Applied K-Nearest Neighbors (KNN) on Dust Supression Prototype

Abstract

Steam power plants that use coal as fuel have serious problems during operation. Before heading to the combustion process, coal is stored in an open field area. However, this results in fine particles of coal dust being exposed to wind and polluting the surrounding environment. The purpose of this study is to minimize the impact of pollution from coal dust by using the dust suppression tool. The tools that have been run manually or conventionally can be operated automatically to facilitate the operator in controlling dust suppression without the need to go to the field. This research proposes a prototype dust suppression equipped with dust and temperature sensors, the sensor data is a representation of the condition of the coal storage area which is processed using the K-Nearest Neighbors method to classify whether the condition of the storage area is normal or dusty. When conditions are dusty, the pump activates and directs bursts of water at the coal to minimize dust. In the application of the K-Nearest Neighbors method, center point 1 is obtained for normal conditions, with a dust density of 0.4353 mg / m3 and a temperature of 27.5818 °C. Whereas center point 2 for dusty conditions has a dust density of 2,374 mg / m3 and a temperature of 28.2667 °C. From 40 testing data in real-time, a success rate of 87.5% was obtained.