IoT-based Online Condition Monitoring of Power Transformer

Conventional power system monitoring techniques mostly rely on human observation, which is error-prone and requires ongoing supervision. However, the successful transformation of a traditional grid to a smart grid requires smarter monitoring and control. A crucial component like transformer needs continuous monitoring to increase its efficiency and reliability thereby reducing its loss and downtime. The development of IoT (Internet of Things) technology provides a revolutionary way to improve monitoring practices. The preventive analytics is made easier by the incorporation of IoT, allowing early detection of any problems and prompt resolution to avoid downtime. This study attempts to develop an advanced online condition monitoring system for power transformers in an open-source cloud-based IoT platform, ThingSpeak. A prototype is developed in the lab using Wi-Fi integrated Node MCU microcontroller with a capability to keep an eye on vital indicators including voltage, current, temperature, humidity and oil level by integrating a variety of sensors. A novel technique of capacitance based oil level detector is also designed and presented in this paper, which can be a potential replacement of the mercury based Buchholz relay. The real-time data visualization and analysis capabilities are shown by using ThingSpeak platform as well as a customized MIT App developed for easier access to real-time data in mobile phone or any other handheld device. Additional feature of SMS notifications via IFTTT (If This Then That) is integrated in the proposed model to alarm and trigger action at the time of anomalies.