Study on Production Decline of Complex Fracture Network in Tight Oil Reservoirs Based on Embedded Discrete Fracture Model

Abstract

Abstract. The Blasingame decline analysis method plays a pivotal role in characterizing and interpreting the production behavior of fractured horizontal wells in tight reservoir systems. In practical tight reservoir systems, fluid transport frequently deviates from classical Darcy flow as a result of the presence of a threshold pressure gradient (TPG) and the stress-sensitive behavior of the rock matrix. To represent the complexity of flow in tight reservoirs, a mathematical model is developed based on the embedded discrete fracture model (EDFM). The proposed framework explicitly incorporates both the TPG and stress sensitivity, enabling a more realistic representation of flow behavior in tight formations. Typical curves of Blasingame production decline are obtained by the normalized pressure algorithm. The production decline curves can be classified into four flow stages. Furthermore, the impacts of key parameters—including stress sensitivity, threshold pressure gradient, primary fracture half-length, the number of secondary fractures within the stimulated reservoir volume (SRV), and the distribution of secondary fractures—on production decline behavior are investigated. With the increase of stress sensitivity coefficient and TPG, the pressure drop rate increases, and the Blasingame curve moves downward. The half-length of primary fractures has less effect on the curve. The number of secondary fractures increases, and all stages of the curve move upward. The Blasingame curve is at the top when secondary fractures are vertical to the primary fractures. When secondary fractures are parallel to the primary fractures, the curve is at the bottom. This study provides a clearer understanding of how nonlinear flow mechanisms and fracture network characteristics influence production decline behavior in tight oil reservoirs, which is useful for decline curve interpretation and fracture performance assessment.