Martin College

<span lang="EN-US">A wireless power transfer system designed for electro-vehicle recharge and low-power device charging is explained in this document through resonant inductive coupling technology. Once switched on the pulse generator and IRF540 MOSFETs from the IC CD4047 drive high-frequency signals through the transmitter coil. IR sensors function as operational safety tools by detecting valid receivers which activate a relay control system for transmitter power management and reduce unnecessary energy consumption. A full-wave rectifier along with the 7805-voltage regulator enables the receiver unit to deliver fully stable 5 V DC output. System status is displayed through a user interface equipped with an LCD and real-time billing information runs on ThingSpeak IoT platform for visualization. Tests show that the system reaches a maximum power transfer efficiency of 90% alongside successful relay operation lasting less than 150 ms. The system provides an inexpensive solution to build smart wireless charging infrastructure networks that remain energy-efficient and expandable through its built-in control and monitoring functions.</span>

Post-Acute Sequelae of SARS-CoV-2 (PASC) is posing an extreme risk environment with severe consequences, especially for middle-aged and older adults and those with chronic health conditions such as cardiovascular diseases, cancers, respiratory illnesses, and diabetes. Physical disorders include severe and persistent body pains, fatigue, and difficulty with body movements. Similarly, mood disorders such as depression, mood swings, feelings of hopelessness, and difficulty concentrating are significant predictors of PASC. Individuals, predominantly middle-aged and elderly people, are facing increased physical and mental consequences, including frequent hospitalizations, medically unstable conditions, and in some cases, unexpected fatalities. Timely identification of PASC is essential to mitigating the severity of associated health issues. This research recommends a latent transfer model to integrate patient data from different regions, to extract insights into the data, and to deliver personalized healthcare solutions. This study features the potential of the latent transfer learning model to improve data simplification, generalization, personalization, insight detection, variability, scalability, and adaptability to improve public health outcomes.

The vehicle-to-grid (V2G) concept has gained significant attention in the last decade due to its potential to enhance direct current (DC) microgrid stability and reliability. Electric vehicles (EVs) play a central role in distributed energy storage systems, optimizing efficiency and enabling the integration of renewable energy sources. This study offers a unique three level CLLC resonant converter developed for off-board EV chargers to promote bidirectional power transfer between DC microgrids and EVs. The suggested converter uses resonant CLLC components and two three-level full bridges to effectively handle a broad range of EV battery voltages (200 V–700 V). To ensure effective power conversion, the first harmonic approximation (FHA) model is used to analyse the converter's resonant frequency characteristics. The proposed system achieves high efficiency (>95%), with voltage stability maintained at 750 V under various load conditions. The converter's performance was validated through MATLAB based simulations, comparing proportional integral (PI) and proportional integral derivative (PID) control strategies. The PID-controlled system demonstrated superior dynamic response, reduced current ripples, and enhanced voltage regulation compared to the PI-controlled system. This study demonstrates the viability of implementing a three-level CLLC resonant converter for efficient, bidirectional, and wide-voltage adaptation in EV charging infrastructure, thereby contributing to grid stability and renewable energy integration.

Traditional construction techniques have been altered by the incorporation of artificial intelligence (AI) within the software engineering process, necessitating the use of cutting-edge tools and methods to handle the particular difficulties presented by projects utilizing AI. Agile methodologies, known for their adaptability and iterative nature, have emerged as an ideal framework for managing the complexity and uncertainty inherent in AI systems. This paper explores the intersection of Agile software engineering and AI, highlighting strategies for effective collaboration between cross-functional teams, managing data-driven workflows, and integrating AI-specific tools into the development lifecycle. Key topics include iterative model training, AI networks for continuous development and installation, ethical considerations, and leveraging automation to enhance productivity. Case studies and real-world examples demonstrate the successful application of Agile practices in AI projects, offering actionable insights for practitioners. By bridging Agile principles with AI development needs, this paper provides a comprehensive guide for dealing with the quickly changing field of software design powered by AI.

Sentiment analysis expert systems have gained intensive market attention because of the vast increase in customer-generated online feedback through reviews and social media posts and feedback. The study examines basic methods of sentiment analysis which consist of lexicon-based approaches as well as machine learning and deep learning systems. The paper evaluates practical uses of sentiment analysis along with discussing current field obstacles and prospective research paths for the upcoming years. New NLP technology particularly using transformer models improves sentiment analysis systems while enhancing their precision and speed.

