THE APPLICATION OF CLUSTER ANALYSIS BASED ON NEURAL NETWORK METHODS IN IDENTIFICATION RESERVOIR FLOW UNIT
SUN Zhi-xue1, YAO Jun1, SUN Zhi-lei2,3, LU Tao4, TANG Le-ping4, YANG Yong4, HAN Ji-chao1
1. School of Petroleum Engineering, China University of Petroleum, Qingdao 266555,China;
2. Key Laboratory of Marine Hydrocarbon Resources and Environmental Geology, Ministry of Land and Resources, Qingdao 266071,China;
3. Qingdao Institute of Marine Geology, Qingdao 266071,China;
4. Exploitation & Development Research Institute of Changqing Oilfield, Xi'an 710021,China
Identification and evaluation of the reservoir flow units is an important aspect of geological research of development mature oilfield. In this paper, take Sulige Gasfield He8 Group as an example, based on the sub-layer and detailed sediment study determined the characteristics of 10 variables to identify and rank reservoir flow units. Based on neural network algorithm of cluster analysis method to identify reservoir flow units of non-linear model. The established the neural network model can be more comprehensive consideration of various geological factors and the flow units. Using the established model, we can identify flow units for the well without boring, which can better solve the mature oilfield reservoir description abundance of information and accuracy requirements. Quarry applications indicates that in view of Chinese oil reservoir's main causes of continental origin, carry out reservoir flow units quantitative evaluation to matured oilfield have greater practical value.
孙致学, 姚军, 孙治雷, 卢涛, 唐乐平, 杨勇, 韩继超. 基于神经网络的聚类分析在储层流动单元划分中的应用[J]. 物探与化探, 2011, 35(3): 349-353.
SUN Zhi-xue, YAO Jun, SUN Zhi-lei, LU Tao, TANG Le-ping, YANG Yong, HAN Ji-chao. THE APPLICATION OF CLUSTER ANALYSIS BASED ON NEURAL NETWORK METHODS IN IDENTIFICATION RESERVOIR FLOW UNIT. Geophysical and Geochemical Exploration, 2011, 35(3): 349-353.
