Analysis of Waterflooding Through Application of Neural Networks
Notice bibliographique
Résumé
Abstract Petroleum reservoirs demonstrate a very complex behavior that changes with time in a non-linear manner. Application of neural networks for field-wise analysis of waterflooding projects is very appropriate because a structural model between injection and production does not need to be specified in order to predict performance. The neural network approach recognizes that individual well behavior may depend on the well history and the injection/production conditions of surrounding wells. The outcome of this neural network analysis could determine injection and production policies that would lead to determining the minimum injection water leading to maximum oil recovery. This paper presents application of neural network for analyzing data from a Canadian oil field that has been under waterflooding for several years. At first, production data for the last 20 years were obtained. Currently, there are 13 injection and 108 production wells in this pool. This neural network model uses this data and divides the field into several areas based on the performance of waterflooding, which helps the field engineers to focus on parts of the field that waterflooding is not very effective. Additionally, the neural network developed in this study is capable of predicting future oil recovery due to waterflooding. Introduction Recent progress in the available computational power and better understanding of the theory of neural networks has gain increasing attention of engineers and researchers working in petroleum industry. The ability techniques, such as neural network and fuzzy logic, to work with noisy data and solve problems even if information related to detailed physics of the system is not known or the system is too complex to be solved by traditional formal methods has provided new means of addressing these complex processes. Artificial intelligence, and both fuzzy logic and neural networks in particular, can give the petroleum industry new tools for better understanding and controlling recovery processes and therefore achieving efficient and profitable oil recovery[1,2]. One of the most widely used processes in mature oil fields is waterflooding. Field-wise management of these waterflooding processes provides several important challenges. Some of the questions facing engineers and mangers during designing and operating waterflood projects are; the location and pattern used for injection wells, amount and rate of water injection, and so on. Reservoir simulation is the common tool utilized to deal with these issues, however reservoir simulation cannot be used extensively because it is both time consuming and expensive. Lack of knowledge about the reservoir formations and detailed geology of these fields adds additional difficulty to correctly simulate oil fields. Fuzzy logic and neural networks can offer an alternative solution. Today most of oil reservoirs are under production for many years and a lot of production information has been cumulated over time. Using the ability of the neural networks to approximate relationships without knowing the exact mechanisms involved we can "simulate" the reservoir behavior and its response to the changes in recovery parameters. There is no "best" type of the network. Each network topology has its own advantages and disadvantages.
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 machine sur la base complète
Imitation des enseignantsNi prévalence calibrée, ni vérité terrain. Validation humaine à venir. Le volet Gemma est une étiquette directe du modèle pour chaque travail de la base, lue sur la notice réduite au titre. Le volet Codex est un classifieur appris des 10 348 étiquettes directes de Codex et calibré sur les taux pondérés de l'échantillon; les champs sans appui suffisant ne portent aucun appel Codex. Le mode candidate est l'union des deux volets; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont pas des étiquettes humaines.
Scores du classifieur distillé par catégorie (deux têtes)
| Catégorie | Codex | Gemma |
|---|---|---|
| Métarecherche | 0,001 | 0,002 |
| Méta-épidémiologie (sens strict) | 0,001 | 0,000 |
| Méta-épidémiologie (sens large) | 0,000 | 0,000 |
| Bibliométrie | 0,001 | 0,001 |
| Études des sciences et des technologies | 0,000 | 0,000 |
| Communication savante | 0,001 | 0,000 |
| Science ouverte | 0,000 | 0,000 |
| Intégrité de la recherche | 0,001 | 0,000 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,001 | 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 source (Gemma direct ou Codex distillé), 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 ».