Clustering of the water characteristics of the Cirata reservoir using the k-means clustering method
Isma Masrofah, Bramantiyo Eko Putro · AIP conference proceedings · 2020
Pollution prevention must be carried out in the upstream, middle and downstream parts. One of the Citarum watersheds located in the middle part is the Cirata reservoir. Pollution in the Cirata reservoir does not only originate from the reservoir environment, but also from rivers that flow into the Cirata reservoir. Cirata Reservoir besides being a Hydroelectric Power Plant unit, this reservoir is also used by the community for fish farming in the cage aquaculture technique. This study aims to know the water characteristics Cirata reservoir. Through data on the content of water entering the Cirata reservoir, the characteristics of water at each point can be identified. The research data was obtained by obtaining clean water quality testing data from several estuary points of the river flow towards the Cirata reservoir and 4 centers of the Cirata reservoir. Data acquisition is done by taking primary data as well as secondary data related to water quality. Data processing techniques are carried out using Data Mining K-Means Clustering. This study obtained three groups of clusters, namely clusters. The first cluster is Organic Compounds Cluster which is dominated by the presence of organic compounds such as compounds that contain nitrogen. The second is Dominant Minerals Cluster that tends to be dominated by the presence of content of rock and soil minerals. Lastly is Middle Area Cluster which is characterized by the presence of mud rock sediments and dissolved organic matter into the river flow.