Indonesian Journal of Electrical Engineering and Computer Science
Not a member yet
9109 research outputs found
Sort by
Transformer oil degradation detection system based on color scale analysis
Get PDFThe rise in power transformer load results in degradation of the condition of the transformer oil and ultimately a deficiency in the distribution of electrical energy. This degradation can be slowed down by reconditioning transformer oil based on oil color detection. This research aims to design, test and validate a transformer oil color testing system based on color sensor and microcontroller. To obtain an accurate system, tests were carried out on selecting the types of sensors, the color of the chamber walls, and the shapes of transformer oil sample vessel used. The oil color scale of the samples was determined visually according to the ASTMD1500, 2009 standard as a benchmark. The test results showed that the TCS3200 color sensor was able to detect the color of all transformer oil samples. White chamber wall and test tube as oil sample containers were chosen to increase system accuracy. Overall, the system is able to detect the color of transformer oil, convert to the ASTMD1500, 2009 standard transformer oil color scale, determine the condition of the transformer oil and conclude the level of transformer oil degradation according to CIGRE-761, 2019. Validation results showed the system had an accuracy level of 92.65%
Convolutional neural network-based strategies for efficient content-based image retrieval
Get PDFRecent years have seen a meteoric rise in the usage of enormous image databases due to advancements in multimedia technologies. One of the most critical technologies for image processing nowadays is image retrieval. This study uses convolutional neural networks (CNNs) for content-based image retrieval (CBIR). With the ever-growing number of digital photos, practical methods for retrieving these images are crucial. CNNs are incredibly efficient in many computer vision applications. Improving the efficacy and precision of image retrieval systems is the primary goal of our research into using deep learning. The paper starts with a thorough analysis of the current state of CBIR methods and the difficulties they face. Afterwards, it explores CNN’s design and operation, focusing on CNN’s capacity to learn hierarchical features from images autonomously. This paper also looks at how the model performs when it alters its hyperparameters, transfer learning techniques, and CNN topologies. The insights obtained from these experiments enhance the comprehension of the elements impacting CNN effectiveness in CBIR. Finally, our study shows that CNNs can change the game for image search by transforming CBIR systems. This research adds to the expanding body of information about using cutting-edge deep learning algorithms to make image retrieval more efficient and accurate
Blockchain and smart contracts based system for criminal record management
Get PDFReducing crime rate in a country is the most important concern of developing robust systems to automate the criminal record-obtaining process. Generally, the criminal record is managed manually, which makes the information collection from other criminal records very difficult. Therefore, investigations that could be carried out using criminal records to understand the purpose of crime and countering it are outdated. However, the integrity, security, and traceability of data exchange, especially for the judicial sector are the most frequent issues faced by information systems of public organizations. In this paper, we present a study of using blockchain technology and smart contracts to design a new architecture for a decentralized system to manage criminal record storage. This proposed architecture automates the process of getting a criminal record by moving past the techniques employed in developing traditional systems of data management such as centralized systems. In this study, blockchain technology is used to ensure data security, integrity, and traceability as well as ensure timely access to criminal records, and smart contracts are used to allow traceability and authenticity. This architecture will significantly reduce the impact of corruption in law enforcement by eliminating fraud cases, which will revolutionize E-governance in the Moroccan country
Challenges of implementing protection systems in smart grids: a review
Get PDFBased on the emergence of increasingly advanced technology, the conventional power grid can be upgraded to a smart grid by adding bidirectional communication, computer algorithms, and equipment that uses artificial intelligence (AI). A smart grid is a revolution in the current electricity network that can control the two-way generation and transmission process by utilizing an intelligent system so that the distribution of electric power can be handled optimally and in real time. The challenge of the smart grid is that there are distributed generators and microgrids that must be controlled in real time with rapidly changing loads. To meet these criteria, several points are proposed, i.e., finding an effective procedure to construct self-healing capability; developing a protection system based on AI; and proposing a systematic procedure to realize self-healing and protection systems with the help of a multi-agent system (MAS). Multi-agent systems are one of the AI approaches. Each agent can work independently and can also communicate with one another and with other devices on the network. Agents used as models can be classified into several categories, such as grid component agents, distributed resource agents, end-user agents, failure control agents, data analysis agents, and graphical visualization agents
Digital afterlife: challenges and technological innovations in pursuit of immortality
Get PDFDigital immortality, the idea of endless life and ultimate happiness in a virtual afterlife, has become a subject of human fascination. This article reports the results of a comprehensive research project focused on identifying the challenges and potential options related to digital immortality. Analyzing 39 relevant studies, our research concentrates on two main themes: the barriers to achieve the digital immortality and the tools created to preserve digital memories. Our findings reveal that the challenges associated with digital immortality are deeply rooted in legal, ethical, and social issues. Importantly, our focus is the challenges related to digital content left by the deceased, its collection method, and integrity in digital immortality research, as content forms the basis for achieving this objective. Furthermore, the research highlights the need for more advanced technology, as the number of studies is limited and current progress is primarily future-oriented. However, our analysis demonstrates that the digital content left by the deceased is paramount, as it constitutes the raw material for achieving digital immortality
