A M Manzur Hossain Khan — Fri, 22 Dec 2023 00:00:00 +0600

The 4th Industrial Revolution, more commonly referred to as Industry 4.0, has brought about a wave of multifaceted changes across the industrial spectrum around the world, and it has triggered the digitalisation of supply chains and their management regardless of the type of organisation. With increasing interconnectivity through various sectors, digital supply chain (DSC) practices and intentions have also become integral to higher education institutions. As streamlined, automated administrative processes and virtual classes conducted through online platforms become the norm, digitalisation has been catalysed in the education sector. However, several sociocultural, economic, and psychographic factors influence the adaptation of new technologies, especially in developing countries such as Bangladesh. This study uses the composite index approach to determine the Index derived from the correlation between the factors and their impact on the DSC practices and intentions. The study indicates that Trust (T) is the primary influencer, along with Performance Expectancy (PE), closely followed by Facilitating Value (FV), Facilitating Conditions (FC), and Digital Literacy (DL).

Aspect-based Sentiment Analysis Model for Evaluating Teachers' Performance from Students' Feedback

Abhijit Bhowmik, Noorhuzaimi Mohd Noor, Md Saef Ullah Miah, Debajyoti Karmaker — Fri, 22 Dec 2023 00:00:00 +0600

Evaluating teachers' performance is a fundamental pillar of educational enhancement, guiding the evolution of pedagogical practices and fostering enriched learning environments. This study pioneers an innovative approach by harnessing sentiment analysis within an aspect-based framework to decipher the intricate emotional nuances embedded within students' feedback. By categorizing sentiments as positive, negative, and neutral, we delve into the diverse perceptions of teaching aspects, offering a multifaceted portrait of educators' contributions. Through meticulous data collection, preprocessing, and a deep learning sentiment analysis model, we dissected student comments into distinct teaching aspects. The subsequent sentiment analysis unearthed positive, negative, and neutral sentiments. Positive sentiments highlighted strengths and effective communication, while negative sentiments illuminated areas for growth. Neutral sentiments provided contextual equilibrium, forming a holistic tapestry of teachers' performance. The proposed model achieved 86\% F1 score for classifying sentiments into three classes.