Optimización en la planeación de pozos por medio de la predicción de tiempos, costos y NPT´S, aplicando un modelo de machine learning para la campaña de perforación de Castilla y Castilla norte 2020

Currently the company Ecopetrol SA, takes into account the technical-historical data such as those stored in OpenWells and Power BI, to evaluate the performance during the drilling phase of the wells week by week, instead of taking advantage of this information together with the variables involved in the complexity matrix, to optimize well planning through the implementation of a predictive model, thus generating added value on the stored information. Considering the above, the present degree work was carried out in order to optimize well planning for the 2020 Castilla y Castilla Norte drilling campaign by applying the selected machine Learning models, which predict cost days and NPT's associated with problems in open hole. Therefore, a methodology aimed at the implementation of three supervised machine learning models was designed, based on the information from the 2019 Castilla y Castilla Norte drilling campaign. Subsequently, the prediction of the models was implemented and evaluated. for the same field in 2020.

• Fecha de publicación: 2021
• Institución: Universidad de América
• Repositorio: Lumieres
• Idioma: spa
• OAI Identifier: oai:repository.uamerica.edu.co:20.500.11839/8274
• Acceso en línea: https://hdl.handle.net/20.500.11839/8274
• Palabra clave: Matriz complejidad; Método supervisado; Predicción de tiempos; Complexity matrix; Supervised method; Time prediction; Tesis y disertaciones académicas
• License: Atribución – No comercial – Sin Derivar