NTU Journal of Engineering and Technology

Innovative technologies of future generation networks such as Cyber-Physical System (CPS), Mobile Ad Hoc Network (MANET), Vehicular Ad-Hoc Network (VANET), Internet of Things (IOT), and Wireless network commonly known as Wi-Fi have emerged, which require a distinguished understanding of the main challenges and constraints that face the design and implementation of an Intrusion Detection Systems (IDS) for such type of networks. Moreover, a dramatic increase in the rate of cyber-attacks has increased, and new cases of intrusions, bugs, novel attacking tactics, and vulnerabilities are evolving daily. Intrusion Detection Systems (IDS) are one of the solutions against these attacks. Thus, IDS needs to improve its performance in terms of its ability to detect new attacks and respond to threats. Getting suitable datasets for evaluating various research designs in IDS design domains is a significant challenge"“. The machine learning (ML) design approach can quickly identify trends and patterns of intrusions, bugs, tactics, and cyber vulnerabilities with minimum human intervention. This paper reviews datasets for the research community. Furthermore, it explores the challenges of Dataset for intrusion detection based on Machine learning. It glances through a period of 6 years of intrusion detection datasets, explores what is currently applicable, outlines criteria for selecting the best Dataset, and explores future directions for creating relevant datasets.

In this paper, the Auto-Reclosing Relay (ARR) was designed and implemented with monitoring on the Internet of Things (IoT) and local SCADA systems. This system detects the faulty phase depending on the values of the angles of the actual impedances (postive sequence). The hardware of this system includes three PZEM-004T modules that are used to measure the electrical power parameters and the ESP8266 is employed to gather data from detectors and send it via the web to the Blynk server and to local SCADA via USB cable. The local SCADA system of the power line is used to monitor and control in HMI which is programmed by using the C# language.This paper gives a complete description of the compact system (ARR, SCADA and IoT),which works in an acurate and successful mechanism.

The focus of this study is to generate maps of clay and clay mineral deposits in the Kirkuk, Bor, and Jambor structures using Landsat imagery. The authors utilized the band ratio (5/7) and band composite (3,4,5) techniques in their analysis. Fifty-two sites were selected as testing data to evaluate the accuracy of the generated map, which showed a correlation of 0.74. The study concludes that remote sensing techniques provide a cost effective and accurate means to map the locations of clay and clay mineral deposits, offering an alternative to traditional mapping methods.

This work aims to investigate the influence of the presence of three cylinders with rounded edges on the drag coefficient. Three heated cylinders, whose shapes varied from triangle to circle, were placed in a 2-dimensional horizontal channel with laminar, steady flow. The effect of five different Reynolds numbers (Re = 100, 200, 400, 800, and 1200) and three vertical distances (ST = 1.5D, 2D, and 3D) has been studied. A constant heat flux (q'' = 5000 W/m2) subjected to the cylinders’ walls and the rounded corners radii of the triangular cylinders was (1 mm, 1.6667 mm). The finite volume method is used to solve the governing equations (ANSYS FLUENT 2022 R2). It was observed that the coefficient of drag decreases as the Re increases and decreases with decreasing longitudinal distance. The drag coefficient decreased from (37.01%) to (72.3%) when the Reynolds number increased from (100) to (1200) at ST = 3D and from (33.1%) to (62.4%) at ST = 1.5D.

The widespread use of single-phase rectifiers, particularly in residential buildings for LED lighting fixtures, has led to an increase in harmonics in power distribution systems. These harmonics pose significant challenges to power grids and require a thorough assessment of their impact on distribution networks. Therefore, the purpose of this work is to clarify how higher capacitance values at rectifiers' output, which are essential for reducing DC voltage waveform ripple, affect the quality of the input power supply. This study employs analyses and simulations conducted via MATLAB/Simulink to investigate the AC/DC converter circuit and its associated load. The analysis includes calculating output current and voltage ripples, as well as evaluating the total harmonic distortion (THD) in the input current with different capacitor values. The results highlight the adverse consequences of increasing filter capacitance values, which affect both the input current purity and the power factor. Notably, the THD showed a significant increase, reaching approximately 134%, with a low power factor of 0.5 when the output ripple voltage decreased from 307 V to 64 V using a capacitor filter of 1500 ?F.

In This paper compares the control strategies for three phase voltage source inverters, which have been the subject of extensive research in the last few years. Even if the inverter topologies that require the fewest switching elements yield the most efficient results, the switching technique that is as effective as the topology, at least. In addition, the inverter's selected switching technique will effectively suppress harmonic components while generating the optimal output voltage and current. To manage voltage source inverters, to regulate three phase voltage source inverters, several PWM approaches are frequently employed. Sinusoidal PWM and hysteresis band PWM are two examples of these techniques. MATLAB/SIMULINK has been used to model these modulation techniques that are applied in voltage source inverters to generate variable frequency and amplitude output voltage and current.
Furthermore, a comparison is made between the overall harmonic distortions of the output voltages and current. Simulation tests have revealed that there is greater THD in the output current and voltages of voltage source inverter when sinusoidal pwm is used. Hysteresis band pwm has shown the ability to produce a lower overall harmonic distortion and more efficient output voltage and current.

