Automatic Partitioning of Water Distribution Networks Using Multiscale Community Detection and Multiobjective Optimization

Partitioning water distribution networks into district metered areas (DMAs) is a challenge for water utilities due to the complexity of network topology, especially for large-scale and highly looped water supply systems. This paper presents a novel approach, using complex network theory and multiobjective optimization, for efficiently partitioning of water distribution networks. In this approach, a water distribution network is first mapped on to a weighted undirected graph, and its strength is measured by an extended modularity by a random walk strategy. Then, the Louvain algorithm is employed for network partitioning based on the modularity. Finally, boundary pipes between DMAs are determined by applying an evolutionary algorithm to minimize the number of boundary pipes and maximize the network pressure and water age uniformity. The approach has been tested on a real-world distribution system to demonstrate the effectiveness of the method. The results indicate that the method is effective at partitioning the network into DMAs with a similar pressure and water age.