Volume of International Discoveries and Youth Academic Works, Achievements, and Novelties - VIDYAWAN

Next-generation hydrogen fuel cells must raise power density and efficiency while simultaneously reducing platinum-group-metal loading, local thermal gradients, water flooding and durability losses. This study proposes an entropy-guided, functionally graded membrane-electrode assembly (EG-FGMEA) that couples four spatial design decisions: a reinforced thin perfluorosulfonic-acid membrane, a through-plane cathode ionomer gradient, graded catalyst accessibility, and a hydrophilic-to-hydrophobic microporous-layer transition. A steady-state, two-phase, non-isothermal model resolves charge transport, oxygen diffusion, membrane hydration, liquid saturation and heat generation. A deterministic Latin-hypercube design explores 180 candidate architectures and is filtered through a three-objective Pareto criterion comprising net power density, a voltage-decay proxy and normalized entropy generation. Relative to a conventional uniform catalyst layer, the selected architecture increases the modelpredicted peak net power density by 24.0% and raises voltage at 1.5 A cm-2 by 51 mV. Maximum cathode liquid saturation falls from 0.59 to 0.34, while the integrated irreversibility index decreases by 20%. Sensitivity analysis identifies effective oxygen diffusivity, area-specific resistance and ionomer-gradient amplitude as the dominant controllable variables. The analysis predicts a practical design window near 88-94 degC, 60-75% cathode relative humidity, 1.6-1.9 bar absolute pressure and 0.10-0.14 mgPt cm-2 total platinum loading. These results define testable design hypotheses rather than experimental proof and provide a transparent framework for subsequent catalyst-coated-membrane fabrication and validation.

Artificial Intelligence (AI) has emerged as a transformative paradigm in electrical power systems, enabling real-time optimization, predictive control, and intelligent energy distribution. This paper proposes a unified AI-driven framework for power management integrating machine learning, deep reinforcement learning, and edge intelligence across smart grids, microgrids, and distributed energy resources (DERs). The proposed architecture enhances efficiency, resilience, and sustainability by enabling adaptive load balancing, predictive maintenance, and dynamic demand-response control. A comparative analysis demonstrates that AI-based systems can achieve significant improvements in forecasting accuracy, operational cost reduction, and system stability. Recent studies show load forecasting errors can be reduced below 2% using advanced AI models. The paper further explores challenges such as computational overhead, data integrity, and cybersecurity. Finally, future research directions involving federated learning, digital twins, and AI-integrated energy markets are discussed.

The deterioration of apple leaves through disease imposes a tremendous impact on the quality of fruit production from apple orchards. Therefore, swift and accurate identification of these diseases is necessary for effective fruit management. Traditional methods for identifying apple leaf diseases utilize expert humans who inspect the leaves visually. However, visual inspections are labor-intensive, highly subjective, and cannot be conducted effectively in larger orchards. The rapid advancement of deep learning research is facilitating automated identification of apple leaf diseases from images of contaminated leaves using Convolutional Neural Networks (CNNs) that are able to learn visual characteristics such as spots, lesions, colour changes, or texture differences on apple tree leaves. However, when using a single CNN model to accomplish this task, one may encounter difficulties due to environmental variability, similarities between classes of diseases, and generalisation across variable field conditions. The current work proposes a hybrid ensemble CNN model for the analysis of apple leaf disease by providing a solution to these challenges. The ensemble model consists of EfficientNet-B0, DenseNet121, and ResNet50. Feature-level fusion of these models concurrently enables the extraction of strong and complementary feature representations from images for improved performance. The proposed hybrid CNN ensemble model achieved a classification accuracy of 99% when evaluated on a dataset of apple leaf disease images. Furthermore, the proposed model consistently outperformed backbone models such as EfficientNet-B0, VGG16, ResNet50, and MobileNetV2. Therefore, this proposed model has the potential to be an effective, reliable, scalable method to perform disease monitoring in apple production systems.