TALOS: optimization of the CNN for the detection of the tomato leaf diseases
Get PDFEarly detection of plant diseases using convolutional neural network (CNN)is crucial for maximizing crop yield and minimizing economic losses. Manual inspection, the frequent technique, is inefficient and error prone. While CNN’s offer potential for accurate and quick disease recognition, their performance is highly dependent on effective hyperparameter tuning. This process is time consuming, resource intensive, and needs significant expertise due to the vast hyperparameter space, since it can be hard to figure out which is ideal for optimal performance. An effective optimization tool, tunable automated hyperparameter learning optimization system (TALOS), is proposed, which automates the tuning of hyperparameters by systematically exploring the hyperparameter space and evaluates different combinations of parameters to find the optimal configuration that maximize the model’s performance. The performance of this approach is recognizable through its exploration of five different hyperparameters across a search space of 32 combinations, yielding optimal parameters by the second round. Using 3030 tomato leaf images from a benchmark data set, the model achieves a remarkable 94.7% validation accuracy with 33647 trainable parameters. Thus, automated hyperparameter tuning approach not only optimizes model performance but also reduces manual effort and resource requirements, paving the way for more effective and scalable solutions in agricultural technology
A novel dataset and part-of-speech tagging approach for enhancing sentiment analysis in Kannada
Get PDFThe problem addressed in this research is the limited availability of labelled datasets and effective sentiment analysis tools for the Kannada language. Existing challenges include linguistic variations, cultural diversities, and the absence of comprehensive datasets designed specifically for sentiment analysis in Kannada. This research aims to enhance sentiment analysis capabilities for the Kannada language, addressing challenges posed by linguistic variations and limited labelled datasets. A novel Kannada dataset derived from SemEval 2014 task 4 was created using a conversion process. The dataset was processed using part-of-speech tagging, and a specialized model called K-BERT (Kannada bidirectional encoder representations from transformers) was introduced and implemented using Python within the Anaconda environment. Performance evaluation results showcased K-BERT's superiority over traditional machine learning (ML) algorithms and the BERT model, achieving an accuracy of 0.98, precision of 0.97, recall of 0.97, and F-score of 0.98 in sentiment classification for Kannada text data. This work contributes a unique Kannada dataset, introduces the K-BERT model specifically designed for Kannada sentiment analysis, and emphasizes the importance of collaborative efforts in advancing natural language processing (NLP) research for multilingual environments
Engraved hexagonal metamaterials resonators antenna for bio-implantable ISM-band applic
Get PDFThis study will introduce a metamaterial antenna designing for use in biomedical implants. The antenna is compact and utilizes four slot complementary metamaterial hexagonal resonators of uniform shape and size. By incorporating the metamaterial into the antenna design, its size is reduced while the performance is enhanced. Simulation results show that the antenna achieves satisfactory peak gain values of -22.6 dBi and a 34.5% increase in bandwidth. Operating within the 2.4-2.5 GHz industrial, scientific, and medical (ISM) frequency bands, the antenna measures 7×7×1.27 mm3 and consists of substrate layers with patch radiation, four metamaterials hexagonal resonators on the upper surface, a ground layer, and a second superstrate layer. The study also addresses the challenges and problems associated with the interaction between the antenna and human tissue, while aiming to maintain antenna performance, properties, and minimize its impact on tissues. Evaluation of when using a 2.45 GHz operating frequency, the specific absorption rate (SAR) shows values of 489.87 W/kg for 1 g of averaged tissue and 53.738 W/kg for 10 g of averaged tissue. The results of placing the antenna in human skin tissue are safe for use in the human body and appropriate for biomedical applications. Simulations conducted using computer simulation technology (CST) and high frequency structure simulator (HFSS) software emphasize the excellent performance of the engraved metamaterial antenna
Novel five-patch compact microstrip Yagi-alike antenna for Ka-band applications
Get PDFThis paper discusses the process of designing and fabricating a novel compact microstrip patch Yagi-like antenna having five-patch radiating element at operating frequency 31 GHz with a bandwidth of 1 GHz. The developed design aims to optimize the antenna performance. The overall dimension of the antenna being 17× 14 × 0.8 mm3, based on RT Duroid 5880 substrate having dielectric loss tangent of 0.0009 and relative permittivity 2.2. The effectiveness of the performance of proposed design was evaluated using the electromagnetic solver Ansoft high-frequency structure simulator (HFSS) and validated by the laboratory measurements on the antenna prototype. The measured results are consistent with the simulation prediction. The designed antenna achieved directional radiation and the performances with voltage standing wave ratio (VSWR) < 1.32, return loss -17 dB and gain of 6 dBi. The measured results are compared with those existing in literature. The proposed antenna design has proven very effective in terms of the intended design and parameters which make it suitable for satellite application and wireless communication
Enhancing data cleaning process on accounting data for fraud detection
Get PDFData cleaning is a crucial step in fraud detection as it involves identifying and correcting any inaccuracies or inconsistencies in the data. This can help to ensure that the data being used for fraud detection is reliable and accurate, which in turn can improve the effectiveness of fraud detection algorithms. Due to the overwhelming amount of data, data cleaning specific for fraud detection is a very important activity for the auditor to find the appropriate information. Therefore, a new accounting data cleaning for fraud detection is needed. In this paper, an enhancement of the process of fraud detection by accounting auditors through the implementation of accounting data cleaning technique is proposed. The proposed technique was embedded in a prototype system called accounting data cleaning for fraud detection (ADCFD). Through experiment, the performance of the proposed technique through ADCF is compared with those obtained from the IDEA system, using the same dataset. The results show that the proposed enhanced technique through ADCFD system performed better than the IDEA system