Optimization and improvement of water anomaly measurement based on DBSCAN clustering approach

Abstract

In the context of coping with the leakage problems of water supply systems brought about by urbanization and population growth, this study employs DBSCAN density clustering analysis, truncated mean analysis, and descriptive statistics to construct a generalized mathematical model for the detection and identification of outlier points in water supply flow data. By applying these methods, we integrated the distribution characteristics of the data, the outlier identification of the truncated mean method and the noise point detection of the density clustering algorithm. The experimental results show that the model performs well in terms of rationality and sensitivity, providing strong support for improving the efficiency and sustainability of flow anomaly identification in smart water systems. This research provides innovative and practical solutions to practical problems in urban water supply systems.