Assessing the Relation between Mud Components and Rheology for Loss Circulation Prevention Using Polymeric Gels: A Machine Learning Approach

The traditional way to mitigate loss circulation in drilling operations is use preventative and curative materials. However, it difficult quantify the amount of materials from every possible combination produce customized rheological properties. In this study, machine learning (ML) used develop a framework identify material composition for applications based on desired characteristics. relation between properties mud components polyacrylamide/polyethyleneimine (PAM/PEI)-based assessed experimentally. Four different ML algorithms were implemented model data various at concentrations testing conditions. These four include (a) k-Nearest Neighbor, (b) Random Forest, (c) Gradient Boosting, (d) AdaBoosting. Boosting showed highest accuracy (91 74% plastic apparent viscosity, respectively), which can be further hydraulic calculations. Overall, experimental study presented paper, together with proposed ML-based framework, adds valuable information design PAM/PEI-based mud. models allowed wide range rheology assessments fluid formulations mean up 91%. case has shown that appropriate materials, reasonable could achieved prevent by managing equivalent circulating density (ECD).