Fremgangsmate for optimalisering av produksjon i et oljereservoar med usikkerheter

摘要

An intrinsic parameter of the deposit, and a parameter connected with well development options, are selected. These parameters influence hydrocarbon production from the well. An analytical model is developed expressing the criterion of production over time, as a function of the selected parameters. A finite number of values of production criteria are taken into account, the values being obtained from the production flow simulator. Starting from the analytic model so determined, an uncertainty function is associated with one of the parameters intrinsic to the deposit. A distribution of one of the parameters connected to options of well production development is determined, such that the criterion of production is optimized. Before association of the uncertainty function, the relative mutual influence of parameters is quantified. Parameters having negligible influence on production are removed. The quantification employs a statistical test, either the Student- or the Fisher test. One of the parameter values is fixed and the value of a parameter connected with development options is determined, such that production criteria are optimized. Further variants are proposed, based on the preceding principles. Neural network modeling is employed. At least one intrinsic parameter of the deposit is of discrete, continuous and/or stochastic type.