摘要
运用有机地球化学方法分析岩芯、岩屑样品的有机碳含量存在着昂贵、费时且不准确等问题 .利用测井方法的优点是经济、准确 .在测井评价中使用人工神经网络法具有极大的优越性和适用性 .本文结合Kohonen和BP网络方法 ,完成了塔里木台盆区 19口井的寒武、奥陶系烃源岩层段的识别与评价 .并通过测井资料处理成果和岩芯有机地化资料、地质录井情况的相互检验、证实 ,基本上能够满足评价烃源岩的需要 ,从而为利用测井资料进行烃源岩评价做出了新的尝试 .
It was very troublesome to evaluate source rock with the traditional method of geochemistry. Log-identifying has the advantages of economy and accuracy. Compared with the traditional methods of logging, the artificial neural network has many advantages and adaptabilities. Based on the analysis of prior methods, combining Kohonen network and BP network successfully, I practically apply it to 19 wells in Cambrian-Ordovician of Tarim. It is indicated that this method is useful?efficient and reliable hence open up a new area of source rock evaluation.
出处
《地球物理学进展》
CSCD
2002年第1期137-140,共4页
Progress in Geophysics
基金
国家九五重点攻关项目"塔里木盆地有效烃源岩展布及其特征"(991110 30 1)资助
