Abstract

In drilling engineering, mud cake is formed when drilling fluid invades the near-wellbore zone, which can reduce drilling fluid leakage and reservoir pollution and maintain wellbore stability. To study the three-dimensional spatial characteristics of mud cakes, such as heterogeneity, particle accumulation, and porous media, CT scan images of cores and mud cakes before and after drilling fluid intrusion are used as the research object. The composite filtering and segmentation method for typical regions of images based on grayscale values and the batch segmentation method for large regions of images based on deep learning are investigated. On this basis, a visualized digital mud cake is constructed, and microscopic evaluation indexes such as the percentage, equivalent radius, throat length, sphericity, and coordination number of pores and particles in the mud cake are established. The application results show that the number of isolated pores in the outer mud cake formed by static filtration of drilling fluid is more than that of dynamic ones, the connectivity of pores is poor, and the particles have a better sealing effect on pores. The combined effect of the outer and inner mud cake results in different effects on core intrusion. The construction technique of digital mud cake can study the structure at a micron scale. This technique can provide more detailed guidance for engineering situations, such as optimization of drilling fluid systems in fractured formations, damage evaluation of reservoirs by drilling fluid, and influence of mud cake on cementing quality.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call