| 基于FSVM的抽油泵示功图识别方法 |
| |
| 引用本文： | 时文刚,戴静君,王永.基于FSVM的抽油泵示功图识别方法[J].石油矿场机械,2004,33(4):46-48. |
| |
| 作者姓名： | 时文刚 戴静君 王永 |
| |
| 作者单位： | 1. 北京石油化工学院,机械工程学院,北京,102617
2. 大庆石油管理局,黑龙江,大庆,163255 |
| |
| 摘 要： | 针对常规支持向量机多值分类器中存在的不可分类区域问题，引入模糊数学中的模糊隶属度函数，提出了一种改进的多值SVM分类器。即模糊支持向量机(FSVM)的模型，并将其用于抽油泵多种典型故障示功图的模式分类。实验表明，与常规SVM分类器相比，改进的SVM分类器具有更好的模式分类能力和推广能力。
|
| 关 键 词： | 支持向量机 模式识别 抽油泵 示功图 |
| 文章编号： | 1001-3482(2004)04-0046-03 |
| 修稿时间： | 2004年5月10日 |
FSVM based recognition approach for downhole pump dynamometer cards |
| |
| SHI Wen-gang,DAI Jing-jun,WANG Yong.FSVM based recognition approach for downhole pump dynamometer cards[J].Oil Field Equipment,2004,33(4):46-48. |
| |
| Authors: | SHI Wen-gang DAI Jing-jun WANG Yong |
| |
| Affiliation: | SHI Wen-gang~1,DAI Jing-jun~1,WANG Yong~2 |
| |
| Abstract: | On the basis of fuzzy membership function in fuzzy theory, an improved SVM multi-classifier, a Fuzzy Support Vector Machine, is presented to solve the unclassified field problem in regular support vector machine multi-classifiers. Furthermore, it is used to classify downhole pump dynamometer cards of some typical faults. The experiment result shows that the improved FSVM classifier has better classification and generalization ability compared with the regular SVM classifier. |
| |
| Keywords: | support vector machine pattern recognition downhole pump dynamometer card |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
