RESUMEN
Learning operators with deep neural networks is an emerging paradigm for scientific computing. Deep Operator Network (DeepONet) is a modular operator learning framework that allows for flexibility in choosing the kind of neural network to be used in the trunk and/or branch of the DeepONet. This is beneficial as it has been shown many times that different types of problems require different kinds of network architectures for effective learning. In this work, we design an efficient neural operator based on the DeepONet architecture. We introduce U-Net enhanced DeepONet (U-DeepONet) for learning the solution operator of highly complex CO2-water two-phase flow in heterogeneous porous media. The U-DeepONet is more accurate in predicting gas saturation and pressure buildup than the state-of-the-art U-Net based Fourier Neural Operator (U-FNO) and the Fourier-enhanced Multiple-Input Operator (Fourier-MIONet) trained on the same dataset. Moreover, our U-DeepONet is significantly more efficient in training times than both the U-FNO (more than 18 times faster) and the Fourier-MIONet (more than 5 times faster), while consuming less computational resources. We also show that the U-DeepONet is more data efficient and better at generalization than both the U-FNO and the Fourier-MIONet.
RESUMEN
Over the past few years, there has been significant interest in the potential of hybrid nanoparticle-acid fluid (HNAFs) for improved oil recovery. This comprehensive study investigates the effects of nanoparticles and acid on interfacial tension (IFT) to establish a relationship between brine properties and the oil/brine IFT. This investigation is one of the first regional studies conducted utilizing candidate field data from the Middle East. Based on the literature review and screening studies conducted, a seawater (SW)-based HNAF was formulated with nanoparticles (SiO2, Al2O3, and ZnO) and HCl to measure their effect on IFT. A total of 48 formulations of HNAFs, nanofluids with and without acid, were analyzed with crude oil from a candidate field. IFT measurements were subsequently conducted using the pendant drop method under ambient conditions and in a high-pressure, high-temperature reservoir environment. Results showcased that IFT reduction was observed by increasing the acid concentration with SiO2 and Al2O3, although a reverse trend was observed with ZnO. Moreover, it was observed that IFT varied with increasing concentrations of nanoparticles, and at certain acid concentrations, IFT reduced significantly with higher nanoparticle concentrations. From the Amott studies, a clear signature was achieved, with ZnO exhibiting a total of 31.4% oil recovery, followed by SiO2 (27.3%) and Al2O3 (23.7%). The results of this study may assist in defining a screening criterion for future displacement (oil recovery) studies involving the presented nanoparticles. The results also reveal further the mechanisms involved in IFT reduction by hybrid nano-acid fluids and their potential for significant applications in the Middle East.