Efficient well placement optimization under uncertainty using a virtual drilling procedure

Abstract An Automatic Well Planner (AWP) is used to efficiently adjust pre-determined well paths honor near-well properties and increase overall production. AWP replicates modern geosteering decision-making where adjustments pre-programmed are driven by continuous integration of data obtained from logging-while-drilling look-ahead technology. In this work, combined into a robust optimization scheme develop trajectories that follow reservoir in more realistic manner compared common representations for purposes. Core operation relies on an artificial neural network coupled with geology-based feedback mechanism. Specifically, each path candidate outer-loop procedure, customizes according the particular geological realization ensemble models. While placement searches typically rely linear representations, develops customized moving sequentially heel toe. Analog drilling operations, determines subsequent trajectory points processing neighboring information. Studies performed using Olympus ensemble. two derivative-free algorithms Asynchronous Parallel Pattern Search (APPS) Particle Swarm Optimization (PSO), implemented NTNU’s open-source framework FieldOpt. Results show that, both APPS PSO, solutions outperform straight-line parameterization all three tested scenarios, which varied simplest scenario sole producer single-realization environment full multiple producers.