多元回归方法在油田开发中的应用  

Application of Multiple Regression Method in Oilfield Development

作　　者：宫长路[1] 田晓东[1] 付百舟[1] 赵春颖[2] 王天智[1] 张继风[1] 

机构地区：[1]大庆油田有限责任公司勘探开发研究院,黑龙江大庆163712 [2]大庆油田有限责任公司第四采油厂,黑龙江大庆163711

出　　处：《大庆石油地质与开发》2006年第4期65-67,共3页Petroleum Geology & Oilfield Development in Daqing

摘　　要：多元回归在油田开发中得到了广泛的应用,常用的采收率经验公式就是一种典型应用。为了提高生产历史较短开发区块的预测精度,从3个方面研究介绍了该方法的应用情况:一是预测递减指数和初始递减率;二是预测年对年自然递减率;三是预测油田注采比。实际应用表明,建立的多元回归公式在短期预测时达到了较高的精度,具有较高的应用价值。Multiple regression method has been widely applied in oilfield development,and common recovery empirical formula is one typical method.In order to increase prediction accuracy of development blocks with short production history,this paper introduces multiple regression method through three following aspects: prediction of decline index and initial decline rate;prediction of annual decline rate;prediction of injection-production ratio.Real application indicates that multiple regression formula can have high accuracy during short-term predication with high application value.

关 键 词：多元回归 递减率 注采比 

分 类 号：TE329[石油与天然气工程—油气田开发工程]