Adhiyamaan College of Engineering

IoT is one of the significant enabling technologies in the contemporary energy landscape. This work addresses intelligent load management under fault, under-load, and overload conditions. Electrical equipment requires automatic and rapid response to avoid damage and prevent service interruption. In contrast to traditional circuit breakers that disconnect the entire system, the proposed architecture employs ESP32-based intelligent control, together with ACS712 current sensors, to provide accurate per-load current measurements and load-specific overcurrent protection. The system is also capable of selective load isolation, meaning that it maintains the operation of healthy circuits and automatically disconnects only the faulty load without affecting other loads. Cloud-based monitoring is implemented via the ThingSpeak platform, enabling real-time remote system surveillance through any internet-connected device. The faulty section of the system is readily identified from the current readings visualised on the cloud dashboard and mobile interface. The proposed system is powered primarily by solar photovoltaic sources and can also operate from the utility grid, providing flexibility across renewable and conventional supply scenarios. Experimental testing demonstrates protection response times in the sub-200 ms range and high-accuracy current monitoring with a mean absolute error below 0.05 A. The key contributions include: selective load protection, real-time IoT-based energy analysis, and a cost-effective open-hardware architecture. The work advances the safety and efficiency of smart energy control systems in distributed renewable energy environments.

Rapid urbanization and rising vehicle density have made urban traffic congestion a significant problem. An integrated framework for intelligent urban traffic flow optimization utilizing mathematical modeling, optimization methods, and AI-driven computational methodologies is presented in this chapter. Adaptive traffic control is made possible by modeling congestion dynamics using graph theory, network flow models, machine learning, and reinforcement learning. The findings of the simulation show increases in traffic throughput, journey time, and delay reduction. The suggested paradigm for intelligent and sustainable transportation systems is practically applicable, as demonstrated by case studies from Indian cities.

Twenty-four new hybrid Pyrrole-tryptamines compounds were synthesized and thoroughly characterized using various techniques such as FT-IR, 1H NMR, 13C NMR, and LC-MS. An evaluation of each compound was conducted based on criteria including their PASS, BBB, ADME, pharmacophore model, and bioactive score. Antibacterial testing of all derivatives indicated that compound 5j exhibited strong activity against MRSA. Investigations into membrane damage, supported by SEM images, cellular content leakage, potassium efflux, and changes in lipid profiles, confirmed the anti-MRSA properties of compound 5j. In an in silico molecular docking analysis, compound 5j achieved a binding score of − 10.02 against the MRSA protein 6FTB, while streptomycin scored − 10.25. Additionally, when compared to standard doxorubicin against 3T3L1 cell lines, compound 5j demonstrated a less toxic IC50 effect of 669.80 µM on normal cell lines. These findings warrant further research on compound 5j for the potential development of a medication to treat MRSA infections.

There is a growing interest from opinion-making bodies and the government to promote the use of solar photovoltaic (PV) systems in the industrial sector. But most grid-connected solar PV plants are equipped with anti-islanding protection that causes the solar inverter to disconnect immediately when the utility grid is down. While this protection helps ensure the safety of utility employees and equipment, it also means that solar energy cannot be used during outages. As such, the industry relies almost entirely on diesel generators (DGs) for power supply, even during the day when solar power is available. The SABIC Research and Technology Centre in Bangalore has a 540-kW grid-connected rooftop solar photovoltaic system. The site consumes about 2,160 kWh of electricity per day. The solar plant indeed makes a noticeable contribution to the load under normal conditions. However, the PV system shuts down during grid power outages due to anti–islanding protection, and the facility runs on DG power only, even though solar power is available. This bottleneck has cost the facility nearly 2.4 lakh kWh of potential solar energy generation over the last two years alone, leading to increased diesel use, higher operating expenses, and higher carbon emissions. This study presents a real-time solar–diesel generator coordination control strategy that enables high penetration and controlled use of solar power during grid outages. The proposed controller adaptively adjusts the output of the solar inverter based on the load demand, the diesel generators operating limits, and the system voltage and frequency. It also prevents reverse power flow to the diesel generator and maintains a minimum spinning reserve for stable operation. The discussed system is simulated and modelled in MATLAB/Simulink. Simulation results demonstrate that different scenarios were analysed, including grid outage, load variation, and solar generation fluctuation. The simulation results show that the presented scheme can maintain stable system voltage and frequency and maximise solar power utilisation. The system significantly reduces the diesel generator load and fuel consumption, improving energy efficiency and environmental friendliness for the industrial plant.

With the rapid growth of Mobile Ad hoc Networks (MANETs), decentralized and dynamic communication systems have become essential for applications such as the Internet of Things, disaster recovery, and battlefield communication. Ensuring quality of service in MANETs remains challenging due to dynamic topologies, limited bandwidth, and resource-constrained nodes. This research aims to develop a scalable, secure, and energy-efficient framework for malicious node detection and optimized routing in dynamic MANET environments. This research introduces an Explainable Liquid Spike Edge Graph Attention (ELSEGA) method to detect malicious nodes, enhancing secure routing decisions through interpretable and adaptive feature extraction. The White-Faced Capuchin Optimizer (WFCO) is employed to improve resource allocation efficiency, providing robust global search capabilities and adaptive decision-making for optimized network performance. Federated learning-enhanced quality of service multicast tree routing is implemented to optimize routing decisions while preserving data privacy across distributed nodes. Reconfigurable intelligent surface technology is utilized to improve signal quality and mitigate interference in complex environments. The integration of these techniques achieves substantial improvements in security, routing stability, energy efficiency, detection accuracy, and attack resilience, offering a scalable, privacy-preserving solution for 6G-enabled MANETs. The proposed approach achieves a packet delivery ratio of 97.65