Machine learning petroelastic facies classification in a complex carbonate reservoir from Santos Basin, Brazil
H. Santana, E. Leite, J. Oliveira, N. Mattos
ARTIGO
Inglês
Agradecimentos: This research was carried out in association with the ongoing R&D project registered as ANP nº 21575-6, "Análise Integrada de Multi-Escala de Rochas Carbonáticas do Pré-Sal para Caracterização e Predição de Propriedades de Reservatórios" (Unicamp/Shell Brasil/ANP), sponsored by Shell...
Ver mais
Agradecimentos: This research was carried out in association with the ongoing R&D project registered as ANP nº 21575-6, "Análise Integrada de Multi-Escala de Rochas Carbonáticas do Pré-Sal para Caracterização e Predição de Propriedades de Reservatórios" (Unicamp/Shell Brasil/ANP), sponsored by Shell Brasil under the ANP R&D levy as "Compromisso de investimentos com Pesquisa e Desenvolvimento". We thank ANP for providing the well log dataset and Shell for providing the seismic dataset. We also thank Geosoftware for providing the Jason Workbench software license. Finally, the authors are grateful to the Institute of Geosciences at Unicamp for the infrastructure provided for the development of the research
Ver menos
Este artigo foi apresentado no evento 84th EAGE Annual Conference & Exhibition, 2023
Abstract: Understanding the distribution of rock properties in reservoir models is fundamental for planning hydrocarbon exploration and production. The objective of this study is to obtain three-dimensional distribution models of petroelastic facies in a complex carbonate reservoir in the Santos...
Ver mais
Abstract: Understanding the distribution of rock properties in reservoir models is fundamental for planning hydrocarbon exploration and production. The objective of this study is to obtain three-dimensional distribution models of petroelastic facies in a complex carbonate reservoir in the Santos Basin. The Extreme Gradient Boosting (XGBoost) algorithm was applied to classify facies from geophysical well logs and Ocean-bottom nodes (OBN) 3D seismic angle-stack data. Furthermore, additional geological attributes extracted from dataset were used to improve the classification results. The efficiency of XGBoost in the classification of petroelastic facies was verified and the use of these additional attributes is recommended to predict petrophysical or elastic properties in the pre-salt carbonate reservoirs of the Santos Basin
Ver menos
Fechado
Machine learning petroelastic facies classification in a complex carbonate reservoir from Santos Basin, Brazil
H. Santana, E. Leite, J. Oliveira, N. Mattos
Machine learning petroelastic facies classification in a complex carbonate reservoir from Santos Basin, Brazil
H. Santana, E. Leite, J. Oliveira, N. Mattos
Fontes
Proceedings of the 84th EAGE Annual Conference & Exhibition - Fonte avulsa) Bunnik : European Association of Geoscientists and Engineers, 2023. v. 2023, p. 1-5 |