Imaging methods have broad applications in geosciences. Scanning electron microscopy (SEM) and micro-CT scanning have been applied for studying various geological problems. Despite significant advances in imaging capabilities, and image processing algorithms, acquiring high-quality data from images is still challenging and time-consuming. Obtaining a 3D representative volume for a tight rock sample takes days to weeks. Image artifacts such as noise further complicate the use of imaging methods for the determination of rock properties. In this study, we present applications of several convolutional neural networks (CNN) for rapid image denoising, deblurring and super-resolving digital rock images. Such an approach enables rapid imaging of larger samples, which in turn improves the statistical relevance of the subsequent analysis. We demonstrate the application of several CNNs for image restoration applicable to scientific imaging. The results show that images can be denoised without a priori knowledge of the noise with great confidence. Furthermore, we show how attaching several CNNs in an end-to-end fashion can improve the final quality of reconstruction. Our experiments with SEM and CT scan images of several rock types show image denoising, deblurring and super-resolution can be performed simultaneously.

Summary: The problem of solids cleanout in horizontal wellbores was studied experimentally. The special case of drilling-fluid circulation with no inner-pipe rotation was considered. This case is similar to coiled tubing (CT) drilling in which frequent hole cleanout must be performed. Sand-sized cuttings (ranging from 260 to 1240 µm) were used. Critical velocity and wall shear stress required for starting bed erosion were measured. Water and viscous-polymer base fluids with three different polymer concentrations were used. Results have shown that water always starts cuttings movement at lower flow rates than polymer solutions. Fluids with higher polymer concentration (and higher viscosity) required higher flow rates to start eroding the bed. Critical wall shear stress was also determined from pressure-loss measurements. Analyzing the data revealed that water starts cuttings removal at lower pressure loss than more-viscous fluids. Higher-viscosity fluids always showed higher pressure loss at the start of bed erosion. For the range of cuttings size studied, results show that an intermediate cuttings size was slightly easier to remove. However, the impact of cuttings size was far less than that of fluid rheology. Overall cuttings size was found to have a small impact on hole cleaning. Dimensionless analysis of parameters relevant to the process of cuttings movement was performed. It was shown that dimensionless wall shear stress (in the forms of Shields’ stress and also ratio of shear velocity to settling velocity) at the onset of bed erosion correlated well with particle Reynolds number. On the basis of this finding, two correlations were developed to predict critical wall shear stress. A procedure was developed to calculate critical flow rate as well. Friction-factor data for the flow through the annulus with a stationary cuttings bed are also reported.

Summary: An experimental study was conducted to investigate turbulent flow of water over a cuttings bed by use of a large-scale horizontal-flow loop. A nonintrusive laser-based-imaging technique was used to determine instantaneous local velocity near the stationary sandbed/fluid interface in the horizontal annulus. The velocity measured directly at the sandbed/fluid interface was then used for critical evaluation of the accuracy of the assumptions and correlations commonly used for development of mechanistic and semimechanistic sediment-transport models. In particular, effects of turbulent velocity fluctuations on the magnitude of the hydrodynamic drag and lift forces and the interfacial (bed) shear stress are investigated.