Numerical and ANFIS modeling of the effect of fracture parameters on the performance of VAPEX process

The Vapor Assisted Petroleum Extraction (VAPEX) process, a newly developed enhanced heavy oil recovery process, is a promising EOR method for certain conventional non-fractured heavy oil sandstone reservoirs such as reservoirs located in Canada, but its applicability on low permeable naturally fractured reservoirs like Persian Gulf reservoirs is controversial. Previous studies show that the foremost concern for VAPEX application in naturally fractured reservoirs is the low non-economical production rates. The aim of the present work is studying the effect of fracture geometrical properties such as orientation, length, intensity, discontinuity, and position of fractures on the performance of the VAPEX process in a fractured model. Then an intelligent model is proposed for predicting the recovery of the process. To do so, first, a non-fractured, two-dimensional numerical model, i.e. a conventional model is constructed to simulate the VAPEX process. Then, different fracture patterns are applied in the model and the simulation results are analyzed. The qualitative analysis of the simulation results demonstrates that the presence of the fractures enhances the recovery of the VAPEX process. In order to quantify and model the effects of these geometric fracture parameters on the performance of VAPEX process, an intelligent modeling tool, Adaptive Neural Fuzzy Inference System (ANFIS) is employed. For the model results to be generalized, dimensionless forms of geometrical parameters are utilized. In fact, this model employs an image of the fracture pattern which determines the geometrical parameters of the fractures. Considering a constant Pressure-Volume-Temperature (PVT), relevant rock properties and operational condition, the recovery of the pattern can be determined fairly well. The results of this study show that the presence of both vertical and horizontal fractures improve VAPEX recovery, but the effect of vertical fracture are more than horizontal fractures on the VAPEX recovery. Fracture length is an important parameter on the performance of VAPEX as higher production rates achieved in the case of larger length. In addition, discontinuity in the fractures decreases the oil production rate in the VAPEX period. Finally, the effect of position of the fractures on the recovery are very complicated. As the distance of the fractures from the hypothetical line increases, the oil production rate in the VAPEX period increases. Weight in on ANFIS as an intelligent tool by combining fuzzy logic and neural networks can predict an output according to the input parameters. This ability helps quantify the complicated effect of different parameters on the recovery of the VAPEX process.