Abstract-Many enhanced oil recovery projects like steam injection into an oil reservoir are analyzed using composite reservoir models. Most of the models used assume two-region composite reservoirs with highly different properties separated by a vertical front. Applicability of thermal well test analysis method was evaluated previously using simulation studies of both vertical and horizontal steam injection wells with the conclusion that simplifying assumptions of the conventional two-region models may not explain some pressure behavior and may cause significant errors in the estimates. The main objective of this paper is therefore to develop a new analytical model for well test analysis to improve previous models. The model is a three-region composite model with an intermediate region characterized by power-law decline of properties. Fronts are tilted due to the gravity override and are modeled using the multi-layer reservoir concept assigning different front radii to each layer. A commingled model is assumed in which there is no cross-flow between the layers and all the communication happens through the wellbore. Steam condensation is included in the form of heat loss to the formation. Pressure type curves are generated and validated. Effects of the important parameters included in the model are investigated. The model developed in this work will be used in type curve matching for improved well test analysis.

Abstract-A composite reservoir may occur during any enhanced oil recovery project like steam injection into an oil reservoir. Falloff test analysis in steam injection projects offers a quick way to obtain estimates of the swept volume and steam zone properties. Most of the models used for the analysis assume two-region composite reservoirs with different but uniform properties separated by a sharp vertical interface, which is not very realistic. Strange trends seen on the pressure plots and the errors associated with the volume estimates could be related to the simplifying assumptions of the conventional models. The main objective of this paper is to develop a new analytical model for pressure transient analysis to improve previous results with inclusion of more realistic assumptions. The proposed model is a three-region composite model with an intermediate region in which mobility and storativity decrease as power law functions of the radial distance from the front. The fronts are considered to be tilted due to gravity effects. Steam condensation in the form of heat loss from the steam zone is also included in the analytical model. The new sets of type curves for well test interpretation are generated and verified. Effects of several parameters on the shape of type curves are discussed. The developed model can improve estimations of reservoir parameters using type curve matching and explain the anomalies seen on the data, which cannot be described using the conventional models. The general nature of the model makes it applicable to other types of composite reservoirs created either naturally or artificially.

No abstract

Abstract-Pressure data of composite reservoirs are usually analyzed using two-region models separated by a vertical front. This simplification may lead to significant errors in the estimates. In this paper, a multi-region composite model with thin skin at the fronts is improved by adding n layers in the vertical direction to model the effect of gravity. The proposed model is then compared to the recently developed models for thermal well test analysis. The proposed model consists of m regions to model the smooth variation of properties from the inner to the outer region while the composite models used for the comparison assume an intermediate region characterized by power-law decline of properties. Fronts are assumed tilted due to the gravity effects by use of multi-layer reservoir concept with no cross-flow between the layers. The model can be validated against conventional models. Addition of more intermediate regions and skin at front locations are investigated to achieve a better match with the new analytical models. The model developed in this work can be used in type curve matching for improved well test analysis.

No abstract