Early in the producing life of a gas or a gas-condensate reservoir, a material balance analysis is carried out to assess reserve (initial gas-in-place and/or recoverable reserve) and to infer reservoir life. The data required for a material balance analysis are average reservoir pressure, gas deviation factor, two-phase deviation factor, and cumulative production. These data may be erroneous, to a certain extent, because of several reasons. The errors in the input data may be random and/or systematic. A material balance analysis based on a straight time may also be affected by the extend of reservoir depletion, number of data points, and data collection frequency.
This work analyzes the effects of errors in the input data on reserve estimation via a graphical material balance analysis based on a straight line to study the value of additional and/or accurate data acquisition to reduce the errors in reserve estimation for volumetric, gas or gascondensate reservoirs. Random errors are modelled as normally-distributed errors. Systematic errors are modelled as linearly-varying errors. A Monte-Carlo simulation study has been carried out using a dimensionless material balance equation to obtain results applicable to all volumetric, gas or gascondensate reservoirs.
The results of this study suggest that both data quantity and quality are important in estimating the initial gas-inplace (G) accurately. A large quantity of data may offset problems in least-squares fitting due to data inaccuracy to some extent. But data accuracy is vital to estimate G accurately during the early production life of a reservoir.
The results of this study would be helpful in planning for production and well-test data acquisition from gas or gascondensate reservoirs, and in suggesting error bounds for estimated G.
For a volumetric, dry gas reservoir or gas-condensate reservoir with average reservoir pressure above dewpoint pressure, a P/zg vs. cumulative production plot is employed for a material balance analysis. (1) For a gascondensate reservoir with average reservoir pressure significantly below the dew-point pressure, a graph of p/Z2 Ø vs. cumulative production is prepared for a material balance analysis.(1–3). If a significant amount of condensate has been produced, then the cumulative condensate produced should be converted to its gas equivalent and added to the cumulative gas produced to estimate the cumulative production.(l)
The errors in the input data may be random and/or systematic. A material balance analysis based on a straight line may also be affected by the extent of reservoir depletion, number of data points, and data collection frequency. The problems in filling a straight line to a data set when both x- and y-variables are subject to error and the misuses of linear regression have been addressed in other disciplines, including geology and hydrology.(4–12) Significant error in reserve estimation may occur for volumetric, gas or gas-condensate reservoirs because of a considerable extrapolation of erroneous x- and y-axes data for a P/zg (or p/Z2 Ø) vs. cumulative production plot. To the best of our knowledge, the preceding problem in reserve estimation for volumetric, gas or gas-condensate reservoirs has not been discussed in the petroleum engineering literature.