Uncertainty assessment in reservoir performance prediction using a two-stage clustering approach: Proof of concept and field application

Haddadpour, Hosein; Niri, Mohammad Emami

doi:10.1016/j.petrol.2021.108765

Cited by 13 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Concerning geological distributions, multiple generated geological realizations may exhibit similarities at certain distances, resulting in comparable flow responses. The ability to distinguish these realizations enables the generation of simulation outputs based on a reduced subset of realizations, each representing diverse flow responses [24].…”

Section: Euclidean Distance Measurement Between Realizationsmentioning

confidence: 99%

“…Porosity and permeability, integral petrophysical properties within the RQI formulation, hold particular significance in the context of fluid flow modeling in heterogeneous porous media [41,42], especially in the modeling of CO 2 storage and plume migration [43,44]. Studies by [16,24] have explored the effectiveness of RQI in identifying realizations with similar flow behaviors, which yield promising results. The mathematical definition of RQI is given by,…”

Section: Euclidean Distance Measurement Between Realizationsmentioning

confidence: 99%

“…Machine learning and deep learning have recently been explored in many fields to accelerate computationally heavy processes [17][18][19][20]. In this area, UML specifically stands out as a widely adopted method to select RGRs, as evidenced in several case studies [21][22][23][24]. UML is designed to uncover the underlying structures in unlabeled data, employing techniques such as dimensionality reduction and clustering [25,26].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Uncertainty Quantification in CO2 Trapping Mechanisms: A Case Study of PUNQ-S3 Reservoir Model Using Representative Geological Realizations and Unsupervised Machine Learning

Mahjour,

Badhan,

Faroughi

2024

Energies

View full text Add to dashboard Cite

Evaluating uncertainty in CO2 injection projections often requires numerous high-resolution geological realizations (GRs) which, although effective, are computationally demanding. This study proposes the use of representative geological realizations (RGRs) as an efficient approach to capture the uncertainty range of the full set while reducing computational costs. A predetermined number of RGRs is selected using an integrated unsupervised machine learning (UML) framework, which includes Euclidean distance measurement, multidimensional scaling (MDS), and a deterministic K-means (DK-means) clustering algorithm. In the context of the intricate 3D aquifer CO2 storage model, PUNQ-S3, these algorithms are utilized. The UML methodology selects five RGRs from a pool of 25 possibilities (20% of the total), taking into account the reservoir quality index (RQI) as a static parameter of the reservoir. To determine the credibility of these RGRs, their simulation results are scrutinized through the application of the Kolmogorov–Smirnov (KS) test, which analyzes the distribution of the output. In this assessment, 40 CO2 injection wells cover the entire reservoir alongside the full set. The end-point simulation results indicate that the CO2 structural, residual, and solubility trapping within the RGRs and full set follow the same distribution. Simulating five RGRs alongside the full set of 25 GRs over 200 years, involving 10 years of CO2 injection, reveals consistently similar trapping distribution patterns, with an average value of Dmax of 0.21 remaining lower than Dcritical (0.66). Using this methodology, computational expenses related to scenario testing and development planning for CO2 storage reservoirs in the presence of geological uncertainties can be substantially reduced.

show abstract

Section: Euclidean Distance Measurement Between Realizationsmentioning

confidence: 99%

Section: Euclidean Distance Measurement Between Realizationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Uncertainty Quantification in CO2 Trapping Mechanisms: A Case Study of PUNQ-S3 Reservoir Model Using Representative Geological Realizations and Unsupervised Machine Learning

Mahjour,

Badhan,

Faroughi

2024

Energies

View full text Add to dashboard Cite

show abstract

“…Many literatures proposed a method depending on interior index comparisons as one of the initial attempts. [30], on the other hand, offered a twostage clustering approach, recommending that clusters be identified first based on the Ward's approach or the average interconnections (i.e. hierarchical classification), then cluster refining using non-hierarchical classification.…”

Section: Two-stage Clustering and Data Sizementioning

confidence: 99%

Survey on Technique and User Profiling in Unsupervised Machine Learning Method

Kristijansson

Aegisson

2022

JMC

View full text Add to dashboard Cite

In order to generate precise behavioural patterns or user segmentation, organisations often struggle with pulling information from data and choosing suitable Machine Learning (ML) techniques. Furthermore, many marketing teams are unfamiliar with data-driven classification methods. The goal of this research is to provide a framework that outlines the Unsupervised Machine Learning (UML) methods for User-Profiling (UP) based on essential data attributes. A thorough literature study was undertaken on the most popular UML techniques and their dataset attributes needs. For UP, a structure is developed that outlines several UML techniques. In terms of data size and dimensions, it offers two-stage clustering algorithms for category, quantitative, and mixed types of datasets. The clusters are determined in the first step using a multilevel or model-based classification method. Cluster refining is done in the second step using a non-hierarchical clustering technique. Academics and professionals may use the framework to figure out which UML techniques are best for creating strong profiles or data-driven user segmentation.

show abstract

“…Later, the use of AI in this field expanded. Up to this date, many successes have been made in cases such as predicting reservoir properties (i.e., permeability, matrix porosity, and fracture) in dual porosity reservoirs (Alajmi and Ertekin, 2007), history matching for a hydrocarbon field (Haghshenas et al, 2020(Haghshenas et al, , 2021Kolajoobi et al, 2021;Shahkarami et al, 2014), well placement optimization (e.g., Kolajoobi et al, 2023), uncertainty evaluation in reservoir performance prediction (Haddadpour and Emami Niri, 2021), CO2 storage (Van Si and Chon, 2018;Vo Thanh et al, 2020), and Hydrogen Storage (Rahimi et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

Design of Artificial Neural Network for predicting the reduction in permeability of porous media as a result of polymer gel injection

Kamel Targhi,

Emami Niri,

L.J. Zitha

2023

Geoenergy Science and Engineering

Self Cite

View full text Add to dashboard Cite

Uncertainty assessment in reservoir performance prediction using a two-stage clustering approach: Proof of concept and field application

Cited by 13 publications

References 17 publications

Uncertainty Quantification in CO2 Trapping Mechanisms: A Case Study of PUNQ-S3 Reservoir Model Using Representative Geological Realizations and Unsupervised Machine Learning

Uncertainty Quantification in CO2 Trapping Mechanisms: A Case Study of PUNQ-S3 Reservoir Model Using Representative Geological Realizations and Unsupervised Machine Learning

Survey on Technique and User Profiling in Unsupervised Machine Learning Method

Design of Artificial Neural Network for predicting the reduction in permeability of porous media as a result of polymer gel injection

Contact Info

Product

Resources

About