Session: 11-03-01 Data Science Applications in Drilling

Submission Number: 156491

Optimization and Performance Assessment of Decanter Centrifuges for Low-Gravity Solids Removal in Drilling Operations 

The solids control system is an essential component of the drilling-rig’s circulation system. Although shale shakers do the heavy lifting in drilled solids separation, due to the screen size limitations, fine and ultra-fine drilled solids remain in the mud, and they need to be separated. Decanter centrifuges are frequently used for this purpose. Weighting materials are relatively easier to separate. The removal of fine and ultra-fine drilled solids is a challenging process. Failure of removal leads to several dysfunctions, including decreasing ROP, increase risk of formation damage, undesired increase of ECD etc. The aim of this study is to provide a practical yet accurate workflow to first determine the decanter performances, and subsequently a mean to optimize its operational parameters. A large experimental campaign of 100 tests was conducted using a decanter centrifuge (BOS SOLUTIONS DE-1000 VFD). Additional data were also gathered from a past work which used other two industrial decanter centrifuges. The experimental data were used to train, test, and validate a Random Forest classifier meant to help automate the classification between efficient and not-efficient removal for a given set of centrifuge operational conditions. Moreover, an improved mechanistic model was developed (and validated against test data) to determine the theoretical decanter centrifuge low-gravity solids removal efficiency. The modified mechanistic model is then used as the foundation for the presented decanter’s operational parameters optimization tool.

Presenting Author: Silvio Baldino The University of Tulsa

Presenting Author Biography: Silvio Baldino is a petroleum engineering expert with a Ph.D. from the University of Tulsa, backed by over eight years of experience in data science, drilling engineering, and geomechanics. Currently, he serves as an Assistant Professor and Associate Director of the Tulsa University Drilling Research Projects (TUDRP), where he teaches courses in drilling engineering, carbon capture, and geomechanics. His industry experience includes roles as a Data Engineer at Cognite AS and Research Engineer at Shell, where he developed real-time predictive analytics and data-driven drilling solutions. Proficient in Python, SQL, and data integration, Silvio has contributed to cost-saving drilling technologies and holds multiple technical certifications.