Please use this identifier to cite or link to this item:
|Type:||Artigo de periódico|
|Title:||Analysis Of Methodological Impacts Of The Estimation Of Porosity In Oil Reserves By Means Of A Monte Carlo Simulation [análise De Impactos Da Metodologia De Estimativa Da Porosidade Em Reservas De Petróleo Por Meio De Simulação De Monte Carlo]|
|Abstract:||In many papers dealing with estimation of oil reserves, engineers usually assume that well porosity can be modeled as a Gaussian distribution, that is, under this assumption the expected value of porosity can be estimated from the average porosity values from well log and petrophysical data. But, other distributions can be used to model local porosity when Gaussian distribution cannot fit sample data. In this paper, using actual porosity data of a 3-NA-002-RJS well from the Campos Basin, it is shown that for a selected interval, the logistic distribution fits the data better than other distributions and its expected value should be used to estimate the well porosities of the entire population. In such cases, as numerical analysis shows, using arithmetic mean instead of expected value may give rise to errors. The data shows that using an average as porosity estimator will overestimate the P90 and underestimate the P10 estimates.|
|Appears in Collections:||Unicamp - Artigos e Outros Documentos|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.