Using sensitivity kernels in realistic borehole conditions for informing ahead-of-the-bit prediction of hard stringers

Drilling operations can unexpectedly encounter hard stringers-thin, high-strength inclusions/rock layers in a softer background-which cause drilling problems including stick-slip vibrations, near-bit inclination changes, and severe damage to drill-bit and bottomhole assembly (BHA), all leading to inefficient and costly drilling. Early detection of hard stringers allows drilling personnel to adjust parameters proactively, enhancing operational stability. In this study, we propose a methodology referred to as ahead-of-the-bit prediction (ABP) using drill-bit-generated noise as a seismic source and BHA-mounted sensors for cross-correlation analysis of recorded signals. We compute sensitivity kernels in a realistic borehole environment to identify the contributions from direct arrivals, stringer reflections, mud-induced guided waves, and to better understand the physics of the elastic wavefield. The results from this work will enable further development of our methodology for real-time early detection of hard stringers during drilling.