Implementación de un modelo predictivo de machine learning para la estimación de los parámetros óptimos de la ROP y la MSE en la sección 8½’’ y 12 ¼’’ para los pozos perforados con motor de fondo en el Campo Yarigui – Cantagallo durante el 2019

The implementation of a predictive machine learning model to estimate optimal drilling parameters arises from the need for the industry to migrate towards data science seeking to optimize processes. Through this research project, a database corresponding to the wells drilled with a downhole motor during 2019 in the aforementioned field was generated, which was subjected to an exploratory data analysis (EDA). Following this, it was divided for standardization and testing of the predictive model. Once this division was made, a supervised automatic learning algorithm was implemented such as Random Forest Regressor, having as input variables the revolutions per minute (RPM) of the surface and the bottom, the weight on the bit (WOB), the flow rate. (Q), the torque (TQ) and the information corresponding to the tops of the drilled geological formations, and the penetration rate (ROP) and the specific mechanical energy (MSE) were obtained as output variables.

• Fecha de publicación: 2021
• Institución: Universidad de América
• Repositorio: Lumieres
• Idioma: spa
• OAI Identifier: oai:repository.uamerica.edu.co:20.500.11839/8269
• Acceso en línea: https://hdl.handle.net/20.500.11839/8269
• Palabra clave: Energía mecánica específica; Modelo predictivo; Perforación de pozos; Specific mechanical energy; Predictive model; Well drilling; Tesis y disertaciones académicas
• License: Atribución – No comercial