Engineering, Electrical and Electronic Engineering, Multidisciplinary, Renewable Energy, Sustainability and the Environment
9
Scopus Publications
Scopus Publications
IoT-Based Thermal Anomaly Alert System for Wind Turbine P. Balasubramani, Kavin M, Madesh S, Muthu Krishnan M Proceedings of 5th International Conference on Communication Computing and Electronics Systems Iccces 2026, 2026 The use of wind turbines as a source of renewable energy has also led to a growing concern about fire hazards due to overheating, electrical faults, or mechanical breakdowns in the turbine parts. Such anomalies are very important to detect at an early stage to avoid catastrophic failures and to guarantee the safety of operations. This project is based on an IoT-based anomaly thermal detector of wind turbine fire detection (coupled to a microcontroller ESP32), which is the central unit of the proposed project. The system has a combination of several sensors, such as an MQ135 gas sensor that detects harmful gases, a flame sensor that detects real-time fire, and a smoke sensor that detects the presence of combustion byproducts. These also have sensors that constantly check the environmental conditions within the turbine nacelle. The gathered data is reported to the ThingSpeak cloud platform, where it is stored, analysed, and visualised. When abnormal readings are registered, operators can be alerted in real time, which allows for preventing big losses. The system will provide a low-cost, reliable, and scalable fire safety system in wind energy infrastructure, which will aid the improvement of reliability, downtime reduction, and protection of renewable energy infrastructure.
A VLSI-based multi-level ECG compression scheme with RL and VL Encoding P. Balasubramani, S. Swathi Krishna, E. Udayakumar Distributed Time Sensitive Systems, 2025 Wearable sensor nodes produce a lot of data because they are equipped with characteristics for continuous monitoring. aside from the fact that data transmission uses about 3/4 of the sensor node's power. In wireless body area networks, nodes generate a sizable quantity of information throughout the prolonged and continuous monitoring of any physiological signals, boosting transmission power and battery consumption. To save both space and energy, a compression of data without loss technique is recommended for an ECG signal monitoring system. Bit compression is used to create a suggested hybrid lossless multi-level compression system using different lengths of encoding with dictionary selection. When the input bit stream's amplitude values are significantly lower and there are uninterrupted runs of ones and zeros, encoding with variable length is used. Among the most efficient linear encoding approaches applied is code. A huge collection of matching patterns can be produced using an effective bitmask and dictionary selection strategy, which will considerably minimize the amount of memory needed to store a set of repeating random data. The sql is employed to implement the coding strategy, through higher ratio of compression than of current design.
A Resource Efficient Approximate MAC using Hierarchical Compressors with Balanced Error Control P. Balasubramani, Arthi D R 4th International Conference on Automation Computing and Renewable Systems Icacrs 2025 Proceedings, 2025 A novel approximate computing approach has been developed to achieve energy efficient Multiply Accumulate (MAC) operations for error tolerant applications. The proposed architecture employs an interleaved structure of approximate compressors to counterbalance positive and negative errors during accumulation, thereby minimizing cumulative error propagation. In the least significant bit (LSB) region, custom-designed approximate 4:2 compressors are utilized to generate complementary error patterns that maintain balanced accuracy while reducing hardware complexity. To further enhance power efficiency, 5:2 and 7:2 compressors are incorporated in the partial product reduction stage, effectively reducing the number of logic level and overall switching activity. A statistical error modeling technique is applied to optimize the placement of positive and negative compressors, ensuring consistent computational accuracy across all MAC operations. As a result, the proposed design archives over 35% reduction in processor level power consumption compared to existing multiplier architectures. Simulation and synthesis results confirm that the architecture maintains high computational stability and accuracy, making it suitable for applications such as image processing, neural network computation, and digital signal processing. Overall, the proposed interleaved compressor-based MAC unit provides an effective, scalable, and low-power solution for next-generation approximate computing systems, offering an excellent balance energy efficiency and computational precision.
