| 基于样本自举的电力变压器状态评估 |
| |
| 引用本文： | 郭磊,张圣楠,张雷,刘佳,陈丽. 基于样本自举的电力变压器状态评估[J]. 电力科学与技术学报, 2013, 0(4): 10-14 |
| |
| 作者姓名： | 郭磊 张圣楠 张雷 刘佳 陈丽 |
| |
| 作者单位： | [1]国家电网华北电力调控分中心,北京100053 [2]重庆市电力公司检修分公司,重庆400039 [3]中国电力出版社,北京100053 |
| |
| 基金项目： | 国家自然科学基金（51177143）;浙江省自然科学基金（LZl2E07002）;国家重点基础研究计划（“973”计划）（2013CB228206） |
| |
| 摘 要： | 针对传统数据学习型方法需要大量故障历史实测数据的缺点，提出一种基于数据样本自举的电力变压器状态评估方法．通过对变压器的故障样本数据进行自举扩充，克服了工程中某类样本数据较少的问题，提高了学习型分类器的训练量，从而提高其预测分类的精度．并基于支持向量机分类方法进行变压器故障分类评估，显著提升了评估精度．
|
| 关 键 词： | 变压器故障 状态检修 自举法 支持向量机 |
Power transformer condition evaluation based on samples bootstrap method |
| |
| GUO Lei,ZHANG Sheng-nan,ZHANG Lei,LIU Jia,CHEN Li. Power transformer condition evaluation based on samples bootstrap method[J]. JOurnal of Electric Power Science And Technology, 2013, 0(4): 10-14 |
| |
| Authors: | GUO Lei ZHANG Sheng-nan ZHANG Lei LIU Jia CHEN Li |
| |
| Affiliation: | 1. North Subsection of State Grid Corporation of China,Beijing 100053,China; 2. Maintenance Branch of Chongqing Electric Power Corporation, Chongqing 400039,China 3. China Electric Power Press, Beijing 100053,China) |
| |
| Abstract: | In order to solve the problem that traditional data learning methods depend on too much historical faults data, a new samples bootstrap based method for power transformers condition evaluation was proposed in this paper. The bootstrap can extend transformer faults data samples, and the classification accuracy can be highly improved. Using SVM （Support Vector Machine） classi{ication method, the transformer faults classification evaluation was realized and the evalua- tion accuracy was also highly improved. |
| |
| Keywords: | transformer faults condition maintenance bootstrap SVM |
| 本文献已被 维普 等数据库收录！ |
|
