A Graph-Enhanced DeepONet Approach for Real-Time Estimating Hydrogen-Enriched Natural Gas Flow under Variable Operations

2025-04-09

AI Summary

Accurate real-time estimation of hydrogen molar fraction in hydrogen-blended natural gas (HENG) networks under varying operating conditions remains challenging, jeopardizing operational safety. Method: This paper proposes a graph-enhanced DeepONet framework: (i) it encodes pipeline topology as a graph and embeds it into the branch network to incorporate physical priors; (ii) it introduces a dual-network coordination mechanism that decouples dynamic operating-condition representation from sparse sensor-based modeling. The method integrates graph neural networks (GNNs), deep operator learning, and sensor-driven data modeling. Contribution/Results: Experiments demonstrate a 37.2% reduction in hydrogen molar fraction estimation error under diverse transient conditions compared to conventional end-to-end models. The framework achieves millisecond-scale state inference for large-scale HENG networks, significantly enhancing generalizability and interpretability—establishing a new paradigm for safe, intelligent regulation of HENG systems.

Technology Category

• Intelligent Robots: State Estimation
• Machine Learning: Graph-based Machine Learning
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing

Abstract

Blending green hydrogen into natural gas presents a promising approach for renewable energy integration and fuel decarbonization. Accurate estimation of hydrogen fraction in hydrogen-enriched natural gas (HENG) pipeline networks is crucial for operational safety and efficiency, yet it remains challenging due to complex dynamics. While existing data-driven approaches adopt end-to-end architectures for HENG flow state estimation, their limited adaptability to varying operational conditions hinders practical applications. To this end, this study proposes a graph-enhanced DeepONet framework for the real-time estimation of HENG flow, especially hydrogen fractions. First, a dual-network architecture, called branch network and trunk network, is employed to characterize operational conditions and sparse sensor measurements to estimate the HENG state at targeted locations and time points. Second, a graph-enhance branch network is proposed to incorporate pipeline topology, improving the estimation accuracy in large-scale pipeline networks. Experimental results demonstrate that the proposed method achieves superior estimation accuracy for HCNG flow under varying operational conditions compared to conventional approaches.

Problem

• Estimating hydrogen fraction in natural gas pipelines accurately
• Improving adaptability to variable operational conditions
• Enhancing real-time flow estimation in large-scale networks

Innovation

• Graph-enhanced DeepONet for real-time HENG flow estimation
• Dual-network architecture for operational condition characterization
• Graph-enhance branch network incorporating pipeline topology