Natural Gas Industry B (Oct 2020)
A new well test interpretation model for complex fracture networks in horizontal wells with multi-stage volume fracturing in tight gas reservoirs
Abstract
Multi-stage volume fracturing of horizontal wells is the main means to develop tight gas reservoirs. Complex fracture networks of various shapes are generated around the wellbore after volume fracturing. At present, however, most of the well test models suitable for fracturing horizontal wells take all hydraulic fractures as single main fractures, which results in a large error between well test interpretation result and actual situation. As a result, the fracture network characteristic parameters of the stimulated areas cannot be obtained accurately. To this end, a well test model for complex fracture networks in tight-gas fracturing horizontal wells was established on the basis of the non-structural discrete fracture model. Then, this model was solved by using the finite element method with combined triangular elements and linear elements. And accordingly, the well test type curves of a horizontal well under different fracture network patterns (rectangular, elliptical and hyperbolic) were prepared. Based on this, well test type curves were analyzed from the aspects of characteristics and influential factors and were compared with those obtained from the conventional single-fracture model. Finally, the new model was applied in well test interpretation of one multi-stage volume fracturing horizontal well in the gas reservoir of Permian Shan 1 formation in the Qingyang Gas Field of the Ordos Basin. And the following research results were obtained. First, the biggest difference of well test type curve between the fracture network model and the conventional single-fracture model occurs in the early stage, the characteristics of first linear flow regime are replaced with the characteristics of pseudo-radial flow regime in the stimulated area. Second, the end time of the pseudo-radial flow regime in the stimulated area is mainly dominated by the size and shape of the stimulated area. The larger the stimulated area is, the longer the pseudo-radial flow regime lasts. As the shape of the stimulated area approaches to be elongated, the characteristics of the well test type curve obtained by the new model are more consistent with those by the single-fracture model. Third, the pressure derivative value of the pseudo-radial flow regime in the stimulated area is mainly dependent on the conductivity and density of the fracture network. The higher the density or the conductivity of fracture network in the stimulated area is, the earlier the wellbore storage effect regime ends, the lower the pressure derivative value of the pseudo-radial flow regime in the stimulated area is and the more obvious the characteristics of the horizontal line are. In conclusion, case study results confirm that the new model is reliable and practical and can provide accurate reservoir parameters as well as the size of the effectively stimulated area by volume fracturing and the conductivity of fracture network, which is conducive to evaluating the stimulation effect of volume fracturing and predicting the postfrac production performance.
