Abstract
This work presents a novel approach through the use of the connected reservoir storage model to accurately predict production performance of gas reservoirs, which have a complex reservoir geometry and have a large uncertainty of reservoir energy support. Based on the deconvolution of measured pressures and rate data from a well, a gas reservoir performance will be modeled using a group of normalized quantities with which the wells behavior can be accurately predicted in all flow regimes. No foreknowledge of reservoir geometries, petro-physical properties or fluid properties are required in the development of this analysis.
Reservoir storage is a time function defined as the product of the total compressibility and the reservoir volume at a particular time. Unlike with classical or curve fitting decline curve analysis, the connected reservoir storage model is based on the pressure diffusivity theory by the normalized production rate and normalized cumulative production volume through deconvolution methods. With the knowledge of reservoir volumes, the long term well performance of gas reservoirs can be predicted.
Through the development of the connected reservoir storage model, three profiles can be ascertained, and are interrelated, to shed valuable insight into the production potential, and health, of an unconventional gas reservoir. The three curves are the pressure drop curve, normalized curve and connected reservoir storage curve. Utilizing the three curves stated above embodies a more powerful reservoir performance analysis technique for predicting the production potential of complex reservoirs.
The connected reservoir storage model will be presented from the fundamental material balance approach and mathematical equations will be derived for this model as it relates to gas reservoirs. This concept will then be validated through the development of synthetic models with varying permeabilities, porosities and well geometries through the use of a reservoir simulation suite with known reservoir parameters along with the utilization of deconvolution techniques. Limitations involved with this method will be highlighted in the development procedures.
The connected reservoir storage model allows for the bridging of the production decline and reservoir pressure response from production data found in the public domain. This concept allows one to evaluate and forecast current and future volume production, reservoir pressure behavior, average reservoir pressure and determine original fluids in place with no prior knowledge of reservoir geometries, petro-physical data or fluid properties. With the development of this model decline curve analysis (DCA) can be extended into the transient regime of the production period.
