极限学习机在岩性识别中的应用  

Lithologic identification based on ELM

作　　者：蔡磊[1] 程国建[1] 潘华贤[1] 

机构地区：[1]西安石油大学计算机学院,陕西西安710065

出　　处：《计算机工程与设计》2010年第9期2010-2012,共3页Computer Engineering and Design

基　　金：国家自然科学基金项目(40872087)

摘　　要：基于传统支持向量机(SVM)训练速度慢、参数选择难等问题,提出了基于极限学习机(ELM)的岩性识别。该算法是一种新的单隐层前馈神经网络(SLFNs)学习算法,不但可以简化参数选择过程,而且可以提高网络的训练速度。在确定了最优参数的基础上,建立了ELM的岩性分类模型,并且将ELM的分类结果与SVM进行对比。实验结果表明,ELM以较少的神经元个数获得与SVM相当的分类正确率,并且ELM参数选择比SVM简便,有效降低了训练速度,表明了ELM应用于岩性识别的可行性和算法的有效性。Based on the problem that training speed is slow and parameter selection is difficult in traditional support vector machine （SVM）, a method based on extreme learning machine （ELM） for lithology identification is presented. ELM is a new learning algorithm of single-hidden layer feedforward neural networks （SLFNs）. It can not only simplify the parameter selection process, but also improve the training speed of the networks. In determining the optimal parameters, the lithology classification model is established, and the classification result of ELM is compared to traditional SVM. The experimental results show that, ELM with less number of neurons has similar classification accuracy compared to SVM, and it is easier to select the parameters which significantly reduce the training speed. The feasibility of ELM for lithology identification and the availability of the algorithm are validated.

关 键 词：机器学习 极限学习机 前馈神经网络 岩性识别 支持向量机 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]