In the past, the reservoir material balance (voidage) effects occurring between the end of the measured (known) production history and future Inflow Performance Relationship (IPR) time levels have been commonly ignored in the computation of the future IPR behavior. Neglecting the reservoir voidage that occurs during the time interval between the end of the known production history and the future IPR time levels results in erroneous estimates of the future IPR behavior.
A detailed description is given of the mathematically rigorous technique that has been used in the development of a multilayer well performance simulator that properly accounts for the reservoir voidage effects. Some of the more significant results are also presented of an extensive effort to develop an accurate and computationally efficient well performance simulation model.
The reservoir can be considered to be multilayered, with mixed reservoir layer completion types and outer boundary shapes, drainage areas and boundary conditions. The well performance model can be used to simulate performance in three different operating modes:
constant wellhead rate,
constant bottomhole pressure, and
constant wellhead pressure.
The transient performance of vertical, vertically fractured and horizontal wells can be simulated with this well performance model. The well performance model uses mathematically rigorous transient solutions and not simply the approximate solutions for each of the well types, as do most of the other commercially available well performance models.
Numerous investigators have reported analysis procedures for estimating the inflow performance of a well at future time levels. A comprehensive study that reported a technique for estimating future Inflow Performance Relationship (IPR) behavior was given by Brown and coworkers. The assumptions of infinite-acting, steady-state or pseudosteady-state flow behavior in the reservoir and steady-state conditions in the wellbore are required for this analysis technique to be applicable.
A common limitation of the analysis procedure reported by Brown is that the flow regime exhibited by the reservoir at a given time level must be known and the appropriate reservoir inflow performance model has been used. An invalid assumption has also been employed in much of the earlier work on well performance modeling in which the Dirichlet inner boundary condition (constant pressure) solutions have been evaluated from the transient results of the Neumann inner boundary condition (constant rate) solutions.