ECTI Transactions on Electrical Engineering, Electronics, and Communications

A Recent Comprehensive Review of The Research Challenges and Opportunities of Automatic Generation

Mesfin, Chandra — Wed, 03 Apr 2024 00:00:00 +0700

The main goal of modern power systems is to balance the power generated at the end of the generation with the demand for power on the consumer side. However, the consumer’s development needs are constantly shifting load demand. Therefore, it is necessary to monitor and meet the load demand continuously. Customers gain from receiving cheaper electricity, more alternatives, and better services as the power system problem approaches the deregulation paradigm. Automatic Generation Con- trol is presented in this thorough literature review (AGC). The current AGC designs for power systems are analyzed and categorized in this study for present-day and foreseeable intelligent power systems. Then, various control techniques- including classic control, optimal control, artificial intelligence control, adaptive and self- tuning control, and other optimization techniques, are discussed for the modeling of AGC in regulated and deregulated environment contexts. Multiple control groups are also created from the proposed AGC control mechanisms and examined. The paper’s conclusion lists numerous new AGC research directions and gaps.

Comparison of Deep Learning and Incremental Learning Model for Net Load Forecasting

Charnon Chupong, Nitikorn Junhuathon, Sirichai Dangeam, Boonyang Plangklang — Thu, 29 Feb 2024 00:00:00 +0700

This paper presents hourly net load forecasting, which is the forecasting of the difference between the hourly demand and the hourly power produced from the Photovoltaic (PV) system, which is the load that the utility should supply to the consumer. By comparing the forecasting of the 3 models, 1) Long Short-Term Memory (LSTM), which is a deep learning model, 2) Fully Online Sequential Extreme Learning Machine (FOS-ELM), which is an incremental learning model that does not require initial training data and 3) Online Sequential Extreme Learning Machine (OS-ELM), a model that can be incrementally learned as FOS-ELM. In addition, we proposed the initial training method for the OS-ELM model by taking the first sample obtained from working to synthesize a sufficient amount of sample for the initial training of the OS-ELM model. It was found from the experiment that in the case of fixed PV penetration rate, the LSTM model had slightly lower of error in forecasting than the other two models. In the case of increasing PV penetration rate, the FOS-ELM, and OS-ELM models, with incremental learning capacity, had significantly lower errors in forecasting than the LSTM model. When comparing only the OS-ELM model using the proposed method with the FOS-ELM model, it was found that the OS-ELM model gave lower errors in forecasting than the FOS-ELM model because it was initially trained by the synthetic sample properly,

Multi-Resolution Analysis Features followed by Facial Part Detection for Face Recognition

Rashid , Haider J. Abd — Wed, 03 Apr 2024 00:00:00 +0700

Face recognition is considered the main physiological and behavioral biometrics due to the following advantages; simplicity of the face capturing, feature uniqueness and distinctness, and availability of the image acquisition devices. In this paper, a new approach to face recognition system (FRS) utilizing the Two-Dimensional Discrete Multi-Wavelets Transform (2D DMWT) followed by Vector Quantization (VQ) to the detected face and facial parts (DF and DFP) is proposed. Faces and facial parts (Nose, Mouth, Left-Right Eyes) are detected in the preprocessing step. Face and facials are the main parts that represent each person in the feature extraction step. For dimensionality reduction and features selection, the 1-level of 2D DMWT decomposition is employed in the two representations. For each person in the second representation, four groups are constructed using the training poses, each group for each facial part. Furthermore, VQ and Kekre Fast Codebook Generation (KFCG) are applied to the detected faces and the four groups derived from the first and second representations, respectively. The Euclidean distance is utilized in the classification phase. Four databases, namely, YALE, FERET, FEI, and Georgia Tech. are used to test the FRS. These databases have different facial diversity, such as pose rotation, light condition, expressions, etc. Kfold Cross-Validation (CV) is utilized to analyze the experimental results. The proposed system improves the recognition rates and the storage requirement compared to the state-of-the-art approaches.

Single Phase Multilevel Inverter Based on Single Carrier Pulse Width Modulation Algorithm

Essam Hendawi — Thu, 29 Feb 2024 00:00:00 +0700

High frequency pulse width modulation techniques play very important role in multilevel inverter MLI applications where those techniques overcome the problems associated with their low frequency modulation counterpart techniques. Multicarrier sinusoidal pulse width modulation (MC-SPWM) is a common technique for achieving high frequency modulation and control in MLIs. The complexity, high cost, synchronization are very important issues when implementing MC-SPWM in real time. This paper presents a simple, high effective algorithm to overcome the problems and difficulties of MC-SPWM. The proposed algorithm employs a unique carrier signal instead of multicarrier waveforms thus significantly reduce the complexity and cost as well. The algorithm is applied to 15-level, 7-level and 13-level multilevel inverters to prove its effectiveness. Simulation and experimental validation of the system verify the high performance of the proposed algorithm.

