ADBU Journal of Electrical and Electronics Engineering (AJEEE)
Not a member yet
73 research outputs found
Sort by
Advancing Communication: A CNN-Powered Framework for Assamese Sign Language Recognition
Get PDFThis paper introduces an advanced method for recognizing Assamese Sign Language (ASL) using deep learning, specifically Convolutional Neural Networks (CNN). The proposed framework encompasses three primary phases: data preparation, model development, and real-time gesture recognition. A dataset of 10 static gestures representing distinct Assamese alphabets was created, utilizing OpenCV for frame capture and MediaPipe for hand landmark detection. The collected data underwent preprocessing, and the hand gesture images were further split into training and testing sets. A CNN model was then trained for gesture classification, yielding a perfect accuracy score of 100%, along with exceptional precision, recall, and F1 scores across all 10 categories. The findings demonstrate the CNN model's proficiency in extracting and learning key features of ASL gestures, enabling the system to accurately recognize and convert the signs into text in real-time. This real-time recognition ability holds significant potential for enhancing communication between the deaf and hearing communities in Assam. Future work will focus on expanding the dataset to incorporate dynamic gestures and optimizing the system for real-time deployment in mobile applications
Spectral Analysis of Musical and Random Noise Signals with Audio Classification: A Comparative Study
Get PDFThis study presents a comparative analysis of two types of audio signals: random noise and music audio. The random noise was recorded in a crowded public bus on an urban road, while the music audio was captured using the same device in a controlled environment, free from external disturbances. Statistical and spectral analyses, including histogram analysis, power spectrum, Fast Fourier Transform (FFT), and Auto-Correlation Function (ACF), were employed to evaluate the signals. The noise signal showed a bell-shaped histogram with samples concentrated around the mean, an ACF that decreased with lag, and the FFT spectrum, which shows that the energy is distributed uniformly across all frequencies. In contrast, the music signal displayed a sharp histogram, distinct voice and music components, rhythmic repetitions, and specific frequencies in its ACF and FFT spectra. Additionally, Support Vector Machine (SVM) classification was performed on both signals. The noise signal was accurately categorized into three noise levels with an accuracy of approximately 82%, while the music audio was categorized into voice, music, and noise components with 90% accuracy. The superior classification performance for music is attributed to the clear separation between voice and music components. This study demonstrates the potential of SVM in audio signal classification, providing a robust approach for distinguishing between noise and music audio
The Role of various types of plasma treatments to improve and enhance the various insulation and dielectric properties of the materials-A Review
Get PDFEnhancing the insulation and dielectric properties may be regarded as one of the major research topics that can be undertaken for study and analysis. When any insulating material is exposed to extreme weather conditions, the degradation of such insulating materials is a very common phenomenon due to variations in temperatures, external weather extremities, wind, dust, and many other naturally occurring phenomena. Thus, the protection of insulating materials has become a vital area of study for researchers and scholars to enhance the lifespan, insulation, and dielectric properties and also to prevent the degradation of such insulating materials due to extreme weather conditions when exposed to an open environment. The various withstanding properties of the materials like breakdown strength, flashover voltage, contact angle, thermal conductivity and various dielectric properties are being studied in this paper. Also, the applications of various non-thermal plasma treatments generated by using various discharge methods like Dielectric Barrier Discharge (DBD), Corona discharge, Microwave discharge and Radiofrequency discharges and the effects of those non-thermal plasma treatments in improving and enhancing the insulation and dielectric properties of the insulators have been extensively studied and reviewed
A Review on Optimization of PV-TEG Hybrid Model
Get PDFThe integration of photovoltaic (PV) panels and thermoelectric generators (TEGs) in a hybrid energy system is a promising development in renewable energy technology. The hybrid system is based on combining the strengths of both PV and TEG technologies to maximize energy production and efficiency. While PV panels are efficient in converting sunlight into electricity, their performance often suffers due to heat losses, especially at high temperatures. The use of TEGs addresses this challenge through the capture of waste heat produced by the PV panels and its conversion into further electrical energy by the Seebeck effect, thereby increasing the overall energy output. In this review, a systematic literature analysis has been conducted through various case studies, and simulation-based research relevant to PV-TEG hybrid systems. These were assessed based on performance metrics, design configurations, material innovations, and thermal management techniques. The review methodology includes a structured analysis of experimental studies, and numerical simulations. Sources have been selected based on relevance to system performance, material properties, and design innovations. This analysis highlights that PV-TEG hybrid systems can enhance overall energy conversion efficiency by 10–20% in optimized configurations. It is seen that power output improved with bismuth telluride-based modules, enhanced cooling using heat sinks and phase change materials, and better power tracking using intelligent MPPT algorithms. Thus, it is crucial to pay more attention to thermal management, material selection, and system design in order to optimize PV-TEG hybrid systems. Proper thermal management enhances TEG performance, and there are efficiency gains concerning the PV panels under all environments. Advanced heat sink design with high-performance thermal interface materials can significantly enhance both energy conversion efficiency and heat dissipation. It is observed that PV-TEG systems can reduce reliance on conventional energy sources, offering more sustainable and efficient alternatives to power generation. Challenges with material compatibility, cost, and scalability remain major barriers to the practical implementation of these systems, and continued research into finding cost-effective materials and innovative system designs will be necessary. In future studies, the focus should be put on the discovery of new thermoelectric materials and scaling up the techniques involved in system integration. PV-TEG hybrid systems would then be a far better choice for the world's global energy crisis
