摘要
在油气勘探领域中,当使用测井资料进行油气层分类识别时,运用传统的方法具有一定的局限性。本文使用了数据挖掘分类算法中的支持向量机(SVM)方法,并实际应用到新疆塔里木盆地油气层识别中。实验中分别采用了支持向量机算法和BP神经网络算法进行对比检验,结果表明通过支持向量机算法建立的油气层识别模型具有更高的识别检验性能,体现了支持向量机在处理多类分类问题中的优越性。
In the area of oil and gas exploration, when identifying them by wire log data, there are some flaws in traditional method. This paper uses the Support Vector Machine （SVM） method which is based on the classification algorithm of Data Mining, and applies it to the oil and gas formation identification of Sinkiang Tarim Basin. In this paper, the experiment both uses the SVM and BP Neural Network algorithms, and the result shows that the oil and gas formation identification model made of SVM algorithm are the most useful identification capability. So the SVM algorithm is superior in application to deal with the multi-classification problem.
出处
《微计算机信息》
2010年第5期44-46,共3页
Control & Automation