The rapid proliferation of Internet of Things (IoT) devices across healthcare, industrial automation, smart agriculture, smart grid, and smart-city domains has magnified concerns surrounding data integrity, device authentication, and network resilience. Conventional client-server and cloud-centric IoT security models suffer from single points of failure, limited transparency, and poor scalability as the number of connected endpoints grows into the tens of billions. Blockchain technology, with its decentralized, immutable, and cryptographically verifiable ledger structure, has emerged as a promising complement to traditional IoT security mechanisms. This paper presents a comprehensive study of blockchain-based security for IoT networks from an Electronics and Communication Engineering (ECE) perspective. We review the layered architecture of IoT systems, identify dominant threat surfaces at the perception, network, edge, and application layers, and propose an integrated five-layer blockchain-IoT security architecture suited to resource-constrained embedded hardware. We present a comparative evaluation of consensus mechanisms — Proof of Work (PoW), Proof of Stake (PoS), Practical Byzantine Fault Tolerance (PBFT), Delegated Proof of Stake (DPoS), Proof of Authentication (PoAh), and Proof of Authority (PoA) — with respect to energy consumption, latency, and scalability for constrained IoT nodes. A threat model mapping common blockchain-IoT attack vectors to mitigation strategies is developed, along with case-study discussion of healthcare (IoMT), supply chain, and smart-grid deployments. The paper concludes with open research challenges — including scalability, interoperability, quantum-resistance, and regulatory alignment — and outlines future research directions relevant to hardware-software co-design in the ECE domain

Financial literacy, socio-demographic factors and behavioral factors were added and investigated the investment behavior of the women academician in Rajasthan, India. The primary cross-sectional study was conducted on 400 respondents from Higher Educational Institutions (HEIs) of Rajasthan by using structured questionnaire. Analysis was done by descriptive statistics and correlation analysis, cross tabulation, and Ordinary Least Squares (OLS) regression. Results indicated moderate financial literacy, with roughly 32.5% of respondents getting the financial literacy questions right, there was clearly some room for improvement in terms of financial literacy regarding understandings of inflation and compounding. The analysis further reveals that preference was given to more traditional financial instruments such as gold (52%) and fixed deposits (68.5%) with low risk. Results of the OLS regression models indicate that financial literacy (β = 0.284) is Along with income level and risk attitude, the factors of statistically significant in determining the investment behavior. Based on the micro level, this study belongs to the body of literature based on the micro level data on the not so much studied group of women academicians with regional Indian scenario and not only that but also gives the financial education its importance so as to empower women and help them in making educated investment decisions. The findings will try to document the current scenario of women academicians in terms of their financial literacy and their investment habits with an aim for highlighting the financial literacy issue among women academics. This paper has been prepared to find out financial literacy level of the women academicians and Investment behavior of women academicians and to solve the problem of low financial literacy of women academicians.

The Marathwada region of Maharashtra, India, faces chronic agricultural distress driven by severe water scarcity, recurring droughts, and rapid climate variability. Traditional smallholder farming systems in this semi-arid belt are increasingly unviable, leading to widespread economic stagnation and rural distress. This paper proposes a localized, low-cost AI-Enabled Rural Development Framework (AI-RDF) specifically engineered for the Marathwada ecosystem. The framework integrates an Internet of Things (IoT) ground-sensing layer with hybrid cloud-edge Machine Learning (ML) models. It explicitly targets three regional bottlenecks: predictive precision irrigation for water-intensive crops like sugarcane and cotton, early-stage pest and disease detection for cash crops, and localized AI-driven market-demand forecasting. Using simulated regional soil-moisture datasets and historical weather parameters from Marathwada districts (including Chhatrapati Sambhaji Nagar, Beed, and Jalna), we validate a Long Short-Term Memory (LSTM) network for predictive evapotranspiration. Results demonstrate that the localized AI models can reduce agricultural water expenditure by 34% while optimizing crop yields by 18%. Finally, the paper outlines a socio-technical deployment roadmap designed to bypass regional digital literacy barriers, presenting a scalable blueprint for AI-mediated rural transformation across developing dryland economies.