Application of Linear Programming for Optimal Allocation of a Diesel- PV Hybrid Power System
Get PDFTo address escalating electricity prices and more frequent power cuts in developing countries, hybrid energy systems, including solar and diesel generation, are being installed to ensure reliability and reduce costs. Linear programming can size hybrid systems for operation and cost-effectiveness. Linear programming helps determine the optimal dispatch strategy to improve system reliability, reduce emissions, and lower the associated costs in hybrid generation networks through the formative articulation of objective functions and constraints. The work presented in this paper shows the application of linear programming as an efficient optimization tool for solving an allocation problem involving a hybrid power system problem. Eight load points will be allocated to eight hybrid power sources involving diesel generators and a solar photovoltaic (PV) system. By solving the linear programming, it is shown that the optimal allocation of the power to the load can be done using diesel generators alone or using hybrid power generators, i.e., by using solar power generators along with diesel generators. The allocation algorithm can efficiently allocate the hybrid power generators to the load, thereby helping mitigate power loss and ill effects of pollution from diesel generators
Sentiment Analysis of Assamese Text Reviews: Supervised Machine Learning Approach with Combined n-gram and TF-IDF Feature
Get PDFSentiment analysis (SA) is a challenging application of natural language processing (NLP) in various Indian languages. However, there is limited research on sentiment categorization in Assamese texts. This paper investigates sentiment categorization on Assamese textual data using a dataset created by translating Bengali resources into Assamese using Google Translator. The study employs multiple supervised ML methods, including Decision Tree, K-nearest neighbour, Multinomial Naive Bayes, Logistic Regression, and Support Vector Machine, combined with n-gram and Term Frequency-Inverse Document Frequency (TF-IDF) feature extraction methods. The experimental results show that Multinomial Naive Bayes and Support Vector Machine have over 80% accuracy in analyzing sentiments in Assamese texts, while the Unigram model performs better than higher-order n-gram models in both datasets. The proposed model is shown to be an effective tool for sentiment classification in domain-independent Assamese text data
A review on Day-Ahead Solar Energy Prediction
Get PDFAccurate day-ahead prediction of solar energy plays a vital role in the planning of supply and demand in a power grid system. The previous study shows predictions based on weather forecasts composed of numerical text data. They can reflect temporal factors therefore the data versus the result might not always give the most accurate and precise results. That is why incorporating different methods and techniques which enhance accuracy is an important topic. An in-depth review of current deep learning-based forecasting models for renewable energy is provided in this paper
Voltage Stability Analysis of Electrical Transmission System Using Reactive Power Sensitivity Indicator
Get PDFNowadays the biggest challenge of the electrical power transmission system is voltage collapse. As a result, these days a major outlook has been paid with the aid of a variety of research on voltage stability. In this paper, a method has been introduced to determine the voltage stability of an IEEE 14-bus power system. This technique is based on Reactive Power Sensitivity Indicator. Using this indicator, weak buses are identified among the 14 buses of the system under study. Newton-Raphson method of Load Flow Analysis is coded in MATLAB programming to find out different parameters of the IEEE 14-bus system and used for stability analysis. A FACTS device has been installed in the weakest bus to enhance the voltage stability of the network. The results exhibit the effectiveness of the proposed technique
Design and Implementation of a Microcontroller-based Automatic Temperature-sensing Relay
Get PDFA temperature-based relay is designed using a microcontroller, which can be programmed easily for changed settings. The system is built around an Arduino Uno microcontroller that is connected to relays, an LED, and an LM35 temperature sensor. The temperature sensor provides an electrical output corresponding to the temperature in °C and can be used to measure temperature. To determine the temperature of a given room, the Arduino Uno microcontroller decodes the temperature reading from the temperature sensor and compares it to a temperature value that has been set through the program. The microprocessor then automatically switches the relay ON or OFF depending on the comparison algorithm, which is indicated by an LED turning ON or OFF. The primary goal of this work was to design and build an automatic temperature-controlled relay, based on a microcontroller. It has been accomplished with high accuracy. The relay tested for different temperature conditions shows a wide variability in the range of operation. This experimental work serves as an example of how embedded systems are used in electrical device design
A Survey on Coordinated Charging Methods for Electric Vehicles
Get PDFElectric vehicles (EVs) is regarded as one of the most effective ways to reduce oil and gas use. EVs (electric vehicles) have many advantages over ICEVs (internal combustion engine vehicles), including zero pollution, little noise, and exceptional energy efficiency. Even though an EV is known to have a three times higher fuel efficiency than an ICEV, the driving range is often significantly lower because batteries have a lower energy density than gasoline or diesel. Over the next few decades, it is anticipated that the number of electric vehicles will increase significantly due to concerns about pollution and technological advancements in the sector. Utilizing a variety of energy sources will boost energy security while reducing emissions and fuel usage. A paradigm shift has been observed with the switch from internal combustion to electric car technology. For electric vehicles to become widely used, a charging infrastructure must be developed. However, there is a cap on the amount of electricity that can be used to charge the vehicles in a charging station. Rearranging charging times, specifically charging coordination can help optimize the distribution of the available power among the vehicles. In this paper, a review of the various coordinated charging methods has been presented. A detailed comparison of the methods has been done