A System Identification Approach for Spatiotemporal Prediction of CO2 Storage Operation in Deep Saline Aquifers
Notice bibliographique
Résumé
Abstract Estimation of subsurface storage performance before obtaining storage credit is a key requirement in development of a CO2 sequestration hub. Traditionally, reservoir modelling tools have been used, in similar engineering applications. However, physics-based models are computationally expensive for early decision making processes, particularly in deep saline aquifers due to large geographical spread and limited geological data. In this work, we present a system identification approach to rapidly emulate the geological CO2 storage operation. Leveraging 8 years of field performance data at the Aquistore CO2 injection site, we built non-isothermal EOS-based fluid flow simulations; multiple realizations were calibrated with periodic monitoring data of downhole injection rate, pressure, and temperature. We then tested the possibility of applying system identification techniques based on the proper orthogonal decomposition (POD). The POD models were formulated to include spatial variations in petrophysical properties, irregular boundaries, and multiple CO2 injection inputs. Additionally, we included multiple synthetic realizations of CO2 injection into saline aquifer with dependent and independent variables. The training and validation of POD models included a robust and complete data sets of the Aquistore injectivity performance at multitemporal resolutions, and time-lapse seismic surveys from the storage and overlying caprock formations. The accuracy and efficiency of POD models were measured using multiple quantitative metrics, including global root mean squared error, training time, and forecast time. POD was found a powerful tool to reduce the spatiotemporal dimensionality of the large Aquistore dataset and to speed up the training process. It also produced acceptable global errors compared to the scale of the downhole measured responses. The infographics of the entire pressure/saturation/temperature field variations, using a set of basis vectors and time-varying coefficients, indicated that POD is capable to capture the CO2 plume shape. The visualization of POD predictions suggested to employ smaller grid size around the injection well for higher accuracy and larger grid size near the model boundary for higher efficiency (when generating the training set using CMG GEM simulator). The Aquistore dataset included multiple injection and shut-in periods during the past 8 years. This highlighted the significance of multi-temporal issues when the inclusion of finer time resolutions during start and end of CO2 injection improves the performance and accuracy of POD models. However, when CO2 plume extent is significantly large compared to the reservoir boundaries, the number of time steps needed proper management to keep the training process within a reasonable time. POD-based proxy models offer huge reduction in data dimensionality and computational time. The results from our system identification approach using the Aquistore field data deliver insights into handling the multi-temporal multi-spatial nature of dynamic input data for prediction of CO2 storage performance; this approach has potential applications in other subsurface energy systems.
Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.
Comment cette classification a été obtenuedéplier
Prédiction distillée sur la base complète
Imitation des enseignantsNi prévalence calibrée, ni vérité terrain. Validation humaine à venir. Apprise à partir de 10 348 étiquettes directes de Codex et de 10 348 étiquettes directes de Gemma. Le mode candidate est l'union des têtes enseignantes seuillées; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont ni des étiquettes humaines ni des étiquettes directes de modèles de pointe.
Scores Codex et Gemma par catégorie
| Catégorie | Codex | Gemma |
|---|---|---|
| Métarecherche | 0,000 | 0,000 |
| Méta-épidémiologie (sens strict) | 0,000 | 0,000 |
| Méta-épidémiologie (sens large) | 0,000 | 0,000 |
| Bibliométrie | 0,000 | 0,000 |
| Études des sciences et des technologies | 0,000 | 0,000 |
| Communication savante | 0,000 | 0,000 |
| Science ouverte | 0,000 | 0,000 |
| Intégrité de la recherche | 0,000 | 0,000 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,000 | 0,000 |
Scores machine (provisoires)
Les deux têtes enseignantes du modèle étudiant, lues sur ce travail. Un score ordonne la base pour la relecture; il n'affirme jamais une catégorie, et le statut de validation accompagne chaque rangée tel quel.
Scores de référence d'un modèle non mature (critères de maturité non atteints, 7 itérations). Un score ordonne; il n'affirme jamais une catégorie.
score_only:v0-immature-baseline · tel quel depuis la passe de notation : score_only signifie que le nombre peut ordonner les travaux, et qu'aucune étiquette de catégorie n'en découleClassification
machine, non validéePrédiction automatique; un appel candidat d’une seule tête enseignante, pas un consensus.
Le détail, modèle par modèle et score par score, se trouve en fin de page sous « Comment cette classification a été obtenue ».