Next-generation hydrogen fuel cells must raise power density and efficiency while simultaneously reducing platinum-group-metal loading, local thermal gradients, water flooding and durability losses. This study proposes an entropy-guided, functionally graded membrane-electrode assembly (EG-FGMEA) that couples four spatial design decisions: a reinforced thin perfluorosulfonic-acid membrane, a through-plane cathode ionomer gradient, graded catalyst accessibility, and a hydrophilic-to-hydrophobic microporous-layer transition. A steady-state, two-phase, non-isothermal model resolves charge transport, oxygen diffusion, membrane hydration, liquid saturation and heat generation. A deterministic Latin-hypercube design explores 180 candidate architectures and is filtered through a three-objective Pareto criterion comprising net power density, a voltage-decay proxy and normalized entropy generation. Relative to a conventional uniform catalyst layer, the selected architecture increases the modelpredicted peak net power density by 24.0% and raises voltage at 1.5 A cm-2 by 51 mV. Maximum cathode liquid saturation falls from 0.59 to 0.34, while the integrated irreversibility index decreases by 20%. Sensitivity analysis identifies effective oxygen diffusivity, area-specific resistance and ionomer-gradient amplitude as the dominant controllable variables. The analysis predicts a practical design window near 88-94 degC, 60-75% cathode relative humidity, 1.6-1.9 bar absolute pressure and 0.10-0.14 mgPt cm-2 total platinum loading. These results define testable design hypotheses rather than experimental proof and provide a transparent framework for subsequent catalyst-coated-membrane fabrication and validation.

Artificial Intelligence (AI) has emerged as a transformative paradigm in electrical power systems, enabling real-time optimization, predictive control, and intelligent energy distribution. This paper proposes a unified AI-driven framework for power management integrating machine learning, deep reinforcement learning, and edge intelligence across smart grids, microgrids, and distributed energy resources (DERs). The proposed architecture enhances efficiency, resilience, and sustainability by enabling adaptive load balancing, predictive maintenance, and dynamic demand-response control. A comparative analysis demonstrates that AI-based systems can achieve significant improvements in forecasting accuracy, operational cost reduction, and system stability. Recent studies show load forecasting errors can be reduced below 2% using advanced AI models. The paper further explores challenges such as computational overhead, data integrity, and cybersecurity. Finally, future research directions involving federated learning, digital twins, and AI-integrated energy markets are discussed.

The deterioration of apple leaves through disease imposes a tremendous impact on the quality of fruit production from apple orchards. Therefore, swift and accurate identification of these diseases is necessary for effective fruit management. Traditional methods for identifying apple leaf diseases utilize expert humans who inspect the leaves visually. However, visual inspections are labor-intensive, highly subjective, and cannot be conducted effectively in larger orchards. The rapid advancement of deep learning research is facilitating automated identification of apple leaf diseases from images of contaminated leaves using Convolutional Neural Networks (CNNs) that are able to learn visual characteristics such as spots, lesions, colour changes, or texture differences on apple tree leaves. However, when using a single CNN model to accomplish this task, one may encounter difficulties due to environmental variability, similarities between classes of diseases, and generalisation across variable field conditions. The current work proposes a hybrid ensemble CNN model for the analysis of apple leaf disease by providing a solution to these challenges. The ensemble model consists of EfficientNet-B0, DenseNet121, and ResNet50. Feature-level fusion of these models concurrently enables the extraction of strong and complementary feature representations from images for improved performance. The proposed hybrid CNN ensemble model achieved a classification accuracy of 99% when evaluated on a dataset of apple leaf disease images. Furthermore, the proposed model consistently outperformed backbone models such as EfficientNet-B0, VGG16, ResNet50, and MobileNetV2. Therefore, this proposed model has the potential to be an effective, reliable, scalable method to perform disease monitoring in apple production systems.

