Physics‑Informed Neural Network Modelling of Water Distribution Systems for Integrated Flow and Quality Simulation
Abstract
We propose a unified physics informed neural modelling framework for the simultaneous emulation of hydraulic states and water quality in water distribution systems (WDS). Building on recent advances in physics informed graph neural networks for hydraulics and spatio temporal PINN-GNN hybrids for chlorine transport, our approach (PINNWDSFQ) couples a physics aware Graph Neural Network (GNN) emulator of hydraulic variables (flows, heads, pressures) with a spatio temporal graph PINN for reactive advective transport. The hydraulic GNN employs message passing based on conservation laws to recover nodal/edge flow fields with a small number of layers and low computational cost, matching high fidelity simulator outputs while enabling fast inference across network sizes. The quality module discretizes pipes with virtual nodes and imposes PDE constraints (advection-reaction) using physics informed loss terms in a GNN encoder-processor-decoder, thereby preserving network topology and chemical reaction kinetics. Both modules are jointly trained in a multitask loss that includes supervised data terms (sparse sensor readings or EPANET surrogate labels) as well as physics constraints and consistency coupling (mass/flux conservation at nodes and flow-transport coupling). We show that the hybrid model obtains emulator level accuracy for hydraulics and state-of-the-art water quality RMSE/MAE (order 1e-3-1e-2 mg/L), scales to large networks with thousands of virtual nodes and runs orders of…
Read the paper
Related studies
- Preliminary Toxicity Analysis of 3DCRT versus IMRT on the High Dose Arm of the RTOG 0126 Prostate Cancer Trial
- Adaptive Radiotherapy for Head and Neck Cancer: Initial Clinical Outcomes from a Prospective Trial
- American Association of Physicists in Medicine Task Group 263: Standardizing Nomenclatures in Radiation Oncology
- Motion Management Strategies and Technical Issues Associated with Stereotactic Body Radiotherapy of Thoracic and Upper Abdominal Tumors: A Review from NRG Oncology
- High-dose-rate interstitial brachytherapy as a monotherapy for localized prostate cancer: treatment description and preliminary results of a phase I/II clinical trial
- Highly efficient training, refinement, and validation of a knowledge-based plan quality control system for radiotherapy clinical trials
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.