| 基于CBR的数控设备故障诊断系统知识表示 |
| |
| 引用本文： | 陈友东,韩美华,叶进军.基于CBR的数控设备故障诊断系统知识表示[J].北京航空航天大学学报,2011,37(12):1557-1561. |
| |
| 作者姓名： | 陈友东 韩美华 叶进军 |
| |
| 作者单位： | 1.北京航空航天大学 机械工程及自动化学院, 北京 100191 |
| |
| 基金项目： | 国家科技重大专项资助项目(2009ZX04009-014-02,2009ZX04013) |
| |
| 摘 要： | 随着数控设备越来越复杂,对设备故障诊断与维修提出了更高的要求,采用专家系统的方式建立故障诊断系统以实现数控设备的快速诊断和维修.针对CBR (Case-Based Reasoning)系统自身的增量学习模式使系统的实例库逐渐增大,系统效能大大降低的缺点.根据数控系统结构的特点采用分层检索的方法,依据该检索方法采用框架型的知识表示方式,提高了大实例库的检索效率.将该方法应用于数控设备的故障诊断系统中,提高了故障诊断的诊断效率.
|
| 关 键 词： | 数控系统 故障诊断 实例推理 知识表示 |
| 收稿时间： | 2010-07-23 |
Knowledge representation for CNC equipment fault diagnosis system based on CBR |
| |
| Chen YoudongSchool of Mechanical Engineering and Automation,Beijing University of Aeronautics and Astronautics,Beijing,China Han MeihuaSchool of Reliability and Systems Engineering,Beijing University of Aeronautics and Astronautics,Beijing,China Ye Jinjun.Knowledge representation for CNC equipment fault diagnosis system based on CBR[J].Journal of Beijing University of Aeronautics and Astronautics,2011,37(12):1557-1561. |
| |
| Authors: | Chen YoudongSchool of Mechanical Engineering and Automation Beijing University of Aeronautics and Astronautics Beijing China Han MeihuaSchool of Reliability and Systems Engineering Beijing University of Aeronautics and Astronautics Beijing China Ye Jinjun |
| |
| Institution: | 1.School of Mechanical Engineering and Automation, Beijing University of Aeronautics and Astronautics, Beijing 100191, China2. School of Reliability and Systems Engineering, Beijing University of Aeronautics and Astronautics, Beijing 100191, China3. School of Computer Science and Technology,Wuhan University of Science and Technology, Wuhan 430065, China |
| |
| Abstract: | As the computer number control (CNC) become more and more complexity, there is a great demand for fault diagnosis and maintenance system to improve the efficiency of failure diagnostics and maintenance. The case-based reasoning (CBR) has been broadly applied in fault diagnosis. However the case base is increasingly large with the incremental learning which results in low solution retrieval efficiency and weak performance. To improve the accuracy of retrieval cases and reduce retrieval time, a method that searches the cases with stratified indexing and feature retrieval was presented in terms of the structure of CNC, and a frame of knowledge representation was detailed by this method. A prototype system was built with the search method and knowledge representation. The experimental results show that this method is effective. |
| |
| Keywords: | computer number control fault diagnosis case-based reasoning knowledge representation |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《北京航空航天大学学报》浏览原始摘要信息 |
| 点击此处可从《北京航空航天大学学报》下载免费的PDF全文 |
|
