Temperature Monitoring System on Distribution Transformer with Web Thermal Camera

Abstract

<p><em>Temperature monitoring using a thermal camera is one type of light inspection that is used as an indicator of the health of a transformer. The system can detect the temperature of a transformer without having to stick directly to the surface of the transformer. The results of temperature measurements from the sensor are displayed in the form of images that can be displayed on a web. In this final project, a transformer temperature monitoring system is made that continues to work in real time to measure the temperature of the transformer. This is because changes in the temperature of the transformer can occur at any time. With this system, workers are expected to always monitor the condition of the transformer without having to make onsite visits. In this final project, ESP32 is used as a microcontroller and MLX90641 thermal camera. The use of a thermal camera in this system is caused by several advantages, namely the thermal camera can take the temperature of all parts of the transformer captured on the camera. In addition, the thermal camera has a range of up to seven meters. In this system, the temperature capture is set at the hottest point of all visible camera pixels. This system is also equipped with a notification system that will work when the temperature of the hamster transformer exceeds 60</em><em>℃</em><em>. The normal temperature threshold of the distribution transformer is 30</em><em>℃</em><em>-80</em><em>℃</em><em>. Or the maximum increase in one day of 40</em><em>℃</em><em> </em><em>from the initial temperature. The temperature of the transformer's hottest point will trigger a notification display in a web. So that after receiving a notification the officer can immediately know the temperature of the transformer and immediately take action. From the results of experiments that have been carried out, it is known that the results of measuring the temperature value of the thermometer and sensor have an accuracy rate of 98.8%. In addition, the system also works well, this is indicated by the color difference that occurs when the transformer temperature is in normal and hot conditions. So that from the image displayed, the officer can already know that the measured transformer is in normal or overheated conditions.</em></p>