Production optimization for water flooding in fractured-vuggy carbonate reservoir – From laboratory physical model to reservoir operation

2019 
Abstract Conventional physical model experiments of carbonate reservoirs are limited to the investigation of production patterns and mechanisms for a single well or a well group, and cannot effectively optimize the oil production at the reservoir scale. For an actual fractured-vuggy reservoir in Tahe Oilfield, China, seismic and drilling data were first analyzed to obtain the distributions of fractures and vugs, and the dynamic tracer monitoring was used to determine the type of connection between fracture-vug units. Then, according to shape similarity, the real fracture-vug units are simplified into 12 visualized physical models made of transparent acrylic plates with a scale of 300:1. Different physical models were further combined together with the connectivity modes derived from dynamic tracer monitoring, thus embodying the entire complex carbonate reservoir to a laboratory-scale physical model for the first time. Experiments were then carried out with these models and results showed that changing the injection and production parameters and increasing the number of flow channels between injector and producer have little effect on the displacement efficiency, while altering the water flow direction and converting injector into producer achieved better effect, due to the increasing fracture-vug units that can be displaced. In the field production, the output of TK652, TK605CH and TK608 wells increased from 11, 5, 1 m3/d to 18, 18, 19 m3/d, respectively, by closing well TK659 and converting well TK6100 (two high water-cut producers) to change the direction of water flow. The laboratory physical modeling technique illustrated in this study provides a simplified and unified way for the production optimization of fractured-vuggy carbonate reservoirs.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    26
    References
    7
    Citations
    NaN
    KQI
    []