Category Archives: Uncategorized

Authors: Dr. Osman Zenk

Abstract: As is known, memcapacitors, which are memory elements, are electrical circuit elements ex-pected to revolutionize various research and many engineering fields thanks to their variable capacitance and non-volatile memory effects. Although there is not yet enough scientific research on memcapacitors in power electronics systems, they have serious scientific discovery potential, especially in terms of reducing output voltage ripple, increasing voltage stability, and improving energy efficiency. In this study, theoretical and simulation results performed in the Matlab/Simulink environment are presented, showing that adding a memcapacitor to the output of a commonly used buck-boost converter, a type of DC-DC converter, significantly reduces the output voltage ripple. In the study, a memcapacitor emulator that can be implemented using commercial components is first proposed and validated. Then, this emulator was used to exam-ine the steady-state output voltage ripple and transient response of the buck-boost converter. The results show that the use of a load-controlled memcapacitor can reduce the output voltage ripple by up to 96%.

DOI: https://doi.org/10.5281/zenodo.18127037

Published by: vikaspatanker

Authors: Maram Mounika, Rakshitha L, Ramya R, Ramya Shree D, Dr Manasa P

Abstract: Inertial Navigation Systems (INS) are mission- critical subsystems in naval platforms, providing continuous nav- igation information independent of Global Navigation Satellite Systems (GNSS). In contested or GNSS-denied environments, the reliability of INS data directly impacts navigation accuracy, weapon control, and overall mission effectiveness. This paper presents the design and implementation of a real-time INS monitoring platform developed for Indian Naval shipboard appli- cations. The proposed system acquires high-speed navigation data from dual Ring Laser Gyro (RLG) based INS units operating at 10 Hz and 100 Hz through RS-422 interfaces. Using industrial- grade USB-to-serial hardware and a LabVIEW-based software framework, the system performs real-time data acquisition, frame validation, parameter extraction, visualization, logging, and replay. Experimental results demonstrate reliable, lossless data capture and synchronized monitoring of heading, roll, and pitch from forward and aft INS units, validating the effectiveness of the platform for onboard monitoring, testing, and post-mission analysis.

Published by: vikaspatanker

Authors: Girish Kishor Ingavale

Abstract: The exponential growth in computational power and industrial processes has led to an increased demand for efficient cooling solutions in data centers. Traditional air-cooling systems are becoming inadequate due to their limitations in managing high thermal loads and their high energy consumption. In response to these challenges, Direct-to-Chip Liquid Cooling Systems (D2C LCS) have emerged as a promising alternative for thermal management in high-density computing environments. This study investigates the application of Artificial Intelligence (AI) and Machine Learning (ML) technologies to optimize the performance of D2C LCS in rack-mounted data center setups. The primary objective of this research is to develop and implement AI-driven models that can predict temperature and fluid flow within D2C LCS, thereby enabling the optimization of cooling strategies. By leveraging advanced algorithms such as Linear Regression and Support Vector Machine, the study aims to enhance thermal efficiency and reduce the energy consumption of data centers. Experimental data was collected from a simulated data center environment equipped with D2C LCS. The data was used to train and validate ML models, ensuring their accuracy and reliability in real-world applications. The results demonstrate that AI-optimized cooling strategies can achieve a 15% reduction in temperature and a 20% decrease in energy consumption compared to traditional air-cooling systems. The findings of this study highlight the significant benefits of integrating AI and ML technologies with D2C LCS for thermal management in data centers. The predictive models and optimized cooling strategies presented herein provide a robust framework for improving the efficiency and sustainability of data center operations. Future research directions include the development of more advanced AI models and the implementation of real-time monitoring systems to further enhance the performance of D2C LCS.

DOI: http://doi.org/10.5281/zenodo.18125704

Published by: vikaspatanker

Authors: Dr. Chetana Reddy, Divya K, Madhumitha B, Maithra K, Melisha K Sunny

Abstract: Power transformers must work well and be reliable to keep the flow of electricity stable and uninterrupted. Standard periodic maintenance often does not find problems in their early stages, which can lead to insulation breakdown, oil breakdown, thermal stress, overloading, and unexpected outages. This paper proposes an IoT-based Transformer Health Monitoring System to address these limitations. The system can continuously and in real time monitor important operating parameters. The system uses sensors for temperature, oil level, load current, and input voltage that are connected to a microcontroller. The microcontroller processes the data and sends it to a cloud-based monitoring platform. The data analyzed by the IOT platform ensures early fault detection for maintenance planning. To support predictive maintenance, the suggested framework provides threshold-based alert notifications, historical logging, real-time data visualization, and remote access. The system creates automated alerts to stop overheating, insulation failure, and possible transformer failures when anomalous conditions are identified. This system monitors multiple transformers at different distributed substation. This IoT- enabled strategy prolongs transformer lifespan, lowers maintenance costs, minimizes downtime, and improves operational safety. The solution offers a scalable architecture for intelligent monitoring across substations and distribution networks and is in line with efforts to modernize smart grids.

