基于神经网络的元素录井岩性识别方法研究及应用
|
摘要
随着PDC钻头的推广及应用,钻井速度加快、录井岩屑细碎,给现场地层岩性识别带来很大的挑战,其中以潜山地层尤为突出。以渤中22-23构造带为例,在元素录井数据标准化的基础上,结合薄片、岩心等资料,确定研究区4类主要岩性的敏感元素。以岩性敏感性元素数据为基础,利用BP神经网络方法构建地层岩性识别模型。通过研究区大量薄片鉴定岩样数据的训练与验证,岩性识别的准确度达到90%以上,表明该方法对提高现场岩性识别的准确率有着较好的效果。
The popularization and application of PDC bits leads to high drilling speed and fine debris in mud logging, which has brought great challenges to lithology identification in situ, especially in buried hill formations. Taking Bozhong 22-23 structural belt as an example, on the basis of standardization of element logging data, combined with thin section and core data, the sensitive elements to four types of main lithology in the study area are determined. Based on the data of lithology sensitive elements, a lithology recognition model of strata is constructed by using BP neural network. Through the training and verification of the model using a large number of thin section identification rock sample data in the study area, the lithology identification accuracy of the model is more than 90%, which shows that this method has a good effect on improving the accuracy of lithology identification in the field.
The popularization and application of PDC bits leads to high drilling speed and fine debris in mud logging, which has brought great challenges to lithology identification in situ, especially in buried hill formations. Taking Bozhong 22-23 structural belt as an example, on the basis of standardization of element logging data, combined with thin section and core data, the sensitive elements to four types of main lithology in the study area are determined. Based on the data of lithology sensitive elements, a lithology recognition model of strata is constructed by using BP neural network. Through the training and verification of the model using a large number of thin section identification rock sample data in the study area, the lithology identification accuracy of the model is more than 90%, which shows that this method has a good effect on improving the accuracy of lithology identification in the field.
引文
[1]谭伟雄,张志虎,马猛,等.X射线元素录井与测井结合识别碳酸盐岩岩性[J].录井工程,2017,28(4):12-15.
[2]孙哲,韦阿娟,江尚昆,等.元素录井技术在渤海潜山岩性识别中的应用[J].特种油气藏,2017,24(5):78-84.
[3]尚锁贵,谭忠健,阚留杰,等.元素录井岩性识别技术及其在渤海油田的应用[J].中国海上油气,2016,28(4):30-34.
[4]陈颖,徐高军.塔河油田风化壳卡取难点及方法应用[J].石油工业技术监督,2017,33(5):10-15.
[5]王振华,张亮.测井曲线标准化方法优选及其效果评价:以延长油田化子坪区块为例[J].西安石油大学学报(自然科学版),2018,33(2):37-43.
[6]王守君,刘振江,谭忠健,等.勘探监督手册(地质分册)[M].北京:石油工业出版社,2013.
[7]单敬福,陈欣欣,赵忠军,等.利用BP神经网络法对致密砂岩气藏储集层复杂岩性的识别[J].地球物理学进展,2015,30(3):1257-1263.
[2]孙哲,韦阿娟,江尚昆,等.元素录井技术在渤海潜山岩性识别中的应用[J].特种油气藏,2017,24(5):78-84.
[3]尚锁贵,谭忠健,阚留杰,等.元素录井岩性识别技术及其在渤海油田的应用[J].中国海上油气,2016,28(4):30-34.
[4]陈颖,徐高军.塔河油田风化壳卡取难点及方法应用[J].石油工业技术监督,2017,33(5):10-15.
[5]王振华,张亮.测井曲线标准化方法优选及其效果评价:以延长油田化子坪区块为例[J].西安石油大学学报(自然科学版),2018,33(2):37-43.
[6]王守君,刘振江,谭忠健,等.勘探监督手册(地质分册)[M].北京:石油工业出版社,2013.
[7]单敬福,陈欣欣,赵忠军,等.利用BP神经网络法对致密砂岩气藏储集层复杂岩性的识别[J].地球物理学进展,2015,30(3):1257-1263.