Saeid Jamshidi

[This corrects the article DOI: 10.1021/acsomega.2c05688.].

The increase in the application of unconventional resources and the presence of heterogeneity in such reservoirs have increased the need to consider multi-scale models. Some of the recent studies indicate that sub-resolution porosity (SRP) has a profound influence on the flow characteristics of porous. Thus, the objective of this work is to present a technique to restore lost details from images of porous media. Interpolation and filtering are some of the conventional techniques that have long been used to improve image resolution. However, this field has changed with the current development of artificial intelligence, especially generative adversarial networks (GAN). GANs consist of two neural networks: A generator and a discriminator, which are trained in an adversarial manner to produce realistic images. When applied to petroleum engineering, GANs are employed in super-resolution tasks, in which the GANs learn to reconstruct high-resolution images from low-resolution inputs. In particular, a synthetic dataset of low-resolution images is generated from an existing dataset of a sandstone sample. In this research, we employ RealESRGAN which is more effective than previous models such as SRGAN. Finally, we constructed a pore network based on generated images and the model's results were almost identical to the actual model.

Summary: Integrated reservoir-production modeling is a collaborative multidisciplinary tool that can facilitate optimization of oil and gas production operations during the field development planning stage of exploiting subsurface resources. The critical issue with this technique is the excessive computational burden of the large integrated model with many input variables, which has not been effectively addressed to date. This study aims to reduce the computational costs and runtimes associated with the production integration and optimization process from oil fields. To do so, the reservoir and the surface network models of an Iranian oil field were coupled to create an integrated model for the optimization of field parameters to achieve the highest oil production rate. In the first step of simplification, polynomial chaos expansion (PCE) was used to establish a surrogate model from the integrated system. Next, Sobol sensitivity analysis, which is a variance-based, global, and model-free sensitivity analysis technique, was performed to reduce the number of input variables by identifying the most influential variables. Finally, the optimization was implemented using genetic algorithm (GA) on the PCE surrogate model of the integrated system with the most important variables. The results from the case study showed that the integrated model can be replaced with the PCE surrogate model while the accuracy is maintained. Moreover, performing sensitivity analysis considerably decreased the number of input variables for optimization by revealing their significance. The proposed methodology in this study can substantially improve the computational efficiency of the optimization for the integrated reservoir-production system.

In this study, the potential of using a polyacrylamide-silica nanocomposite (PAM-S) to control the filtration properties of bentonite water-based drilling muds under different salinity conditions was evaluated. Static filtration tests under low-pressure/low-temperature (LPLT) conditions accompanied by rheological measurements have been carried out to analyze the role of silica nanoparticles (NPs) and nanocomposites (NCs) in the base fluid properties. Moreover, high-pressure/high-temperature (HPHT) static filtration was also investigated to evaluate the thermal stability of PAM-S. Afterward, dynamic filtration has been conducted in a filtration cell equipped with an agitating system with a disk-type impeller to investigate the hydrodynamic and formation of a filter cake under shear flow conditions. Fluid flow velocity and wall shear stress (WSS) distribution over the filter cake were analyzed using an exact 3D computational fluid dynamic (CFD) simulation. A transparent filtration cell with a camera was used to accurately record the fluid flow field inside the filter press and validate the CFD results. The obtained results indicated that adding silica NPs at a concentration of less than 2 wt % increases the fluid loss due to reducing rheological properties such as yield point. While silica NPs could not significantly change the mud properties, the experimental results showed that, under both LPLT and HPHT conditions, the PAM-S NC could reduce the total filtration loss by 70% at a low concentration of 0.75 wt %. Moreover, during dynamic filtration, the results indicated that there is a linear relationship between the cake thickness and the inverse of WSS at different operating pressures. However, no correlation could be found between predeposited mud cake erosion and WSS. At a rotating disk speed of 1000 rpm, more than 60% of the predeposited mud cake was eroded after 30 min for a saline mud sample while for the NC-treated mud sample cake erosion is considerably reduced and reaches up to 20% at 1.5 wt % PAM-S.