Stream-DBSCAN: A Streaming Distributed Clustering Model for Water Quality Monitoring

With the increasing use of wireless sensor networks in water quality monitoring, an enormous amount streaming data is generated by widely deployed sensors. However, current batch mode used for analysis can no longer meet diverse combination monitoring indicators and requirement timely results on all-weather basis. To overcome these challenges analyze a large quickly accurately, we propose stream-DBSCAN distributed stream processing clustering model. First, real-time streams are processed using computing framework Flink. Then, DBSCAN algorithm applied to cluster each dataset as different dimension cluster. Finally, time distribution characteristics same analyzed identify variation rules. The system extract noise points sudden deterioration quality. We tested model datasets three indices, pH, ammonia nitrogen (NH4N), turbidity, Yantai Menlou Reservoir from May August 2019. demonstrate that efficiently perform data. By analyzing results, found daily pollution events consistent with actual situation.