AI Navigate
インサイト最新記事一覧AI大全

The impact of sensor placement on graph-neural-network-based leakage detection

arXiv cs.LG / 2026/3/26

💬 オピニオンSignals & Early TrendsIdeas & Deep AnalysisModels & Research
原文を読む →
共有:

要点

  • The paper examines how the choice and arrangement of sensor locations affect the accuracy of graph-neural-network (GNN) based leakage detection in water distribution systems.
  • It shows that GNN performance for tasks like pressure reconstruction and prediction—and ultimately leak detection—can vary strongly depending on sensor measurements and sensor configuration.
  • The authors introduce a PageRank-Centrality-based sensor placement strategy designed to improve measurement placement for downstream GNN leakage detection.
  • Experiments on EPANET Net1 demonstrate that sensor placement can substantially influence reconstruction, prediction, and leak detection results.
  • The work frames sensor placement as a critical design variable when deploying GNN leakage detection in real water-utility contexts, not just a data-collection detail.

Abstract

Sensor placement for leakage detection in water distribution networks is an important and practical challenge for water utilities. Recent work has shown that graph neural networks can estimate and predict pressures and detect leaks, but their performance strongly depends on the available sensor measurements and configurations. In this paper, we investigate how sensor placement influences the performance of GNN-based leakage detection. We propose a novel PageRank-Centrality-based sensor placement method and demonstrate that it substantially impacts reconstruction, prediction, and leakage detection on the EPANET Net1.

関連おすすめサービス

※当サイトはアフィリエイト広告を利用しています
🎧

Notta搭載AI議事録イヤホン ZENCHORD1

AI時代の仕事術。Notta搭載で会議の議事録を自動生成するスマートイヤホン。

🎙️

AI搭載ボイスレコーダー Plaud

世界100万人が愛用。AIで文字起こし・要約を自動化するボイスレコーダー。

🖼️

画像高画質化AIツール Aiarty Image Enhancer

AIで画像を高画質化。写真・イラストを簡単にアップスケール。