| 引用本文: | 郝思鹏,张济韬,张仰飞,张小莲.融合在线监测数据的变压器状态评估[J].电力自动化设备,2017,37(11): |
| HAO Sipeng,ZHANG Jitao,ZHANG Yangfei,ZHANG Xiaolian.State evaluation of transformer based on information fusion of on-line monitoring data[J].Electric Power Automation Equipment,2017,37(11): |
|
| 摘要: |
| 随着技术的发展，变压器状态评估逐步趋向融合动态在线监测和静态预防性试验数据的综合评价。基于在线油色谱监测数据，提出利用小波模极大值识别快速渐变拐点和跃变点的算法，提高了在线监测评价的准确度。在此基础上，考虑在线监测和预防性试验数据的不同时效，在双层结构的变压器综合状态评价模型中，提出采用时间可信度指标修正各子证据体的可信度，实现了静态数据参与状态评估的动态化处理。实例分析结果表明，融合在线监测数据的变压器综合评估模型较传统变压器状态评估更加准确。 |
| 关键词: 变压器 状态评估 在线监测数据 小波模极大值 多信息融合 时间可信度指标 |
| DOI：10.16081/j.issn.1006-6047.2017.11.028 |
| 分类号:TM41 |
| 基金项目:国家自然科学基金资助项目(51607083)；江苏省产学研前瞻性项目(BY2015009-03)；江苏省自然科学基金资助项目(BK20160778) |
|
| State evaluation of transformer based on information fusion of on-line monitoring data |
|
HAO Sipeng1, ZHANG Jitao2, ZHANG Yangfei1, ZHANG Xiaolian1
|
|
1.School of Electric Power Engineering, Nanjing Institute of Technology, Nanjing 211167, China;2.State Grid Xiangshui Power Supply Company, Xiangshui 224600, China
|
| Abstract: |
| With the development of technology, transformer state evaluation is gradually developing towards comprehensive evaluation based on the fusion of dynamic on-line monitoring data and static preventive test data. Based on on-line oil chromatography monitoring data, an algorithm to identify the rapid gradient inflection points and jump points is proposed by using wavelet modulus maximum values, improving the accuracy of on-line monitoring and evaluation. On this basis, considering the different effects of on-line monitoring data and preventive test data, the confidence level of each sub evidence body is corrected by time reliability index in the double-layer comprehensive state evaluation model of transformer, to realize the dynamic processing deal with the of static data participated in state evaluation. Results of case analysis show that the transformer comprehensive state evaluation model which integrates the on-line monitoring data is more accurate compared with traditional transformer state evaluation. |
| Key words: power transformers state evaluation on-line monitoring data wavelet modulus maximum value multiple information fusion time reliability index |