An access point offers wireless networking to several wireless client nodes in the shared medium when using IEEE 802.11 wireless local area networks in infrastructure mode. The related nodes could be at risk of security attacks. In the field of cybersecurity, it is not sufficient to identify vulnerable points; rather, the difficulty lies in learning how hackers can take advantage of weak points to take countermeasures. This research paper simulates some of the risks caused by an attacker when connecting to an AP, as well as the risks of connecting to open Internet networks in public places, the importance of not using them to transfer personal data, and the need to raise awareness of the importance of protecting wireless networks. The Bettercap tool from the Kali Lunix operating system was used to simulate Address Resolution Protoco (ARP) spoofing attack, Secure Sockets Layer (SSL) Stripping attack, Domain Name System (DNS) spoofing attack, and Denial of Service (DoS) attack, which are based on Man in the Middle (MITM) attacks.

Nanofluids are an innovative class of materials that have been recently developed due to their superior ability to improve thermal performance in industrial applications. It is prepared by dispersing metal particles and metal oxides in the base liquid to ensure proper distribution. Thermal conductivity is a key criterion in developing thermal energy transmission fluids. A literature reviesw was conducted to study the types of nanoparticles used at different concentrations and to review the stability evaluation of nanofluids. Researchers have also suggested adding surfactants such as Colace (docusate sodium), trisodium citrate dihydrate (TSC), and sodium lauryl sulfonate (SLS) to increase stability. The effect of physical and thermal properties was analyzed and the impact of different particle shapes was investigated, where spherical and cylindrical shapes were used because they provide a better heat transfer rate than other shapes. The results show that there is a positive correlation between the concentration of solid nanoparticles, thermal conductivity, and Nusselt number.

This paper discusses a novel approach to control a grid-connected multilevel inverter using Field Programmable Gate Array (FPGA) implemented Fuzzy controlled Coordinate Rotation Digital Computer Phase-Locked Loop PLL (CORDIC-PLL) system for achieving improved inverter performance in injecting active and reactive power into the grid under changing grid conditions of operation. A seven-level cascaded Hbridge inverter topology is used to reduce total harmonic distortion without the need for large filtering components. The CORDIC-PLL system gives fine phase information for synchronization, while the fuzzy logic controller (FLC) makes adjustment in the output of the inverter by changing amplitude factors of carrier signals and phase angles for power injection control. In this way, experimental results depicted the effectiveness of the proposed system with significant improvements related to power quality and system stability.

The manufacturing of long-section pipes is expensive and demands sophisticated procedures, requiring the assembly of smaller pipes by welding. Welding is essential in the oil and gas sectors and several other industries. A welding procedure was executed on two ASTM A106 grade B pipes utilizing GTAW and SMAW techniques for four layers. Testing identified a transverse crack in a welded joint (Test No-01A) at site 20, measuring 4 mm. A PMI test established the chemical composition of the weld joint. A key factor contributing to cracks is the temperature disparity between the weld zone and the base metal. To alleviate this, the joints were warmed prior to welding, The results showed that preheating the metal before the welding process reduces or prevents the occurrence of cracks
by minimizing the temperature gradient between the base metal and the welding temperature. As the PMI test findings revealed discrepancies in the chemical composition of the welded joints prior to and subsequent to welding, attributable to chemical reactions throughout the operation. These findings underscore the influence of temperature and chemical alterations on weld integrity.

In recent years, technology utilizing electroencephalography (EEG) for the control of electrical devices has advanced significantly. Today, anyone may monitor their own brain patterns and waves outside of clinics and medical laboratories. Moreover, in addition to evaluating brain signals, these signals may be utilized to operate daily electronic devices, also referred to as the brain-computer interface (BCI). The suggested system can effectively aid older persons and others with disabilities who experience speech and mobility suffering, enabling them to remain at home safely and comfortably. Ten participants between the ages of 25 and 45 participated in the study. The EEG headset used has a sensor that senses human attention, meditation, and eye-blinking. This paper presents an EEG-based home automation system that uses eye blink detection to control multiple living room devices, achieving an
accuracy between 86.21% and 92.86%. These findings enhance the brain-computer interface and also highlight the potential for improving its immobility.

Obesity is a growing global health concern driven by genetic, behavioral, and environmental factors. Machine learning (ML) offers potential for predicting and classifying obesity, however data accessibility and model scalability present challenges. This study evaluates various machine learning algorithms for obesity prediction, including Random Forest (RF), K-Nearest Neighbors (KNN), Gradient Boosting, and Support Vector Machines (SVM). The dataset comprises 1,610 individuals, considering health, behavioral, and demographic characteristics. The aforementioned metrics were employed to evaluate the model performances; accuracy, precision, recall, and F1-score. Of interest, Logistic Regression had the lowest accuracy score (76.39%), while Gradient Boosting had the highest
(88.82%). Similarly, Gradient Boosting performed well with the other metrics, reinforcing its valid conclusions for obesity classification. There is substantial potential for machine learning, as demonstrated in this study, as it will enable the early detection of obesity, and provide intervention prospects for health management professionals.