An efficient MCMC history matching workflow using fit-for-purpose proxies applied in unconventional oil reservoirs

Dachanuwattana, Silpakorn; Yu, Wei; Sepehrnoori, Kamy

doi:10.1016/j.petrol.2019.01.070

Cited by 13 publications

(2 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They estimated the coefficient of fluid composition in the numerical model, which is a 20-dimensional inverse problem. Dachanuwattana et al [5] developed a surrogate-assisted inverse modeling method for shale reservoirs, and they compared the accuracy of quadratic polynomial, cubic polynomial, k-nearest neighboring, and kriging models. Their results showed that the kriging model has the best effect.…”

Section: Introductionmentioning

confidence: 99%

Deep Graph Learning-Based Surrogate Model for Inverse Modeling of Fractured Reservoirs

Ma,

Zhao,

Zhou

et al. 2024

Mathematics

View full text Add to dashboard Cite

Inverse modeling can estimate uncertain parameters in subsurface reservoirs and provide reliable numerical models for reservoir development and management. The traditional simulation-based inversion method usually requires numerous numerical simulations, which is time-consuming. Recently, deep learning-based surrogate models have been widely studied as an alternative to numerical simulation, which can significantly improve the solving efficiency of inversion. However, for reservoirs with complex fracture distribution, constructing the surrogate model of numerical simulation presents a significant challenge. In this work, we present a deep graph learning-based surrogate model for inverse modeling of fractured reservoirs. Specifically, the proposed surrogate model integrates the graph attention mechanisms to extract features of fracture network in reservoirs. The graph learning can retain the discrete characteristics and structural information of the fracture network. The extracted features are subsequently integrated with a multi-layer recurrent neural network model to predict the production dynamics of wells. A surrogate-based inverse modeling workflow is then developed by combining the surrogate model with the differential evolutionary algorithm. Numerical studies performed on a synthetic naturally fractured reservoir model with multi-scale fractures illustrate the performance of the proposed methods. The results demonstrate that the proposed surrogate model exhibits promising generalization performance of production prediction. Compared with tens of thousands of numerical simulations required by the simulation-based inverse modeling method, the proposed surrogate-based method only requires 1000 to 1500 numerical simulations, and the solution efficiency can be improved by ten times.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep Graph Learning-Based Surrogate Model for Inverse Modeling of Fractured Reservoirs

Ma,

Zhao,

Zhou

et al. 2024

Mathematics

View full text Add to dashboard Cite

show abstract

“…There are several studies conducted for basement reservoirs such as the matrix and fracture characterization (Dakhelpour-Ghoveifel et al 2018;Saboorian-Jooybari et al 2015, 2016, well test analysis (Dejam et al 2018;Mashayekhizadeh et al 2011;Zhang et al 2018) and overview of geological and production characteristic (Gutmanis 2009). Moreover, the history matching process is one of the challenges for unconventional and basement reservoirs (Azim 2016;Dachanuwattana et al 2019;Dang et al 2011;Jeong et al 2013;León Carrera et al 2018;Nguyen et al 2011). These authors proposed an efficient workflow to handle the difficulty of modeling and history matching in fractured basement reservoirs.…”

Section: Introductionmentioning

confidence: 99%

Applying the hydrodynamic model to optimize the production for crystalline basement reservoir, X field, Cuu Long Basin, Vietnam

Thanh

Sugai

et al. 2019

J Petrol Explor Prod Technol

View full text Add to dashboard Cite

Weathered and fractured crystalline basement is known as the important unconventional reservoir in the Cuu Long Basin. Naturally fractured reservoir plays a crucial role in oil exploration to contribute for hydrocarbon production in Vietnam. However, the complexity and heterogeneity of the fractures system in the basement reservoir are challenges for oil and gas production. They require the realistic simulation scenarios to estimate the hydrocarbon potential as well as field development plan of these reservoirs. Thus, this paper aims to propose the feasibility development scenarios to improve oil recovery factor for crystalline basement reservoir, X field, Cuu Long Basin, Vietnam. First, history matching process is validated for the model to fit the actual production data (reservoir pressure, pressure, water cut in each well) in order to approach closer the fluid flow behavior through the reservoir. The manual matching was selected to adjust the actual aquifer size and permeability distribution with limit simulation runs. Next, the highest reliability matching model which approximately reflects the actual fluid flow behavior can be used as the base case to forecast the future reservoir performance through the field development plan. The most potential scenario is to add six new infill production wells, two side track wells and two water injection wells. The forecasted results indicate that this scenario yields 8% more oil recovery factor compared to the natural drive with thirteen producers. This result suggests that the precise field development plan is to increase the efficiency of the production process by increasing the displacement parameters of residual oil and reservoir sweep efficiency by stimulation. The major contribution of this paper demonstrates the merits of the field development plan in fractured basement reservoir. The findings of this study can help better understand the fluid flow behavior using the production history profiles and field development scenarios of crystalline basement reservoir of Cuu Long Basin.

show abstract

Introduction and literature review

Tripoppoom¹,

Yu²,

Sepehrnoori³

et al. 2021

Assisted History Matching for Unconventional Reservoirs

View full text Add to dashboard Cite

An efficient MCMC history matching workflow using fit-for-purpose proxies applied in unconventional oil reservoirs

Cited by 13 publications

References 5 publications

Deep Graph Learning-Based Surrogate Model for Inverse Modeling of Fractured Reservoirs

Deep Graph Learning-Based Surrogate Model for Inverse Modeling of Fractured Reservoirs

Applying the hydrodynamic model to optimize the production for crystalline basement reservoir, X field, Cuu Long Basin, Vietnam

Introduction and literature review

Contact Info

Product

Resources

About