低渗特低渗储层水锁损害定量预测方法  

The Quantitative Prediction Method of Water Blocking Damage in Low and Extra-Low Permeability Reservoir

作　　者：蒋官澄[1] 王晓军 关键 李建成 杨鹏 郭林昊 

机构地区：[1]石油工程教育部重点实验室(中国石油大学(北京)),北京102249 [2]中国石油长城钻探工程有限公司工程技术研究院,辽宁盘锦124010

出　　处：《石油钻探技术》2012年第1期69-73,共5页Petroleum Drilling Techniques

基　　金：国家科技重大专项"复杂结构井优化设计与控制关键"(编号:2011ZX05009-005-03A);国家杰出青年科学基金项目"洗井;固井;油层等损害与保护"(编号:50925414)联合资助

摘　　要：在低渗特低渗油气藏开发过程中储层损害严重,其中水锁损害是主要损害之一,因此建立准确、快速定量预测水锁损害程度的数学模型,对高效开发低渗透油田具有重要意义。采用线性回归分析法建立的水锁损害预测模型不能准确反映储层损害实质,预测精度较低。将常规线性回归模型转变为非线性回归模型,能够更好地反映各损害因素与损害程度间的关系;采用逐步回归法选择回归因子,建立了新的水锁损害定量预测模型,并对新模型进行了检验。结果表明,新建立的水锁损害定量预测模型预测结果准确性高,15组验证样本的预测准确率均大于80%,平均预测准确率为91.69%,可为油气层保护和储层解堵措施优化设计提供科学依据。There are very serious damages during the exploitation of low permeability and extralow permeability reservoirs, and water blocking is the most remarkable damage. So the mathematical model which can accurately and rapidly predict water blocking damage level will play an important role in the ef fective development of low permeability oilfield. Linear regression analysis was used to establish prediction model of water blocking damage in the past, but the model could not accurately reflect the real damage essence to the reservoir, especially the prediction accuracy was low. As we all know that the nonlinear regression model can better reflect the relationship between damage factors and extent of damage than the conventional linear regression model. The stepwise regression method was adopted to select regression factors and the new water blocking damage prediction model was established. The test results for the new model indicated that, the new model had higher accu racy than the conventional one, for example, the prediction accuracy rate of 15 tested samples were greater than 80%,and the average prediction accuracy rate was 91.69%. It will provide evidence for reservoir protection and measures to optimize the design of reservoir unplugging.

关 键 词：低渗透储集层水锁损害定量预测非线性回归逐步回归法 

分 类 号：TE312[石油与天然气工程—油气田开发工程]