摘要
青海乌南油田上油砂山组(N22)和下油砂山组(N12)储集层平均孔隙度分别为13.0%和13.6%,平均渗透分别为3.88×103μm2和2.93×103μm2,属于低孔、低渗油田。在低孔、低渗储层,由于油气储层中测井资料受储层岩性、地层水性质和储层物性等影响较大,造成含油气储层测井曲线异常特征不明显,单一应用测井资料识别油、气、水层困难。应用BP神经网络技术对乌南油田低孔、低渗储层的测井资料与录井资料进行综合处理,利用测井信息的丰富性和高分辨率的优势与录井资料识别油、气、水层的直观准确性互相结合对低孔、低渗储层进行油气识别。
Reservoirs in the Upper Youshashan Formation ( N2^2 ) and the Lower Youshashan Formation ( N2^1 ) of Wunan oilfield in Qinghai Province are low in porosity (with average porosity at 13.0% and 13.6% respectively) and permeability (with average permeability at 3. 88 × 10^3 μm^2 and 2. 93 10^3 μm^2). In this type of reservoirs, logging data is greatly affected by the reservoir lithoglogy, formation water characteristics, and reservoir physical properties, causing difficulties in distinguishing the curves of oil/gas-bearing layers and differentiating oil and gas layers from water layers according to logging data only. In this paper, BP neural network technology is applied to the integrated processing of wire logging data and surface logging data from the low-porosity and low-permeability reservoirs in Wunan oiffield. The abundance and thigh resolution of wire log information are combined with the directness and accuracy of surface log data to identify hydrocarbon zones in reservoirs with low porosity and low permeability.
出处
《石油与天然气地质》
EI
CAS
CSCD
北大核心
2007年第3期407-412,共6页
Oil & Gas Geology
关键词
测井
气测录井
地化录井
BP神经网络
低孔
低渗储层
油气识别
wire log
gas log
geochemical log
BP neural network
low-porosity and low-permeability reservoir
hydrocarbon identification