利用动态资料研究巴喀油田裂缝分布

Integrating dynamics data to characterize naturally fractured reservoir—a case study of Baka oil field, Turpan Hami basin, China.

利用动态、静态资料，研究了吐哈盆地台北凹陷巴喀油田微裂缝的分布及其对开发过程的影响。对储集层的静态描述包括基质的孔渗特征和微裂缝的类型、方位、被充填的特征、发育程度以及控制裂缝发育的地质因素等。限于岩心的有限性（数量少、井眼尺度小）以及地质因素的复杂性，对裂缝的静态研究难以定量描述有渗流能力的裂缝分布，而动态资料则反映有效裂缝的宏观分布规律。因此用米采油指数（Ｊｏ）研究了有效裂缝的分布规律。对于采油井，米采油指数使用动态监测资料求得；而对于注水井则使用注水资料首先求得米注水指数（Ｊｗ），然后通过Ｊｗ／Ｊｏ比值（由实际井的资料求得）将Ｊｗ转换为等效的Ｊｏ。基于２１口采油井的米采油指数和１３口注水井的等效米采油指数，使用克里格技术作出巴喀油田的米采油指数分布图，该图基本上反映了裂缝发育的非均质性，指示了构造不同部位油井的生产能力和注水井的可注入能力，对生产、判断油井见水及水流方向等生产动态有一定的指导意义。图４参５（郭海莉摘）

Baka field is located in Turpan Hami basin, Xinjiang, China. Fan delta sandstone deposited in Mid Jurassic composes the reservoir that is low permeable and naturally fractured. Elongated high steep anticline with strike thrust faults and dip tear faults is oil pool structure. Water breakthrough has occurred at three wells after injection less than one year. Based on core and outcrop observation and fracture identification from logging, the static geological model of fracture distribution has been set up by studying fracture geometry and basic characteristics (orientation, aperture, extent, density, porosity and permeability). The fracture mineralization features and the factors that determine the occurrence of fractures also were studied, which include the degree of folding and/or faulting, in situ stresses, rock properties and bedding thickness, etc.. Because of small borehole scale and difficulties to map some factors which determine open, permeable fractures, the static model may not be representative of the large scale fracture network which will determine reservoir performance, thus dynamics production data must be integrated for getting a good understanding of fracture's influence. In this paper, the productivity index per meter reservoir ( J o) is used to characterize natural fractures. For production well, J o is calculated from dynamic surveillance data directly. But for every injection well, the injectivity index per meter reservoir ( J w) is estimated from injection data firstly, then J o is calculated from the J o/ J w ratio, this ratio is estimated from several actual wells which were converted from production to injection. In order to make the ratio J o/ J w convincing, further discussion which involves K rw ( S or )/ K ro ( S wi ) was made. Finally geostatistical method was used to distribute the productivity index. The dynamic index map matches the permeability of several wells estimated from well testing very well, and a good explanation about water breakthrough is derived, but some of differences from static model still exist.