Velammal Educational Trust

Electric line faults such as wire breakage or fallen conductors pose a serious threat to public safety, especially when the affected line continues to carry live current until manual intervention is performed. In conventional power distribution systems, fault identification and power isolation are often delayed due to dependence on human reporting and inspection, which increases the risk of accidents and fatalities. To overcome this limitation, this paper presents an IoT‑based real‑time electric line fault detection system with an automated cut‑off mechanism.The proposed system continuously monitors the continuity of the electric line using a sensing mechanism interfaced with a microcontroller. When a line discontinuity occurs due to wire fall or breakage, the system immediately detects the fault and triggers an automated cut‑off response to isolate the affected line, thereby enhancing public safety. Simultaneously, the fault status is transmitted to a web‑based monitoring dashboard, which can be accessed through a laptop or control room system by Electricity Board (EB) officials.The system is designed to operate in real time, providing fast fault detection, reliable alert visualization, and automatic restoration once the line continuity is re‑established. Experimental validation using a prototype demonstrates the effectiveness of the proposed approach in reducing response time and improving operational safety. The proposed system is simple, cost‑effective, and suitable for smart power distribution and safety‑critical applications.

The growing dependence on cloud-based distributed intelligence requires learning systems that can protect data privacy while maintaining efficiency. Federated learning has become an important method to train models without centralizing raw data, but its effectiveness is hindered by heterogeneous client distributions, adversarial risks, unstable convergence, and inefficient resource utilization. To overcome these challenges, this study introduces FedGenBlk, a federated learning framework that integrates genetic algorithm based optimization with blockchain-supported aggregation. The genetic algorithm dynamically tunes hyperparameters and guides client selection, ensuring stable convergence even under non-IID (non-independent and identically distributed) conditions. Meanwhile, blockchain offers tamper-proof and verifiable model aggregation, reducing the impact of malicious updates and Byzantine failures. Experiments conducted on the EMNIST dataset demonstrate that FedGenBlk achieves 91.2% accuracy, with only 1.9% variance between IID and non-IID settings, and sustains robustness under poisoning attacks with just a 3.1% accuracy drop. Additional validation on CIFAR-10 and Fashion-MNIST datasets confirms consistent performance advantages across different data modalities. These results highlight the framework’s effectiveness in enhancing reliability, security, and adaptability of federated learning in cloud infrastructures, making it particularly suitable for healthcare, IoT, financial services, and cross-organizational collaborative intelligence applications.

Ensuring the integrity and openness of its supply chain is a major concern for the pharmaceutical business. Through the development of an ‎auditable, decentralized tracking system that guarantees the authenticity of pharmaceuticals from manufacture to end-user, this system uses ‎blockchain technology to combat the problem of counterfeit medications. The solution reduces information asymmetry, promotes trust, and ‎improves stakeholder cooperation through the use of a shared ledger. Smart contracts enforce pre-established norms and automate compliance, such as confirming the legitimacy of medications and setting off alarms for questionable activity. In addition to preventing fake phar-‎pharmaceuticals from reaching the market, this strategy improves patient safety by facilitating the quick identification and elimination of bogus ‎prescriptions using a decentralized application‎.

In institutions, developing a timetable is a challenging and problematic process, particularly when creating it manually. When managing multiple departments, some common challenges arise like overlapping schedules, unequal staff allocation for classes and insufficient classroom usage. This proposed work presents an AI-driven solution to automate and optimize timetable scheduling in colleges by using intelligent algorithms. The research involves proposed methodologies that include Constraint Satisfaction Problem (CSP) solutions and Genetic Algorithms, allowing the system to allocate courses, faculty members, classrooms and time slots without any conflicts. This model provides an equal and optimal distribution by considering several limitations like faculty availability, workload balance and institution regulations. The proposed research uses the Genetic Algorithm that represents timetables as “chromosomes” evaluated using a fitness function based on moderate or severe limitations producing solutions through crossover and mutation. The system increases the accuracy of scheduling by reducing human effort. It also handles faculty or time alterations in real-time. This smart automation method improves scalability across institutions and creates a path for a smart educational management system. The proposed system provides a significant shift from manual techniques to data-driven scheduling, increasing educational planning and utilizing resources effectively.

Over the past decade, Human Activity Recognition (HAR) has advanced significantly due to smart devices, large datasets and AI breakthroughs, enabling accurate, real-time activity prediction. This study proposes HAR-CSI-NCGNN, a novel framework using Channel State Information and Node-Level Capsule Graph Neural Networks. Data collected in indoor office settings are preprocessed using the PMCKF to reduce noise and enhance signal clarity. Feature extraction is performed using SSCET, capturing key signal dynamics for radar, biomedical and speech-related tasks. The system supports cross-environment recognition while preserving user privacy and improving generalization. Classification is handled by EPTANN, whose parameters are optimized by the Bitterling Fish Optimization Algorithm (BFOA). Implemented in Python, the framework is evaluated against CNN, ANN, and GNN-based baselines. Experimental results show that EPTANN-BFOA achieves 95–100% accuracy, significantly outperforming state-of-the-art models.

Pythagorean fuzzy digital fine b~-door space (PFD bf~-DS) is a novel notion of generalized Pythagorean fuzzy digital fine topological space that is presented in this paper. Additionally, a number of characterizations of Pythagorean fuzzy digital fine b~-door space are examined. Examples were presented to demonstrate the applicability of these ideas and also some characteristics and connections between other Pythagorean fuzzy digital fine topological spaces and Pythagorean fuzzy digital fine b~-door space were examined. Furthermore, an application which utilizes Pythagorean fuzzy digital fine topological space to select the best menstrual hygiene product for rural women is investigated.