Brain MRI and CT Image Fusion Using Multiscale Local Extrema and Image Statistics

Mocherla Venkata Srikanth, A.Suneel Kumar, B.Nagasirisha, T.Venkata Lakshmi — Thu, 29 Feb 2024 00:00:00 +0700

In medical applications such as radiotherapy and guided-image surgery, data fusion for diagnostic imaging has emerged as a critical issue. Because the objective of medical image fusion is to improve patient diagnosis accuracy, the fused image is created by preserving the source images' prominent details and features. It has been demonstrated that the Multi-Level Local Extrema representation has numerous advantages over traditional image modeling approaches. We propose an innovative MLE-based fusion method for multimodal medical images in this paper. In the MLE schema, the proposed algorithm decomposes source images into coarse and detailed layers, then fuses the source images using weights calculated from these detail images using image statistics. We visually and quantitatively compared the efficacy of the suggested approach to that of existing methods using five different types of medical images from various sources. The experimental results showed that the proposed scheme outperforms other current typical schemes in terms of both qualitative image quality and objective evaluation.

Enhancing Fronthaul Network Connectivity by utilizing a RoF-WDM Structure with MM Wave Transmission

abdennour fellag chebra, Ahmed Riad Borsali, Mehdi Rouissat — Thu, 29 Feb 2024 00:00:00 +0700

The future 5G wireless system will provide communication systems with a new advanced features to cover a large area with providing high bandwidth Millimeter wave (MM Wave) processing speed, in which case transmission technologies become more important and require large amounts of data, a large number of channels and lower cost. This paper reports the design of MM wave optical generation with 60 GHz based on RoF-WDM technique for long-distance optical fiber. The bloc scheme consists of 64 channels generated using Dual-Parallel Mach-Zehnder Modulator (DP-MZM) modulators for high data rate optical transmission. The performance was evaluated and analyzed in terms of various parameters such as optical fiber distance, input power and data rate, the simulation results are reported using Bit Error Rate (BER), Q-factor, Optical Signal-to-Noise Ratio (OSNR), and Eye Diagrams. The System efficiency provides an average BER of 4.0309e-10 with optical fiber link of 120 km and 10 Gbps data rate per channel, it also provides 18 Gbps per channel for 100 km of Standard Single Mode Fiber (SSMF). In this work, the integration of different techniques is viewed as a unique perspective of radio over fiber systems towards a wireless communication network.

Internet of Things Network Performance: Impact of Message and Client sizes and Reliability Levels

Jiby J Puthiyidam, Shelbi Joseph — Thu, 29 Feb 2024 00:00:00 +0700

Consistent and timely message transmission is pivotal in the success of Internet of Things (IoT) communications. IoT applications typically involve resourceconstrained devices with limited processing power, memory capacity, and battery life. These characteristics necessitate the use of specialized communication protocols for IoT applications. Message Queue Telemetry Transport (MQTT) is a widely adopted protocol within the IoT ecosystem. Ensuring the appropriate message size and reliability level is crucial for achieving successful MQTT communication. This article empirically assesses MQTT protocol performance across various Quality of Service (QoS) levels while considering different message and client sizes. Initially, we conducted a comparative analysis of message transfer performance between HTTP and MQTT protocols. Subsequently, we delved into the MQTT protocol’s performance across three distinct configurations: one publisher-one subscriber, multiple publishers-one subscriber, and multiple publishersmultiple subscribers. We also examined how message and client sizes impact message transfer latency. When employing a 100-byte payload, we observed that the time delay in a network comprising 150 clients is 60.71% greater than in a network with 50 clients. Similarly, with a message size of 2500 bytes, a network with 50 clients requires 96% less time to deliver than a network with 150 clients.

Development of an On-Grid Low-Voltage Battery Energy Storage System with Balancing Dual-DC-Voltage Quasi Single-Stage Converter

Yuttana Kumsuwan (Journal editor); Neerakorn Jarutus — Thu, 29 Feb 2024 00:00:00 +0700

With low-voltage (LV) battery energy storage systems (BESSs), the quasi single-stage converters (QSSCs) are utilized to reduce power consumption in two-stage conversion systems. Despite a good waveform quality of applying multilevel converters, the unbalancing voltage problems is possible to be contributed, such as decrease in grid quality and complexity in pulse-width-modulation (PWM). In this paper, it is the main challenge to solve these problems for developing the on-grid LV BESS with the QSSC. Balancing voltage control is proposed to control the dc-dc converter. As a result, the dual dc voltages are equalized. The system waveform quality and power quality can be improved. The PWM stage is then symmetrical and simplified. For more simplicity, the two-inphase carrier-based PWM is applied. The continuous non-sinusoidal scheme is preferred to not only provide the full dc utilization, but also improve the grid waveform quality in addition to the improvement from the proposed control method. Besides, the powers are simply controlled using the basic of the current vector control and decoupling from the balancing voltage control, resulting in the grid-connected enhancement without affecting the waveform quality. A 10-kW-microgrid-scale BESS is employed to validate the feasibility of the proposed system and its superiority over conventional systems.