The rapid proliferation of Internet of Things (IoT) devices across healthcare, industrial automation, smart agriculture, smart grid, and smart-city domains has magnified concerns surrounding data integrity, device authentication, and network resilience. Conventional client-server and cloud-centric IoT security models suffer from single points of failure, limited transparency, and poor scalability as the number of connected endpoints grows into the tens of billions. Blockchain technology, with its decentralized, immutable, and cryptographically verifiable ledger structure, has emerged as a promising complement to traditional IoT security mechanisms. This paper presents a comprehensive study of blockchain-based security for IoT networks from an Electronics and Communication Engineering (ECE) perspective. We review the layered architecture of IoT systems, identify dominant threat surfaces at the perception, network, edge, and application layers, and propose an integrated five-layer blockchain-IoT security architecture suited to resource-constrained embedded hardware. We present a comparative evaluation of consensus mechanisms — Proof of Work (PoW), Proof of Stake (PoS), Practical Byzantine Fault Tolerance (PBFT), Delegated Proof of Stake (DPoS), Proof of Authentication (PoAh), and Proof of Authority (PoA) — with respect to energy consumption, latency, and scalability for constrained IoT nodes. A threat model mapping common blockchain-IoT attack vectors to mitigation strategies is developed, along with case-study discussion of healthcare (IoMT), supply chain, and smart-grid deployments. The paper concludes with open research challenges — including scalability, interoperability, quantum-resistance, and regulatory alignment — and outlines future research directions relevant to hardware-software co-design in the ECE domain

Financial literacy, socio-demographic factors and behavioral factors were added and investigated the investment behavior of the women academician in Rajasthan, India. The primary cross-sectional study was conducted on 400 respondents from Higher Educational Institutions (HEIs) of Rajasthan by using structured questionnaire. Analysis was done by descriptive statistics and correlation analysis, cross tabulation, and Ordinary Least Squares (OLS) regression. Results indicated moderate financial literacy, with roughly 32.5% of respondents getting the financial literacy questions right, there was clearly some room for improvement in terms of financial literacy regarding understandings of inflation and compounding. The analysis further reveals that preference was given to more traditional financial instruments such as gold (52%) and fixed deposits (68.5%) with low risk. Results of the OLS regression models indicate that financial literacy (β = 0.284) is Along with income level and risk attitude, the factors of statistically significant in determining the investment behavior. Based on the micro level, this study belongs to the body of literature based on the micro level data on the not so much studied group of women academicians with regional Indian scenario and not only that but also gives the financial education its importance so as to empower women and help them in making educated investment decisions. The findings will try to document the current scenario of women academicians in terms of their financial literacy and their investment habits with an aim for highlighting the financial literacy issue among women academics. This paper has been prepared to find out financial literacy level of the women academicians and Investment behavior of women academicians and to solve the problem of low financial literacy of women academicians.

The Marathwada region of Maharashtra, India, faces chronic agricultural distress driven by severe water scarcity, recurring droughts, and rapid climate variability. Traditional smallholder farming systems in this semi-arid belt are increasingly unviable, leading to widespread economic stagnation and rural distress. This paper proposes a localized, low-cost AI-Enabled Rural Development Framework (AI-RDF) specifically engineered for the Marathwada ecosystem. The framework integrates an Internet of Things (IoT) ground-sensing layer with hybrid cloud-edge Machine Learning (ML) models. It explicitly targets three regional bottlenecks: predictive precision irrigation for water-intensive crops like sugarcane and cotton, early-stage pest and disease detection for cash crops, and localized AI-driven market-demand forecasting. Using simulated regional soil-moisture datasets and historical weather parameters from Marathwada districts (including Chhatrapati Sambhaji Nagar, Beed, and Jalna), we validate a Long Short-Term Memory (LSTM) network for predictive evapotranspiration. Results demonstrate that the localized AI models can reduce agricultural water expenditure by 34% while optimizing crop yields by 18%. Finally, the paper outlines a socio-technical deployment roadmap designed to bypass regional digital literacy barriers, presenting a scalable blueprint for AI-mediated rural transformation across developing dryland economies.