A REFERENCE PATTERN-BASED DATA AGGREGATION FRAMEWORK FOR IOT-ENABLED TRANSFORMER HEALTH MONITORING: ENHANCING ENERGY EFFICIENCY AND FAULT DETECTION

Abstract

The transformers are important components of electrical power system but their power overloading, aging of infrastructure and severe thermal conditions results in great maintenance problem in these components. The internet of things (IoT) based Transformer health monitoring system (THMS) has come into picture for fault detections, early prediction and maintenance, but the limitation in large scale remote transformer health monitoring systems is energy supply and communication. The proposed work introduces novel reference pattern-based data aggregation framework (RPDAF) for energy-efficient wireless sensor networks data aggregation with the integration of the transformer health monitoring using internet of things (IoT). The proposed method is to calculate compact pattern codes of measured parameters from sensors through a leftmost digit extraction method which will act as a reference and then send the small codes rather than whole data to the cloud for efficient data collection and also for the intelligent reduction of data before sending them to the cloud. Unlike the currentESPDA method, where computational intelligence is needed due to lookup table, the proposed method uses reference values that have been transmitted through the base-station to all sensor nodes. Experimental result using ns-2.33simulation and practical experiment on NodeMCU- Firebase implementation proves the proposed framework provide an average of 5-10% energy savings with respect to other schemes and increases the lifetime of sensor node by around 40%. In the proposed scheme we get accuracy of above 95% for fault detections with reduced response time of 60% for maintenance. Hence, the proposed framework provides a significant and efficient solution for scalable, energy efficient and timely fault detection of transformer health in present electrical power system, especially for countries with mixed portfolio and energy resources such as Pakistan