Automatic optimization method of horizontal well formation model based on natural gamma while drilling

In recent years, the decline of conventional oil and gas resources in China has accelerated the exploration and development of unconventional reservoirs, leading to the rapid advancement of horizontal well technology. Accurate construction of horizontal stratigraphic models plays a critical role in horizontal well log interpretation; however, existing approaches often depend on complex and labor-intensive manual procedures. To address this issue, this study proposes a fast forward simulation method for natural gamma-ray logging based on localized model construction. By employing either the Monte Carlo method or azimuthal sensitivity matrices, the computational efficiency is significantly enhanced compared with conventional full-scale formation simulations. In addition, an automatic generation framework for synthetic gamma-ray logging data is developed to support data-driven interpretation and geosteering applications. Furthermore, an automated while-drilling formation model optimization method based on gamma-ray logging is introduced. The proposed approach has been successfully applied to real horizontal well data from an oilfield in Jilin, China. The results demonstrate efficient and accurate reconstruction of formation profiles, improved formation identification accuracy, expanded application potential, and strong technical support for horizontal well development in complex geological environments.