Graph convolutional networks based contamination source identification across water distribution networks

Water distribution Networks (WDNs) are one of the most important infrastructures for modern society. Due to accidental or malicious reasons, water contamination incidents have been repeatedly reported all over world, which not only disrupt supply but also endanger public health. To ensure safety WDNs, quality sensors deployed across WDNs real-time detection and source identification. In literature, various methods employed improve performance identification (CSI) recent studies show that there is a great potential tackle CSI problem by deep learning models. The success based often requires large size training samples being collected. real-world situations, number events occurring in single WDN rather small, especially newly built WDN. However, existing literature mostly focus on study applying models same knowledge gained from cannot be reused different these ends, application graph convolutional networks, this paper provides solution cross-network can transfer learned Empirically, benchmark task identification, we proposed method achieve comparable accuracy even trained