Optimal Control of Dynamic Gas Networks Using Gaussian Processes with Chance-Constraint Reformulation
Abstract
Natural gas networks must accommodate variable demand and supply while maintaining pressure across pipelines whose stored gas couples operating decisions through time. Probabilistic surrogates can simplify compressor scheduling, but their predictive uncertainty must remain reliable where optimization uses it. We combine direct learning of the minimum pressure margin with assessment at optimizer-selected schedules. Using a warped Gaussian process, we express the posterior pressure requirement as one inequality in its predictive mean and standard deviation. We then parameterize block and Fourier schedules so scalar predictions and analytic derivatives replace trajectory states and time integration in whole-day optimization. In steady state, a surrogate that passes held-out accuracy and coverage checks nevertheless falls below the minimum pressure on 71% of its own dispatches at a nominal 1% tolerance, as the optimizer is drawn to schedules where the margin is overestimated. This diagnostic motivates an enrichment loop that audits selected dispatches against the physics and updates the surrogate by exact conditioning, with fitted quantities held fixed. The loop restores steady-state minimum-pressure risk control at every studied tolerance, with conditioning about four orders of magnitude faster than refitting. In transient operation, we complement this enrichment with empirical threshold calibration to address residual hourly optimism, bringing hourly violation rates closer to their declared tolerances with little compressor work overhead.
Related
- Fraud Detection via Bayesian Positive-Unlabeled Learning with Gaussian Processes and Multilayer Networks
- Optimal Control Under Signal Temporal Logic Constraints: A Robust Reformulation via Augmented Dynamics
- A Dantzig-Wolfe reformulation of the EndoWake model for optimal wind farm layouts
- Deterministic and Random Bipartite Matching on General Networks: Convex Flow Reformulation, Asymptotic Properties, and Fast Algorithms
- Optimal Multitask Representations Under a Rate Constraint: An Information Bottleneck Analysis
BibTeX
@misc{adebiyi2026optimal,
title = {{Optimal Control of Dynamic Gas Networks Using Gaussian Processes with Chance-Constraint Reformulation}},
author = {Taiwo A. Adebiyi and Robert Ferrando and Manuel Garcia and Ruda Zhang and Saif R. Kazi and Kaarthik Sundar},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.10963},
eprint = {2610.10963},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.10963},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Optimal Control of Dynamic Gas Networks Using Gaussian Processes with Chance-Constraint Reformulation
AU - Taiwo A. Adebiyi
AU - Robert Ferrando
AU - Manuel Garcia
AU - Ruda Zhang
AU - Saif R. Kazi
AU - Kaarthik Sundar
PY - 2026
DA - 2026/10/07
PB - arXiv
DO - 10.48550/arxiv.2610.10963
UR - https://arxiv.org/abs/2610.10963
AB - Natural gas networks must accommodate variable demand and supply while maintaining pressure across pipelines whose stored gas couples operating decisions through time. Probabilistic surrogates can simplify compressor scheduling, but their predictive uncertainty must remain reliable where optimization uses it. We combine direct learning of the minimum pressure margin with assessment at optimizer-selected schedules. Using a warped Gaussian process, we express the posterior pressure requirement as one inequality in its predictive mean and standard deviation. We then parameterize block and Fourier schedules so scalar predictions and analytic derivatives replace trajectory states and time integration in whole-day optimization. In steady state, a surrogate that passes held-out accuracy and coverage checks nevertheless falls below the minimum pressure on 71% of its own dispatches at a nominal 1% tolerance, as the optimizer is drawn to schedules where the margin is overestimated. This diagnostic motivates an enrichment loop that audits selected dispatches against the physics and updates the surrogate by exact conditioning, with fitted quantities held fixed. The loop restores steady-state minimum-pressure risk control at every studied tolerance, with conditioning about four orders of magnitude faster than refitting. In transient operation, we complement this enrichment with empirical threshold calibration to address residual hourly optimism, bringing hourly violation rates closer to their declared tolerances with little compressor work overhead.
ER -