Appraising Fracture-Related Mud Losses During Drilling: Fact or Fiction

Abstract Mud losses during drilling have significantly increased the cost of wells, particularly in extended-length horizontal wells. It is widely understood that natural fractures and faults contribute to these mud losses. This solution leverages on Natural Language Processing (NLP) technique to extract structured information from unstructured information in the drilling reports seeks to understand the causes and implications of these losses and to propose solutions for mitigating such issues in future drilling operations. The study interprets seismic data to identify faults and fractures, using Gradient Coherency Cube and Variance attributes to highlight features and confirm seismic interpretations with borehole image data, Production Logging Tool (PLT), production and injection data, Extended Leak Off (XLOT) and Micro-Frac tests and pressure measurements. Analyzing mud losses around the offset wells reveals their extent and severity. Natural Language Processing (NLP), the process of automatically extracting structured information from unstructured text is used to simplify and expedite the mundane task of extracting information from drilling reports. This involves identifying and pulling out specific pieces of data like names, dates, locations, and relationships, transforming vast amounts of text into organized, usable information from all the wells in the reservoir. The analysis highlights the necessity of considering event sequences when linking mud losses to faults or fractures. It mentions that mud circulation might penetrate these features in earlier uncased sections, though most identified fractures are only partially conductive. A mud contamination issue starting at earlier depth caused pressure fluctuations and increased losses. The presence of CO2 and oil suggests that formation fluids have infiltrated the mud, altered its properties and reduced the mud weight. This, combined with pressure fluctuations and mechanical erosion, can degrade the mud filter cake, exposing the borehole to further contamination and losses. The significant increase in dynamic mud losses, while static loss remains low, indicates that mud may have opened natural and new fractures during circulation. Raising mud weight could create excessive overbalance, risking integrity of the formation.