期刊文献+

基于混沌理论和GOA-K-means算法的有载分接开关状态特征分析计算方法 被引量:20

Analysis and Calculation Method of On-Load Tap Changers State Characteristics Based on Chaos Theory and Grasshopper Optimization Algorithm-K-means Algorithm
在线阅读 下载PDF
导出
摘要 为更加准确有效地监测变压器有载分接开关(OLTC)机械状态,针对传统基于K-means的监测方法聚类效果易受其初始聚类中心选择的影响,该文提出一种基于蝗虫算法(GOA)和K-means相结合的OLTC机械状态监测方法。首先针对OLTC振动信号的非线性和混沌特性,利用P-G法和互信息值法计算嵌入维数和延迟时间,对实测的OLTC振动信号进行相空间重构;其次应用Kolmogorov熵对重构后的振动信号混沌特性进行判断;最后为提高聚类精度,针对K-means对初始聚类中心的敏感性,将蝗虫算法引入该算法对其聚类中心进行优化,对重构后的高维振动信号采用优化的K-means聚类方法进行分析。研究结果表明:在OLTC的振动信号识别应用中,优化的K-means聚类算法得到的特征量计算结果具备一定的规律性。研究结果为OLTC的机械运行状态监测提供了一种新的途径。 To monitor the mechanical state of on-load tap changer(OLTC)more accurately and effectively,considering the situation that the clustering effect of traditional monitoring methods based on K-means are affected by the selection of their initial clustering centers,this paper proposed an OLTC mechanical state monitoring method based on the combination of grasshopper optimization algorithm(GOA)and K-means.Firstly,on account of the nonlinear and chaotic characteristics of OLTC vibration signals,the embedding dimension and delay time are calculated by the P-G method and mutual information value method,and the phase space of the measured OLTC vibration signals was reconstructed.Secondly,Kolmogorov entropy was applied to judge the chaotic characteristics of the reconstructed vibration signals.Finally,to improve the clustering accuracy,according to the sensitivity of K-means to the initial clustering center,GOA was introduced into the algorithm to optimize its clustering center,and the reconstructed high-dimensional vibration signals were analyzed by an optimized K-means clustering method.The results showed that in the application of OLTC vibration signal identification,the calculation results obtained by the optimized K-means clustering algorithm have certain regularity,which provides a new way for OLTC machines running state monitoring.
作者 马宏忠 严岩 Ma Hongzhong;Yan Yan(College of Energy and Electrical Engineering Hohai University,Nanjing,211100,China)
出处 《电工技术学报》 EI CSCD 北大核心 2021年第7期1399-1406,共8页 Transactions of China Electrotechnical Society
关键词 蝗虫算法 有载分接开关 优化K-means 振动信号 Grasshopper optimization algorithms on-load tap-changer optimized K-means vibration signal
  • 相关文献

参考文献14

二级参考文献136

共引文献256

同被引文献247

引证文献20

二级引证文献47

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部