Published by: vikaspatanker

Authors: Abhishek Rawat, Basant Agarwal

Abstract: In recent years, extensive research has been conducted on vibrations in air-craft. Vibration can cause some serious failure in the structure. The increase in disquisition in vibration has led to taking design considerations in the bod-ies. This paper specifically focuses on the vibration in bodies causing defor-mation in aircraft wing. Modal analysis of two different wing is done using Fi-nite Element Analysis. Three different material namely: Aluminium Alloy, Copper Alloy and Titanium Alloy have been incorporated to describe the ef-fects of free vibration on different wings.

Published by: vikaspatanker

Authors: Pranav Dhondibhau Gawade, Sarthak Vivek Sagare, Sujan Anna Kambale, Piyush Vinod Chaudhary, Neelam N Kavale

Abstract: The proposed Smart Community Health Monitoring and Early Warning System for rural areas offers a transformative, cost-effective alternative to expensive, sensor-dependent technologies by prioritizing syndromic surveillance and community-led data collection. Recognizing that traditional IoT infrastructure often fails in remote regions due to high maintenance costs and power instability, this model empowers community health workers to act as "human sensors," manually reporting clinical symptoms like fever and diarrhea via an offline-capable mobile interface. By integrating these health reports with periodic, low-cost chemical water testing, the system utilizes a centralized analytical engine to run statistical aberration detection algorithms that compare real-time trends against historical baselines. This proactive framework identifies potential pathogenic outbreaks at their nascent stage, triggering a tiered Early Warning System (EWS) that alerts local authorities through automated SMS and voice calls. Ultimately, this research demonstrates that public health resilience is not solely dependent on high-tech hardware but can be achieved through strategic data management, community participation, and smart analytics, providing a scalable and sustainable blueprint for disease prevention in resource-constrained environments globally.

Published by: vikaspatanker

Authors: Priya Palanimurugan, Miss. R.Christy Alice, Dr.Thulasi Bharathi.M, M.sakthivel

Abstract: This essay delves into the complex confluence of emotional realism, cinematic expression, and mental health in the cinema of Balu Mahendra, India's most empathetic director. Through his sensitive explorations of human vulnerability, psychological trauma, and resilience, Mahendra remapped the way the Tamil cinema represented the human mind and its delicate complexities. Exceeding commercial norms, his films like Moondram Pirai (1982), Marupadiyum (1993), Sandhya Raagam (1989), and Veedu (1988) demonstrate a deep psychological realism that humanizes characters normally pushed to the periphery by social or emotional pain. This research uses a psychological model based on trauma theory, studies on empathy, and humanistic psychology to examine how Mahendra's film language turns pain into poetry and silence into emotional conversation. The study places Mahendra's films within the larger framework of Indian cinema's shifting approach to mental health, highlighting how his stories avoid the melodramatic spectacle commonly linked with psychological illness. Rather, his characters are characterized by a quiet dignity that mirrors the internal struggles of memory, loss, identity, and moral dissonance. The paper also investigates the aesthetic aspects of Mahendra's visual realism—his natural lighting, long takes, and close-ups—as methods that conjure emotional truth and ask viewers to enter a reflective psychological zone. Finally, this essay maintains that Balu Mahendra's films work as sympathetic case studies of the human mind, providing social commentary and emotional counseling. His world of film challenges viewers to see mental illness neither as weakness nor as supernatural affliction but as a vital aspect of human experience. In this process, Mahendra's body of work helps in the destigmatization of mental pain and promotes a different cinematic language based on compassion, realism, and psychological complexity.

