文件名称:
Machine Learning and Model Based Diagnosis using Possible Conflicts
开发工具:
文件大小: 209kb
下载次数: 0
上传时间: 2011-12-16
详细说明： This work presents an on-line diagnosis algorithm for dynamic systems that combines model based diagnosis and machine learning techniques. The Possible Conflicts method is used to perform consistency based diagnosis. Possible conflicts are in charge of fault detection and isolation. Machine learning methods are use to induce time series classifiers, that are applied on line for fault identification. The main contribution of this work is that Possible Conflicts are used to decompose the physical system. This decomposition allows defining the input-output structure of an ensemble of classifiers. Hence the structural knowledge provided by the Possible Conflicts is exploited by the machine learning methods. Possible Conflict decomposition may be used for class selection or for class and attribute selection. Experimental results on a simulated pilot plant show that class selection has an important potential to increase the classifier accuracy for several learning algorithms. The effect of an additional attribute selection depends on the kind of machine learning method, although it improves the accuracy of the most precise classifiers, especially at the first stages of the diagnosis processes, just after a fault has been detected. ...展开收缩
(系统自动生成,下载前可以参看下载内容)
下载文件列表
相关说明
- 本站资源为会员上传分享交流与学习,如有侵犯您的权益,请联系我们删除.
- 本站是交换下载平台，提供交流渠道，下载内容来自于网络，除下载问题外，其它问题请自行百度。
- 本站已设置防盗链，请勿用迅雷、QQ旋风等多线程下载软件下载资源，下载后用WinRAR最新版进行解压.
- 如果您发现内容无法下载，请稍后再次尝试；或者到消费记录里找到下载记录反馈给我们.
- 下载后发现下载的内容跟说明不相乎，请到消费记录里找到下载记录反馈给我们，经确认后退回积分.
- 如下载前有疑问，可以通过点击"提供者"的名字，查看对方的联系方式，联系对方咨询.
