Pore Structure and Permeability Characterization of Tight Sandstone Reservoirs: From a Multiscale Perspective

Cheng, Youyou; Luo, Xiang Rong; Hou, Bin; Zhang, Suiwang; Tan, Chengqian; Xiao, Hui

doi:10.1021/acs.energyfuels.3c01693

Cited by 5 publications

(4 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent decades, tight oil has become a major focus of research in the global petroleum geology field. Due to the fact that the accumulation conditions and mechanisms of tight oil are significantly different from conventional reservoirs, and that various factors, such as geology, development, and engineering, contribute to challenges in tight oil horizontal well development, including rapid production decline, low individual well productivity, and low recovery rates [1,2]. Therefore, forecasting the production capacity of tight oil horizontal wells has become a crucial basis for planning and deploying unconventional oil and gas field development.…”

Section: Introduction 1traditional Methodsmentioning

confidence: 99%

Using a Deep Neural Network with Small Datasets to Predict the Initial Production of Tight Oil Horizontal Wells

Yang,

Tan,

Cheng

et al. 2023

Electronics

Self Cite

View full text Add to dashboard Cite

Due to its abundant reserves, tight oil has emerged as a significant substitute for conventional petroleum resources. It has become one of the focal points of exploration and research, and a new hot spot in global unconventional oil and gas exploration and development. This has led to a significant increase in the demand for forecasting the production capacity of tight oil horizontal wells. The deep neural network (DNN), as a mature model, has demonstrated significant advantages in many fields. However, due to the confidentiality and uniqueness of oilfield data, acquiring large datasets has become a challenge. Traditional methods using small datasets for training DNN models result in low accuracy and overfitting issues, which hinders the development of neural networks in the petroleum industry. This study aims to predict the initial production capacity of tight oil horizontal wells by using a small dataset of 650 data points through a DNN model. The research results indicate that pre-trained and fine-tuned DNNs outperform shallow neural networks, supporting vector machines, and DNN trained with traditional methods in terms of better generalization performance. Their accuracy reached 91.3%, demonstrating that it is reasonable to use a small dataset with pre-trained and fine-tuned DNN models.

show abstract

Section: Introduction 1traditional Methodsmentioning

confidence: 99%

Using a Deep Neural Network with Small Datasets to Predict the Initial Production of Tight Oil Horizontal Wells

Yang,

Tan,

Cheng

et al. 2023

Electronics

Self Cite

View full text Add to dashboard Cite

show abstract

“…Permeable rocks are desirable reservoirs because they allow for the efficient flow of oil and gas through their pores, facilitating the extraction of these resources from the subsurface . Additionally, permeability plays a crucial role in determining the relative effectiveness of different oil recovery methods. , Intrinsic permeability is solely determined by the characteristics of the medium itself and is affected by the microscopic structure of its pores. − The evaluation of the complex nature of the microscopic pore structure entails an analysis of various properties, including the distribution of pore sizes, the distribution of pore throat sizes, the connectivity of pores, and the geometry of individual pores. Every single morphological component has been shown to affect the permeability.…”

Section: Introductionmentioning

confidence: 99%

Real-Time Pore-Scale Investigation of the Effects of Uniform, Random, and Heterogenous Porous Structures on Intrinsic Permeability Using Two-Dimensional Microfluidic Chips

