Scalable and Robust Physics-Informed Graph Neural Networks for Water Distribution Systems

2025-02-11

AI Summary

Urban water distribution network (WDN) planning, expansion, and rehabilitation under climate change and population growth require scalable, robust, and real-time decision support. Method: We propose a physics-informed graph neural network (PI-GNN) surrogate model featuring a novel GNN architecture, constraint-aware physical consistency training, and physics-preserving data normalization—explicitly embedding hydraulic conservation laws and pipe physical constraints. Results: The model significantly outperforms state-of-the-art deep learning methods across multiple real-world WDN datasets, scales to million-node networks, and exhibits strong out-of-distribution generalization to demand surges and pipe diameter distribution shifts. It establishes a new paradigm for high-fidelity, computationally efficient surrogate modeling in complex water systems, enabling real-time operational and strategic decision-making.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks

Abstract

Water distribution systems (WDSs) are an important part of critical infrastructure becoming increasingly significant in the face of climate change and urban population growth. We propose a robust and scalable surrogate deep learning (DL) model to enable efficient planning, expansion, and rehabilitation of WDSs. Our approach incorporates an improved graph neural network architecture, an adapted physics-informed algorithm, an innovative training scheme, and a physics-preserving data normalization method. Evaluation results on a number of WDSs demonstrate that our model outperforms the current state-of-the-art DL model. Moreover, our method allows us to scale the model to bigger and more realistic WDSs. Furthermore, our approach makes the model more robust to out-of-distribution input features (demands, pipe diameters). Hence, our proposed method constitutes a significant step towards bridging the simulation-to-real gap in the use of artificial intelligence for WDSs.

Problem

• Develop scalable deep learning for water systems
• Enhance robustness in AI for infrastructure
• Improve physics-informed graph neural networks

Innovation

• Improved graph neural network architecture
• Adapted physics-informed algorithm
• Physics-preserving data normalization method