Smart Harmful Gas Detection and Monitoring System using Machine Learning Balasubramani P, Aravind T, Nisanthini P, Pavithra R Proceedings of 5th International Conference on Evolutionary Computing and Mobile Sustainable Networks Icecmsn 2025, 2025 Emission of toxic gases is a significant cause of environmental degradation and health hazards to people in the industry, and efforts to monitor and control the emissions require smart approaches. This paper introduces a harmful gas capture and monitoring system with AI-enabled capabilities to monitor the emissions of a factory in real-time, combining embedded IoT with advanced machine learning. The system is based on an ESP32 microcontroller and uses major environmental sensors, MQ135 to detect harmful gases, DHT11 to measure the temperature and humidity, and a flame sensor to detect safety. The sensor data are constantly collected and sent to the ThingSpeak Cloud, which allows the seamless remote monitoring of the sensor data using a mobile platform. The data is fed into a Python-based machine learning model to detect, classify, and predict harmful gas concentrations with high precision. In case the gas concentration rises above a set of predetermined limits, an early alarm is sounded, and timely preventive and remedial measures can be undertaken. The system is a scalable and cost-efficient, proactive approach to sustainable industrial gas management and improved workplace safety, by integrating edge computing with cloud-based analytics. This interconnection underlines the promise of AI-based IoT systems in aiding data-oriented environmental decision-making and emerging intelligent industrial methods.
LoRaWAN-Driven Patient Monitoring and Accurate Location for Emergency Medical Services Balasubramani P, Swetha M, Sharu Balaa C R, Venmathi M Proceedings of 7th International Conference on Inventive Material Science and Applications Icima 2025, 2025 In emergencies, prompt medical assistance is essential. Conventional healthcare systems frequently face difficulties in swiftly evaluating and addressing patients' health situations, particularly when individuals are unable to convey their state effectively. This delay may result in adverse health consequences or even mortality. Current manual monitoring techniques may prove unfeasible in emergency situations. The research work has created an IoT-based health monitoring and GPS tracking system employing LoRaWAN technology to tackle these difficulties. This system provides ongoing monitoring of essential health metrics, including heart rate, blood oxygen level, systolic and diastolic pressures, and body temperature, ensuring prompt detection of any irregularities. Precise monitoring of the patient's location enables rapid deployment of emergency responders to the exact place. LoRaWAN technology facilitates data transmission over distances of up to 8 kilometers, guaranteeing dependable connection in remote locations. Automatic alerts to healthcare practitioners and emergency contacts following the identification of crucial health thresholds or distress signals facilitate immediate response. Through the integration of these aspects, our system ensures ongoing surveillance, swift diagnosis of health crises, accurate location tracking, and prompt medical assistance, thereby markedly enhancing patient safety and the efficacy of emergency response.
Modeling and Simulation of Smart Biogas Bottling Plant Using IoT K Satheeshkumar, M Deepak, K Dhineshkumar, P Dineshkumar, P Gopinath, P Balasubramani 2024 3rd International Conference on Trends in Electrical Electronics and Computer Engineering Teeccon 2024, 2024 This paper is to concentrate, on the generation of energy based on biodegradable wastes from domestic and cattle wastes. The intention of this paper is to attenuate biodegradable wastes and generate a feasible alternative to liquid petroleum gas (LPG). The demand for monitoring and control processes is increasing day by day due to environmental issues. Better monitoring and control algorithms can be verged and proposed in this paper by using the Internet of Things (IoT). It will be used to predict the present level of biogas to manage the input accordingly. With the help of IoT technology phones receive data instantaneously. It is one of the decentralised sources of energy in future. Retained Kitchen waste leads to public health hazards, so this paper assists in avoiding various diseases like malaria, and typhoid and also meets the social concerns in our society. This work provides a suitable means for biogas production with a smart and continuous monitoring system.
Efficient Image Transmission in Underwater Communication using OFDM Modulation P. Balasubramani, S. Suresh, S. Kiruba shankar, S. Kowsika, S. Guhan Proceedings of the 2023 2nd International Conference on Electronics and Renewable Systems Icears 2023, 2023 This design describes the transmission of aquatic images over an OFDM system. Different modulation schemes are used to transmit images over wireless technology. Due to channel fading, only a subset of carriers can be used for successful data transmission in an OFDM system. Channel state information can be used at the transmitter to best match predictive decisions to reject image frames if they are DWT-compressed in use. Compressed data is uploaded to the OFDM system. Next, examine descriptions to the correct subcarriers and to make the individual shard channel status data available at the transmitter. This indicates that sub-channels are good or bad for ocean metamorphism via OFDM channels. The descriptors assigned to currently active channels are in descending order of priority based on the sender's 1-bit channel state information. Allocations to the problematic subchannels described are omitted in the transmitter to reduce system power consumption. Through analysis accompanying Demonstration of effectiveness of proposed method by MATLAB simulation best signal-to- noise ratio and in terms of saving system performance without sacrificing quality
PIC based power loom automation using IOT International Journal of Scientific and Technology Research, 2019
