Graph Neural ODE Digital Twins for Control-Oriented Reactor Thermal-Hydraulic Forecasting Under Partial Observability

2026-04-08

AI Summary

This work addresses the challenge of achieving real-time, high-fidelity prediction of whole-plant thermal-hydraulic states in advanced reactors under partial sensor observability. To this end, the authors propose a digital twin model that integrates physics-informed graph neural networks with neural ordinary differential equations (GNN-ODE). The model captures continuous-time system dynamics through fluid/heat-transfer-aware message passing and employs a topology-guided missing-node initializer to enable autoregressive forecasting. This study presents the first application of GNN-ODE coupling to reactor thermal-hydraulic modeling, demonstrating millisecond-level inference—105× faster than conventional simulation—alongside high predictive accuracy (MAE of 0.91 K and 2.18 K for 60 s and 300 s temperature forecasts, respectively) and excellent reconstruction fidelity for unobserved states (R² = 0.995). Moreover, the model exhibits an emergent capacity to learn constitutive relationships, successfully recovering Reynolds-number scaling laws and accurately tracking power-step transients.

Abstract

Real-time supervisory control of advanced reactors requires accurate forecasting of plant-wide thermal-hydraulic states, including locations where physical sensors are unavailable. Meeting this need calls for surrogate models that combine predictive fidelity, millisecond-scale inference, and robustness to partial observability. In this work, we present a physics-informed message-passing Graph Neural Network coupled with a Neural Ordinary Differential Equation (GNN-ODE) to addresses all three requirements simultaneously. We represent the whole system as a directed sensor graph whose edges encode hydraulic connectivity through flow/heat transfer-aware message passing, and we advance the latent dynamics in continuous time via a controlled Neural ODE. A topology-guided missing-node initializer reconstructs uninstrumented states at rollout start; prediction then proceeds fully autoregressively. The GNN-ODE surrogate achieves satisfactory results for the system dynamics prediction. On held-out simulation transients, the surrogate achieves an average MAE of 0.91 K at 60 s and 2.18 K at 300 s for uninstrumented nodes, with $R^2$ up to 0.995 for missing-node state reconstruction. Inference runs at approximately 105 times faster than simulated time on a single GPU, enabling 64-member ensemble rollouts for uncertainty quantification. To assess sim-to-real transfer, we adapt the pretrained surrogate to experimental facility data using layerwise discriminative fine-tuning with only 30 training sequences. The learned flow-dependent heat-transfer scaling recovers a Reynolds-number exponent consistent with established correlations, indicating constitutive learning beyond trajectory fitting. The model tracks a steep power change transient and produces accurate trajectories at uninstrumented locations.

Problem

• thermal-hydraulic forecasting
• partial observability
• reactor control
• digital twins
• surrogate modeling

Innovation

• Graph Neural ODE
• Digital Twin
• Partial Observability
• Physics-Informed Learning
• Thermal-Hydraulic Forecasting