Numerous facets of the public realm could be revolutionised by the convergence of the Internet of Things (the Internet of Things) and Artificial Intelligence (AI). This project developed a based-on data predictive maintenance system for manufacturing production lines. One of these anticipated improvements that can be included into the factory's production system is “AI of Things (AIoT),” which refers to the use of the IoT to carry out intelligent tasks with the aid of AI integration. By using the information created by IoT sensors progressively, the system utilizes machine learning (ML), a part of AI, to recognize expected blunders preceding they happen. The viability of the innovation was surveyed utilizing IoT information from genuine assembling frameworks, and the results of the appraisal demonstrated that the prescient support structure was effective in distinguishing potential failure indicators and could assist with forestalling some production stops. Comparing the performance of ML techniques revealed that frameworks of the bagging ensemble technique Random Forest and the boosting technique XGBoost appeared to outperform each of the techniques in the assessment. The factory's production system now uses the best-performing ML techniques from the present research.

The Leakage Inductance and Series Capacitance (LLSC) resonant converters are exhaustively employed in a wide assortment of industries involving consumer electronics due to their benefits of good efficiency, higher power density, immunity to electromagnetic interference, low EMI and harmonic distortion, wide production extends, voltage stress is lowered and frequency at high operating characteristic. Three of the most prominent converters with LLSC topologies are explored along with thorough analyses of their merits and disadvantages. The background of LLSC resonant converters are also discussed. A significant amount of research is also being done regarding large - scale production of LLSC resonant converters, namely in order to charge electric automobiles (EVs), solar systems, LED lighting drivers, and power supply for LCD TVs. Eventually, the growth of LLSC resonant converter is explained.

In this research, we investigate the problem of automatically predicting and classifying stock prices, with an eye towards creating and testing a Hybrid AI Feature Selection Method. This work employs a fictitious dataset to offer a new method that integrates Genetic Algorithm (GA) and Recursive Feature Elimination (RFE) to isolate the most important characteristics for predicting stock prices and classifying market fluctuations. The findings demonstrate that the hybrid strategy is effective in reducing the complexity of features and greatly improving model performance over more conventional methods. Furthermore, a simulation of a trading strategy based on the categorization findings reveals its potential to produce more efficient and successful investment methods, highlighting the practical relevance of this study. This research contributes to the developing field of financial technology by laying the groundwork for a new way of thinking about financial prediction and decision making, giving professionals and investors access to cutting-edge resources that can help them make better, more profitable choices.

One of the main ways to evaluate information is through visual learning, which traditionally combines textual reading with visual aids like graphical representations. However, separating the visual method of learning to distinguish between text and visuals has been suggested in light of advancements in learning styles. The ability to swiftly create complex visual information and inexpensively distribute it digitally is another way that technology has improved visual presentation. The purpose of this research paper is to conduct a detailed investigation of the Virtualization of data, storytelling, and Data Visualization’s Narrative, understand how Instruments and methods are used in visualization, and select the appropriate tool for presenting visual data stories and upcoming projects. To address this purpose, this research has reviewed a wide range of different journals and articles to gather information about data visualization, storytelling, and data narratives. Based on the analysis, this research paper finds an all-encompassing perspective on the visual data narrative process, incorporating the phases of discovering insights (data exploration), developing these insights into a tale (creating a story), and sharing this story with others (telling a story). Through the pursuit of these novel research directions, visualizations can facilitate more potent data storytelling.

Industry experts are motivated to collect, collate, and analyse historical data in the legal sector in attempt to predict court case outcomes as the amount of historical data available in this field has increased over time. But using judicial data to predict and defend court judgements is no simple undertaking. Using Machine Learning (ML) models and traditional approaches for categorical feature encoding, previous research on predicting court outcomes using limited experimental datasets produced a number of unexpected predictions. The paper proposes an ensemble model combining Convolutional Neural Network (CNN), attention mechanism and eXtreme Gradient Boosting (XGB) algorithm. This model is primarily based on a self-attention network, which could simultaneously capture linguistic relationships over lengthy sequences like RNN (Recurrent Neural Network) and is nevertheless speedy to train like CNN. C-XGB can obtain accuracy that surpasses the state-of-art model on numerous classification/prediction tasks simultaneously as being twice as speedy to train. The proposed C-XGB model is designed to process the documents hierarchically and calculates the attention weights. Two convolutional layers are used to calculate the attention weights, one at the word level and another at the sentence level. And finally, at the last layer, the XGB algorithm predicts the input case file’s outcome. The experimental results shows that the proposed model outperforms the existing model with 4.67% improvement in accuracy value.