Optimizing Horizontal Well Placement Through Stratigraphic Performance Prediction Using Artificial Intelligence

Abstract

Abstract Accurate predictions of connectivity and heterogeneity pose important technical challenges for successful maturation of conventional and unconventional reservoirs. We present the success of a new reservoir management workflow that uses both artificial intelligence and classic models to define the impact of stratigraphic connectivity and heterogeneity on horizontal-well production performance in a mature heavy oil field. The data-driven model based on fuzzy logic was used to compute a new attribute named dynamic Reservoir Quality Index (dRQI). The classical models used the stratigraphic Lorenz Plots, Reservoir Quality Index (RQI) and Flow-Zone indicator (FZI). Workflows were validated through a lookback process on more than 400 wells used to predict the fine-scale stratigraphic and directional heterogeneities within intervals targeted by horizontal wells, and production performance. The workflow was successfully used to optimize the horizontal well placement for 2019-2020 drilling programs.