Improved identification of fracture total factor characteristics and cross-scale flow channels by interpreting initial production data of horizontal wells

Fracture identification after hydraulic fracturing of tight gas reservoirs is crucial to fracturing evaluation, productivity analysis, and production plan. Owing to fracture branching, bending, extending, and reversing in reservoirs, a simplified straight fracture cannot fully describe fracture characteristics. The novelty of this study is that the coupling mechanism between the complex fracture network characteristics and well production performances is figured out, thus achieving the identification of fracture morphology in fractured reservoirs. In this study, a combined model including Laplace transform, boundary element method, Stehfest numerical inversion, and fractal theory presents transient well performances. The key contributions to productivities at different production stages are dynamically identified via a sensitivity analysis. The results clearly indicate that the fracture total length is the key factor determining productivities due to no fracture interference at initial production stage. Furthermore, the logarithm of productivity linearly depends on the logarithm of production time and the productivity contribution rate equals the ratio of fracture length to total length. Thus, an updated inversion method is proposed by interpreting initial production data based on a tree-shaped fracture network associated with fracture total factor and cross-scale characteristics. The inversion results well match with their targets, with an error of no more than 5 %, ensuring the accuracy and validity of the proposed method. This innovative approach with low requirements on field data is convenient to yield an accurate prediction of fracture morphology, thereby providing a feasible strategy for promoting sustainable production of clean energy in unconventional gas reservoirs.