摘要
针对复杂地球物理数据集合,应用判别分析技术构建不同集合体识别函数时,需要优选出能敏感反映各类集合体特征的测量参量.声波、电阻率、自然伽马等地球物理测井项目对于岩层中的油、水性质反映能力各不相同,以卫星油田葡萄花油藏的20个油层、10个油水同层、15个水层为研究对象,依据构建集合内参量均匀度参数及集合间距离参数,优选出识别3类岩层的敏感测量项目,并依据这些参量建立贝叶斯判别函数,相对于所有测量参量均参与计算的情况,方法识别精度提高了9.7%.
For the complicated geophysics data set, we need to select measured paramters reflecting all kinds of aggregate characteristics impressible when we use discriminant analysis technique to build different aggregate recognition function. Geophysics well logging items. As AC, LLD, GR, ect, have different reflection abilities on the oil or water formation prop- erties.This paper uses 20 oil layers,. 10 oil and water layers, 15 water layers as the object of study on the Weixing oil field Putaohua oil reservoir.Depend on building set parame- ter uniformity parameters and the diatance parameters between collection, we choose the sensitive measurement item which can discriminate three types of rock.And depending on these parameters build the Bias discriminant function. For the situation of all t paraments involved in the calculation, this method make identification accuracy improve 9.7%.
出处
《数学的实践与认识》
CSCD
北大核心
2014年第1期188-192,共5页
Mathematics in Practice and Theory
基金
国家自然科学基金(41274132)
关键词
参量优选
油水性质识别
均匀度参数
距离参数
常规测井
葡萄花油藏
parameter optimization
identification of oil,water properties
uniformity pa-rameters
conventional logging
Putaohua reservoir
