Deep Learning Surrogates for Real-Time Gas Emission Inversion

2025-06-17

AI Summary

Addressing the challenge of real-time identification and quantification of greenhouse gas emissions under dynamic atmospheric conditions, this paper proposes a spatiotemporal Bayesian inversion framework integrating a deep learning surrogate model with sequential Monte Carlo (SMC) sampling. Methodologically, a multilayer perceptron (MLP) surrogate replaces computationally expensive computational fluid dynamics (CFD) solvers within the probabilistic inversion pipeline, enabling millisecond-scale inference while preserving physical consistency—thereby overcoming the longstanding trade-off between real-time performance and robustness. Evaluated on the Chilbolton methane release dataset, the method achieves source localization and flux estimation errors below 8%, matching the accuracy of full CFD-based inversion while accelerating inference by three orders of magnitude. Moreover, it maintains a source localization success rate exceeding 92% even in complex flow fields with multiple obstacles.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Real-time identification and quantification of greenhouse-gas emissions under transient atmospheric conditions is a critical challenge in environmental monitoring. We introduce a spatio-temporal inversion framework that embeds a deep-learning surrogate of computational fluid dynamics (CFD) within a sequential Monte Carlo algorithm to perform Bayesian inference of both emission rate and source location in dynamic flow fields. By substituting costly numerical solvers with a multilayer perceptron trained on high-fidelity CFD outputs, our surrogate captures spatial heterogeneity and temporal evolution of gas dispersion, while delivering near-real-time predictions. Validation on the Chilbolton methane release dataset demonstrates comparable accuracy to full CFD solvers and Gaussian plume models, yet achieves orders-of-magnitude faster runtimes. Further experiments under simulated obstructed-flow scenarios confirm robustness in complex environments. This work reconciles physical fidelity with computational feasibility, offering a scalable solution for industrial emissions monitoring and other time-sensitive spatio-temporal inversion tasks in environmental and scientific modeling.

Problem

• Real-time identification of greenhouse-gas emissions under dynamic conditions
• Bayesian inference of emission rate and source location in dynamic flows
• Replacing costly CFD solvers with fast deep-learning surrogates

Innovation

• Deep-learning surrogate replaces CFD solvers
• Sequential Monte Carlo enables Bayesian inference
• Multilayer perceptron captures spatio-temporal dispersion