Boumedjane,

Al-Maamari,

Rabbani

et al. 2024

Energy Fuels

View full text Add to dashboard Cite

Permeability is a significant fluid flow property that is essential for the evaluation of oil recovery methods. The intrinsic permeability of a porous medium is influenced by factors such as the size distribution of pores and pore throats, the connectivity between pores, the topology of the pores, and the geometry of individual pores. Yet, it is not well understood how multiple elements of pore geometry interact to affect permeability due to the complex and heterogeneous nature of porous structures. This study examines the real-time porescale effects of uniform, random, and heterogeneous porous structures on intrinsic permeability, considering the broader complexity of the pore network as a single unit. The intrinsic permeability exhibited a negative correlation with increasing heterogeneity of the porous structure regardless of whether the porosities were the same or different. At a porosity of 0.59, the uniform pattern exhibited a permeability of 4.07 Darcy, while the heterogeneous pattern had a permeability of 2.37 Darcy. The random pattern demonstrated greater permeability (2.57 Darcy) at a lower porosity (0.47) in comparison to that of the heterogeneous network. Pore-scale experiments and simulations have demonstrated the significant influence of the pore network topology on the intrinsic permeability. As the complexity and heterogeneity of porous structures increase, the combined effect of morphological characteristics on intrinsic permeability becomes more significant than the individual influence of each morphological property. The differences in permeability observed in the experiments were attributed to the presence of stagnant water zones, which could be permanent or temporary. Stagnant water is more likely to be permanent in homogeneous patterns than in heterogeneous patterns. Heterogeneity appeared to promote water movement and redistribution because of variations in local pressure due to variations in pore sizes. The presence of dead-end pores in the random pattern led to a slight fluctuation in the experimental permeability. However, this effect was more prominent in the simulated permeability. Our study found experimental and numerical evidence that the broader heterogeneity of the porous structure significantly impacts permeability. Moreover, it highlights the importance of accounting for the actual microscopic porous structure of the reservoir being targeted when microfluidic devices are employed for enhanced oil recovery techniques.

show abstract

“…The alignment of fractal dimensions with pore structure characteristics serves as a validation of this pore structure classification. Cheng et al 50 employed a multiparameter equation derived via multiple regression analysis, comprehensively assessing the influence of pore-throat parameters at different scales on tight sandstone reservoir properties. It accurately predicts permeability and porosity, offering valuable insights for studying pore structure and permeability in tight sandstones.…”

Section: Introductionmentioning

confidence: 99%

Estimation of pore structure and permeability in tight carbonate reservoir based on machine learning (ML) algorithm using SEM images of Jaisalmer sub-basin, India

Yalamanchi,

Datta Gupta

2024

Sci Rep

View full text Add to dashboard Cite

Analyzing the pore structure in carbonate reservoirs plays a crucial role in predicting fluid flow characteristics within these formations. The goal of the study was to use machine learning techniques for pore structure analysis and estimation of permeability in carbonate reservoirs. We implemented these algorithms by examining 2D scanning electron microscope (SEM) images of carbonate samples from the Jaisalmer sub-basin captured at various magnifications. In the initial stage of the analysis, various binarization algorithms were applied to determine carbonate sample porosity. Among these algorithms, the MaxEntropy algorithm gave a porosity value closely aligned with those obtained through petrography analysis. We employed the watershed algorithm to find the pore network parameters of carbonate samples at various magnifications. We observed that changes in magnification affected pore network parameters, resulting in a reduction in pore size distribution, throat radius, and grain size. Subsequently, we employed the numerical lattice Boltzmann method (LBM) to estimate the permeability of carbonate samples and compared to values derived from well logs. We employed machine learning (ML) algorithms, specifically Artificial Neural Network (ANN) and Support Vector Machine (SVM), to predict the permeability of carbonate samples. The input features for these models were the pore network parameters, while the LBM permeability values served as the output. We examine the prediction performance of these methods against the measured LBM permeability by conducting the error analysis and the coefficient of determination ($${R}^{2}$$ R 2 ) calculation. Our findings revealed that the ANN models outperformed the SVM models. Specifically, the ANN model displayed an impressive R2 value of 0.892, along with root mean square error (RMSE), mean squared error (MSE) and, mean absolute error (MAE) values of 1.927, 3.716 and 1.580, respectively. In contrast, the SVM model yielded an R2 value of 0.849, with RMSE, MSE and, MAE values of 2.324, 5.401 and, 2.166 respectively, when assessed on testing data of measured permeability. This study found that ANN is more dependable, robust, and precise than SVM in forecasting carbonate sample permeability.

show abstract

Pore Structure and Permeability Characterization of Tight Sandstone Reservoirs: From a Multiscale Perspective

Cited by 5 publications

References 40 publications

Using a Deep Neural Network with Small Datasets to Predict the Initial Production of Tight Oil Horizontal Wells

Using a Deep Neural Network with Small Datasets to Predict the Initial Production of Tight Oil Horizontal Wells

Real-Time Pore-Scale Investigation of the Effects of Uniform, Random, and Heterogenous Porous Structures on Intrinsic Permeability Using Two-Dimensional Microfluidic Chips

Estimation of pore structure and permeability in tight carbonate reservoir based on machine learning (ML) algorithm using SEM images of Jaisalmer sub-basin, India

Contact Info

Product

Resources

About