| 基于改进人工神经网络的气水层识别技术 |
| |
| 引用本文： | 田冷,何顺利.基于改进人工神经网络的气水层识别技术[J].测井技术,2009,33(5):449-452. |
| |
| 作者姓名： | 田冷 何顺利 |
| |
| 作者单位： | 1. 中国石油大学石油工程教育部重点实验室,北京,102249;中国石油大学气体能源开发技术教育部工程研究中心,北京,102249;中国石油大学石油天然气工程学院,北京,102249
2. 中国石油大学气体能源开发技术教育部工程研究中心,北京,102249;中国石油大学石油天然气工程学院,北京,102249 |
| |
| 摘 要： | 在改进的神经网络训练算法的基础上．提出了利用神经网络快速识别气、水层的方法。为了迅速、准确地判断储层性质，选用了Kohonen自组织网络和BP神经网络，利用测井参数，建立了长庆气田气、水层识别模型。仿真计算与测井综合解释相对比，样本符合率高达81．3%。分析表明，该方法所需参数少、适用范围广，能定量识别出气水层，从而为制定有水气井改造措施提供较可靠的依据。
|
| 关 键 词： | 测井解释 神经网络 气水层识别 测井参数 长庆气田 |
Gas and Water Layers Identification Based on Improved Neural Network Model |
| |
| TIAN Leng,,HE Shun-li.Gas and Water Layers Identification Based on Improved Neural Network Model[J].Well Logging Technology,2009,33(5):449-452. |
| |
| Authors: | TIAN Leng HE Shun-li |
| |
| Affiliation: | TIAN Leng1,2,3,HE Shun-li2,3(1.MOE Key Laboratory of Petroleum Engineering,China University of Petroleum,Beijing 102249,China,2.Engineering Research Center of Gas Energy Sources Development Technology,Ministry of Education,3.College of Petroleum Engineering,China) |
| |
| Abstract: | On the basis of an improved neural network training method,a way of using neural network to quickly identify and predict gas and water layers is developed.In order to identify reservoir property quickly and correctly,Kohomen self-organized network and Back-Propagation network are employed in building the gas and water layers identification model of Changqing gas field by using log data.Contrast of simulation calculation and integrated log interpretation shows that sample correspondent rate reaches 81.3%.The... |
| |
| Keywords: | log interpretation neural network gas and water layers identification log parameter Changqing gas field |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
