Development of machine learning and tailored image compression algorithms in beehive monitoring system

Abstract

Namanve Thermal Power Plant is a 50 MW generation plant currently being run and maintained by Uganda Electricity Generation Company limited. The generation plant is used as an emergency backup power source for situations when the country experiences deficiencies in the power supply from the main generation plants. At the plant, there are several units that function together during the process of energy production. However, there are occurrences in certain units that could lead to a failure in the whole system. For example, the engines are designed using pneumatically operated valves which are operated using compressed air supplied by the air compressor unit. In case the air compressor trips or malfunctions, the engines will not function and therefore there will be no energy output from the plant. The operator at the plant however faces a challenge of a lack of an efficient monitoring system that enables him or her clearly tell the status of the instrument air compressor unit. This project therefore seeks to design and implement a monitoring system for the instrument air compressor unit using IoT technologies. The proposed system will enable the operator monitor the status of the air compressor in real-time from the control room. The key parameters that our system will focus on include the voltage and current levels of the instrument air compressor unit supply, the status of the phases and the pressure levels of the compressed air storage cylinders.