The Performance of Organic Field Effect Transistor Affected by Different Thickness of Active Semiconductor Layer and Thickness of Gate Insulator

Fadliondi Fadliondi, Budiyanto Budiyanto, Prian Gagani Chamdareno — Thu, 29 Feb 2024 00:00:00 +0700

The purpose of this paper is to fabricate organic field effect transistor and to investigate the effect of the thickness of the pentacene active layer and the thickness of gate insulator layer on MOSFET performance. The fabricated structure is top-contact. When the thickness of the insulator gate layer increases from 10 nm to 30 nm, the magnitude of the drain source current, when V_GS = V_DS = -4 V, decreases from 1813 nA to 214 nA and then the threshold voltage shifts from -1.4 V to -2.4 V. When the thickness of the pentacene increases from 9 nm to 40 nm, the threshold voltage voltage shifts slightly in the negative direction from -1.4 V to -1.6 V for SiO₂ thickness of 10 nm. In case of SiO₂ thickness of 20 nm, the threshold voltage voltage shifts from -1.9 V to -2.2 V. In case of SiO₂ thickness of 30 nm, the threshold voltage voltage shifts from -2.4 V to -2.9 V. Besides that, the mobility decreases from around 0.31 cm²/(Vs) to 0.15 cm²/(Vs) when the pentacene thickness increases from 9 nm to 40 nm.

Control of a Semi-closed Plant Factory with Artificial Lighting-based on Two Different LED with NB-IoT

Thu, 29 Feb 2024 00:00:00 +0700

Plant factory artificial light (PFAL) is an effective technique for producing large amounts of crops per area and high-quality plant growth. This work aims to construct a semi-closed PFAL growth system based on NB-IoT using two types of LED arrays: phosphor-converted LED (pc-LED) and RB-LED. Next, while examining the features of the artificial light spectrum, compare the Curry leaf kale and Chinese kale in seedlings under various LED light sources. An NB-IoT module with the MAGELLAN platform monitored and controlled the temperature, humidity, and illumination of the semi-closed PFAL growing system. The results indicate that cos lettuce cultivated with PCLEDs is likely more photosynthesis-capable than cos lettuce grown with RBLEDs. Compared to RB-LED, the average fresh weight of the cos lettuce from PC-LED was significantly higher. The data gathered from the cloud system under the MAGELLAN platform during the 7-day trial, the control of lighting and watering in the semi-closed PFAL system, and the measurement results of environmental factors were all accurately completed. Organic veggies could be grown in a home or school using the semi-PFAL growing technique.

Selection of Lightweight CNN Models with Limited Computing Resources for Drone Collision Prediction

Rifqi Nabila Zufar, David Banjerdpongchai — Thu, 29 Feb 2024 00:00:00 +0700

The Collision Avoidance System (CAS) is a safety system created to identify and prevent collisions, primarily on drones. The CAS comprises three processes: detection, prediction, and action. The predictive process is crucial as it determines whether a collision will occur, making it the core component of the system. Most drones are equipped with cameras. A visual-based prediction involves the use of a convolutional neural network (CNN). The CNN operates by autonomously learning and extracting hierarchical characteristics from input data through convolution, pooling, and fully connected layers. Currently, there are CNN models called pretrained models that are ready to use. However, not all pretrained models are suitable for compatibility with drones as they possess computational constraints. Our objective is to establish a suitable model selection from a variety of pre-trained CNN models with lightweight architectures. The transfer learning technique is applied to customize these models with the ColANet dataset. Subsequently, we evaluate these models regarding their accuracy, model size, inference time, and power consumption. Finally, the selected model is deployed in real time on a Raspberry Pi 3B+ with data input from a DJI Tello drone camera, and the prediction performance is evaluated.

Topology Design for Cellular IoT: From ILP to ML Perspective

Nakrop Jinaporn — Thu, 29 Feb 2024 00:00:00 +0700

Due to the emerging deployment of cellular IoT, a network topology design appears to be one of the greatest challenges faced by mobile network operators, that is, both the capacity maximization and the overall network cost minimization have been considered as the objective of network planning. In this article, the topology design for cellular IoT is divided into two subproblems: gateway location and gateway connection problems. They are formulated as the integer linear programming problem. For the former subproblem, the best gateway locations and the optimal network cost can be obtained by the optimization approach to form multiple local networks. For the latter subproblem, a connection of selected gateways with the minimum connection cost can be presented by the Kruskal algorithm to form a backbone-like network. This results in a two-layered network with the minimum network cost. According to the results, a significant reduction in the network cost could be obtained with the optimal setting of system parameters. In addition to the optimization approach, the gateway location problem is examined by means of clustering algorithms. The fair gateway placement can be obtained by K-medoids clustering without the time complexity.