节点文献
基于支持向量机的虚拟测井声波速度重构技术研究
The Reconstruction of the Virtual Well-Log Acoustic Velocity Technology Based on Support Vector Machine
【摘要】 对只有地震层速度的待钻井,为精确推算各种地质参数,可利用已钻井地震层速度和测井资料的映射关系,重构待钻井虚拟测井声波速度。在这个过程中采用支持向量机理论(SVM)解决了已钻井地震层速度和测井资料的非线性映射问题,重构了虚拟测井声波速度。采用该方法在准噶尔盆地永字号井进行了虚拟测井声波速度重构。模拟结果与BP网络方法、RBF网络方法相比,在绝对误差、相对误差、相关系数以及数据方差方面都是较优的。通过该方法可以获得高分辨率虚拟测井声波速度剖面,提高钻井模拟质量和钻井工程设计的水平。
【Abstract】 For the well to be drilled where only the seismic layer velocity is available, the mapping relation between the seismic layer velocity and the well-log information can be used to reconstruct the virtual well-log a-coustic velocity (AV) . The reconstruction of the virtual well-log AV is realized based on the Support Vector Machine theory which has solved the nonlinear mapping of seismic lay velocity to log information. This method is applied to reconstruct the virtual well-log AV of the well Yong in the Junggar Basin. In comparison with BP method and RBF method, the SVM method is proved to be the best in terms of absolute error、relative error、correlation coefficient and variance. High resolution virtual well-log AV profile can be obtained using the SVM method, the quality of the drilling simulation and the level of the drilling design can be improved.
【Key words】 support vector machine; seismic interval velocity; drilling simulation; log data; inversion; Junggar Basin;
- 【文献出处】 石油钻探技术 ,Petroleum Drilling Techniques , 编辑部邮箱 ,2006年06期
- 【分类号】P631.8
- 【被引频次】8
- 【下载频次】142