Published by: vikaspatanker

Authors: Priya Palanimurugan, Mr Kalaiselvan S, Dr.Thulasi Bharathi M, M.sakthivel

Abstract: Mental health issues remain a global public health concern, especially among the youth and digitally active population. This study examines the effect of digital storytelling as an intervention tool to increase mental health awareness, reduce stigma and encourage positive behavior changes. A quantitative research design involving 300 graduate students aged 18–25 in diverse educational subjects was employed. Participants were divided into an intervention group, which reflects a control group that receives curate digital stories and traditional information-based materials reflecting real-life mental health experiences.Advanced statistical techniques were used to assess the results in three time points (Post, Post-up). Descriptive data briefly presented demographic data; Alpha of Cronback confirmed the reliability of the scale; Confirmation factor analysis (CFA) valid measurement construction; And the multi -comprehensive analysis of the covalent (mancova) identified important group differences. Repeated measures Anova and Structural Equation Modeling (SEM) further detected time-based reforms in the intention of mental health awareness and behavior, mediate by low stigma. Moderation and latent development analysis highlighted demographic effects and individual trajectory patterns. Conclusions suggest that digital storytelling improves mental health awareness and reduces stigma compared to traditional approaches (P <0.01). The narrative-based method was particularly effective among the pre-risk participants for high digital literacy and mental health materials. The study supports the integration of digital story stories in public health education and mental health advocacy programs. These results contribute to increasing evidence that creative digital equipment can change mental health communication, offering scalable, attractive and human-focused solutions.

Published by: vikaspatanker

Authors: Sakthivel S, Vikash P

Abstract: The rapid expansion of digital financial services has significantly transformed the global financial ecosystem by enabling fast, convenient, and seamless transactions. However, this transformation has also increased the vulnerability of financial systems to fraudulent activities such as credit card fraud, identity theft, phishing attacks, insider fraud, and money laundering. Financial fraud results in substantial economic losses, damages institutional reputation, and undermines customer trust in digital banking systems. Traditional fraud detection mechanisms primarily rely on rule-based systems and manual audits, which are reactive, inflexible, and often incapable of detecting complex and evolving fraud patterns in real time. Advancements in artificial intelligence (AI), machine learning (ML), and data analytics have paved the way for intelligent financial fraud detection systems capable of processing large volumes of transaction data efficiently. By learning patterns from historical transaction data and identifying anomalies, AI-driven systems enable early detection and prevention of fraudulent activities. This paper presents an AI-based financial fraud detection framework that integrates data preprocessing, feature engineering, and machine learning-based classification for real-time fraud analysis. The proposed system aims to improve detection accuracy, reduce false positives, and enhance the overall security of digital financial transactions. Experimental results and analysis demonstrate that intelligent fraud detection systems provide scalable, adaptive, and reliable solutions for modern financial environments.

Published by: vikaspatanker

Authors: Mahavani Kb, Bavithra Rs

Abstract: Federated learning has also become a paradigm shift to making machine learning collaborative and not centralized around sensitive data. Federated learning solves the increasing privacy, regulatory compliance, and data sovereignty concerns by preventing the transfer of model training to centralized model training clients, like hospitals, financial institutions, and IoT devices. Cloud platforms are critical to the operationalization of this paradigm as it offers scalable orchestration, secure aggregation, and communication-efficient frameworks. The paper discusses how cloud native federated learning systems decrease the amount of communication, enhance the model convergence, and provide more robust privacy guarantees without violating regulation of systems like GDPR and HIPAA. By applying federated learning to the medical diagnostic and financial fraud detection domains, the study shows that federated learning can be successful in providing a high level of model accuracy and strong privacy protection. The results indicate the significance of supporting federated learning by cloud-native infrastructure that will allow implementing privacy-safe AI solutions that can be widely adopted in regulated industries. From a privacy and regulatory perspective, cloud-based federated learning systems provide strong guarantees that align with data protection regulations such as the General Data Protection Regulation (GDPR) and the Health Insurance Portability and Accountability Act (HIPAA). By eliminating the need for raw data transfer, federated learning inherently supports privacy-by design principles. When combined with advanced privacy-preserving techniques such as differential privacy, secure multi-party computation, and homomorphic encryption, federated learning further strengthens its compliance with strict legal and ethical requirements. To demonstrate the effectiveness of cloud-native federated learning, this study applies the proposed framework to two critical application domains: medical diagnosis and financial fraud detection. Experimental results show that federated models achieve performance levels comparable to, and in some cases exceeding, those of traditional centralized models, while significantly enhancing data privacy and security. In medical diagnostics, federated learning enables collaborative training across multiple healthcare institutions without exposing sensitive patient records. Similarly, in financial fraud detection, federated learning facilitates cross institutional intelligence sharing without compromising proprietary or customer data.

DOI:

Published by: vikaspatanker