PORE-SCALE VISUALIZATION ON POLYMER FLOODING: APPLICATION OF SINGULAR VALUE DECOMPOSITION-BASED IMAGE ANALYSIS METHOD

Through years of working with visualization studies on micromodels and Hele-Shaw cells, we have always been confronted with difficulties such as numerous images of long flooding processes to be analyzed, irregularities of laser-etched micromodels that complicate the analysis of pore images, poor lighting, and so on. Finally, we aimed to customize a simple method addressing our difficulties. The method is based on singular value decomposition coupled with contour tracing and is developed using the MATLAB programming language. It is built with an approach that allows easy application of the experience of an expert when it is needed. Singular value decomposition (SVD) has been utilized because of its attractive properties that are useful to our objectives, including image compression and image denoising. The method is experimented on pore-scale visualization of polymer flooding in a micromodel, and it showed a reliable performance. In this paper, the pore-scale analysis is chosen rather than calculating the overall oil recovery factor because it has more details to investigate and helps better to evaluate the performance of the proposed method. We calculated the residual oil and connate water saturation as a case study. The method's unique procedure and features are explained step by step through the case study. The method facilitates the analysis by reducing the calculation time and the required storage space. Also, it offers an interesting component to decrease redundancies due to lighting problems.