基于支持向量机的地震属性优选及煤层气预测

Seismic attribute optimization based on support vector machine and coalbed methane prediction

摘要: 为了充分发挥地震属性分析技术的优势,针对地震勘探中的小样本事件,阐述了支持向量机原理,开展了基于支持向量机的煤田地震属性非线性优选的方法研究,并在煤层气含量预测中取得了良好的效果。结果表明:基于支持向量机(SVM)属性优选的煤层气预测效果比运用钻孔插值的效果更精确,较好地解决了小样本的学习问题,可作为煤层气预测的一种有效方法。

Abstract: In order to take full advantages of seismic attribute analysis technology, this paper explained the principle of support vector machine (SVM) based on event of small samples; SVM method for nonlinear optimization of seismic attributes in coal field was applied in prediction of coalbed methane (CBM) content and produced good effect. The study results indicate that the prediction of CMB based on SVM attribute optimization is more precise than using the drilling interpolations method, can better solve the learning problems of small samples and can be used as an effective method for the prediction of CMB.