摘要
在岩性识别方法中,通常采用能够代表岩样的测井参数值,作为确定岩性与测井参数对应关系的基础数据,然后通过数学手段建立岩性识别模式。尝试用多道X荧光仪地球化学测井所测得的岩层元素含量来分类岩性,并利用人工神经网络来解决测井岩性识别问题,在说明BP和SOM网络的模型和算法的基础上,结合某地的实际测井资料,建立神经网络岩性识别模型,进行岩性识别的比较应用研究。结果表明,通过岩层的化学含量来识别岩性具有较高的准确率,较广的识别范围。证明运用尝试用多道X荧光仪测井可以用于解决测井岩性识别问题,具有很好的应用前景。
Lithologic identification methods were usually used to represent rock types of logging parameters,as determined lithology and logging parameters of the relationship between the basic data,and then by means of mathematical models to identify the establishment of lithology.The spectrometer can be used in the text of geochemical logging data by the elements in the rock to classify lithology and ANN was used to solve the problem of identifying lithological logging.First,it describes the model and algorithm of the BP and SOM network.Then an example is used to show how to build up a network model for logging lithological identification and its application in logging lithological identification.The results indicate that the chemical content through the rock to identify lithologic has a high accuracy rate,the broader scope of the identification;and prove that EDS can be used to solve the problems of lithology identification of logging,which has good prospects.
出处
《石油天然气学报》
CAS
CSCD
北大核心
2008年第05X期218-221,384,共4页
Journal of Oil and Gas Technology
关键词
地球化学测井
多道X荧光仪
岩性识别
元素含量
人工神经网络
geochemical logging
multi-channel X-ray fluorescence instrument
lithology identification
element content
artificial neural network
