JK Lakshmipat University

The implementation of solar and wind technologies has become an important aspect of any global sustainability effort, but the sustainability of these technologies can be diminished through steady depreciation of componentry over the long-term as well as surprise component failures due to the natural or environmental effects to the technology that results in downtime and less than satisfactory energy generation. Considerable modern planning and contemporary are needed in order to act proactively and decisively in each example of an incident: needing to have something beyond a conventional maintenance plan, it has developed into the need for a predictive strategy that leads to proactive identification of failure before significant depreciation of componentry results in total failure. The results of the benchmarking indicate, once again, that the hybrid method proposed in this paper outperforms benchmarks and traditional and single model based approaches, resulting in an F1-score of 94.3 % and an AUC-ROC of 95.4. The scheduling with respect to the PMI also resulted in a significant decrease in downtime, 28.6 %, and reduced inefficience, 17.4 %. The system provided not only performance gains, but also tolerated drift, and modeled for changing conditions when deployed in an edge-AI device test, with inference times under 2.1 seconds. Taken together, these results illustrate not only the efficiency of the model but also its practical viability for field deployment. The framework, by enhancing predictability and maintaining flexibility under uncertainty, contributes to the broader goal of sustainable energy management and positions hybrid machine learning as a credible enabler for smart maintenance in renewable infrastructures.

Herbs have historically been central to medicinal practices, representing one of the earliest forms of therapeutic intervention. While synthetic drugs are often highly effective in treating acute conditions, their use is frequently accompanied by adverse side effects. In addition, the growing dependence on synthetic pharmaceuticals has raised concerns regarding affordability, thereby fostering a renewed interest in herbal medicine as a cost-effective and holistic alternative. In response to this need, the current study introduces a computer vision framework for accurate herb identification. A novel dataset, Herbify, was compiled from two different herb datasets and refined through rigorous cleaning, preprocessing, and quality control procedures. The resulting dataset underwent standardization via the Preprocessing Algorithm for Herb Detection (PAHD), producing a refined dataset of 6104 images, representing 91 distinct herb species, with an average of about 67 images per species. Utilizing transfer learning, the research harnessed pre-trained Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs), then integrated these models into an ensemble framework that leverages the unique strengths of each architecture. Experimental results indicate that EfficientNet v2-Large achieved a noteworthy F₁-score of 99.13%, while the ensemble of EfficientNet v2-Large and ViT-Large/16, termed EfficientL-ViTL, attained an even higher F₁-score of 99.56%. Additionally, the research also introduces ‘Herbify’ application, an AI-driven framework designed to identify herbs using the developed model. By directly tackling the principal obstacles in herb identification, the proposed system achieves a highly accurate and operationally viable classification mechanism. The experimental outcomes showcase top-tier performance in herb identification and emphasize the transformative potential of AI-based solutions in supporting botanical applications.Supplementary InformationThe online version contains supplementary material available at 10.1186/s13007-025-01421-5.

This paper explores the potential of hydrogen fuel cell technology as a sustainable alternative to diesel and electric-powered locomotives in Indian Railways (IR), in line with India's National Hydrogen Mission and broader sustainability goals. Focusing on a pilot project in the Jind–Sonipat railway section, the study assesses the economic feasibility and environmental impact of hydrogen-powered trains. A comparative cost analysis reveals that while hydrogen trains involve higher initial capital investment, their operational costs-estimated at ₹164.5 per 1000 passenger-km, are competitive, falling between diesel (₹79.78) and electric trains (₹256.2). Integrating hydrogen fuel cells with renewable energy sources offers significant emissions reductions and supports India's transition toward a low-carbon transport sector. However, large-scale adoption will require further research to address safety concerns related to hydrogen storage and refuelling and operational challenges arising from India's varied climate and topography. With continued technological progress and supportive policy frameworks, hydrogen propulsion presents a promising pathway for cleaner, more sustainable rail transport in India.

Scalability remains a major challenge in blockchain technology, particularly for Layer-1 networks like Ethereum, where high transaction volumes cause congestion and high gas fees. Zero-Knowledge (ZK) Rollups have emerged as scalable Layer-2 solutions, offering enhanced security and lower transaction costs. However, existing zk-Rollups such as zkSync Era, dYdX, and StarkNet each have their advantages and limitations in terms of efficiency, cost, and decentralization. This paper proposes zkFusion, a hybrid zk-Rollup protocol that combines the most effective features of these rollups to achieve faster transaction rates, improved cost-effectiveness, and enhanced security. zkFusion employs a flexible proof mechanism, optimized transaction batching, and a modular data throughput method, allowing users to balance between cost and security. Additionally, it integrates a decentralized sequencer to enhance network reliability and mitigate centralization risks. Comparative evaluation with existing zk-Rollups demonstrates that zkFusion significantly reduces transaction fees, improves throughput, and maintains high security assurances. By blending the innovations of zkSync Era, dYdX, and StarkNet, zkFusion aims to set a new benchmark for scalability, efficiency, and cost reduction in Layer-2 blockchain solutions.

Producer Responsibility Organizations (PROs) are professional organizations that take responsibility for collecting and treating E-waste in an eco-friendly way on behalf of producers. However, they face several challenges in effective e-waste management (EWM). The current study aims to identify and model key challenges in EWM from PROs’ perspective in the Indian context. We identified key challenges to EWM through a literature review and discussions with experts. The expert team validated and mapped key challenges into the EWM process, from collection to disposal. We employed the DEMATEL method to model key challenges. Nine key challenges were identified. Findings suggest that a knowledge-sharing platform is a crucial step in EWM as it allows PROs to share best practices. In addition, collaboration among stakeholders, usage of tracking systems, and eco-design could play a key role in sustainable and effective EWM. This study provides various mitigation strategies to overcome key challenges in the adoption of sustainable EWM. Furthermore, PROs can use the proposed conceptual framework to identify challenges that may be faced during different EWM process stages.