Mechanistic Model Validation of Decline Curve Analysis for Unconventional Reservoirs

The primary objective of this work is to provide the validation and ranking of numerous decline curve analysis models. From the validation cases analyzed in this research best practice guidelines for each of the decline curve analysis models are presented. In addition to the decline curve specific guidelines, general best practices are outlined to be implemented when performing decline curve analysis. A built-for-purpose mechanistic model was developed and used successfully as a validation tool for the decline curve analysis models. The primary validation cases are analyzed using typical properties and parameters from the Haynesville, Eagle Ford, and the Wolfcamp formations. The effect of non-constant pressure production on time-rate data is also analyzed. Investigation into the effect of available time-rate data on the estimation of ultimate recovery at the end of a wells producing life is presented for each of the investigated decline curve analysis models. A methodology is proposed in order to assist with poor time-rate data quality from unconventional reservoirs. The respective necessary equations for the power-law exponential, stretched exponential, and modified hyperbolic decline curve models are derived and presented, as well as the associated modified "qDb" plot (referred to as the rate-integral "modified qDb" plot). It was concluded from this research that the power-law exponential decline curve model is the most valid for the representation of time-rate data from unconventional reservoirs