王永明,陈宇星,殷自力,李宽宏,张振宇,高 源,罗 翔,高 伟.基于 CWD 及分块 SVD 的配电开关故障诊断方法[J].电子测量与仪器学报,2020,34(7):73-80
基于 CWD 及分块 SVD 的配电开关故障诊断方法
Fault diagnosis method of power distribution switch via CWD and block SVD
  
DOI:
中文关键词:  配电开关  机械状态识别  乔-威廉斯分布  分块奇异值分解  极限学习机
英文关键词:distribution switch  mechanical state recognition  Choi-Williams distribution (CWD)  block singular value decomposition(BSVD)  extreme learning machine (ELM)
基金项目:国家自然科学基金(51677030)、国网福建省电力有限公司科技项目(521304190002)资助
作者单位
王永明 1. 国网福建省电力有限公司 
陈宇星 1. 国网福建省电力有限公司 
殷自力 1. 国网福建省电力有限公司 
李宽宏 2. 国网福建省电力有限公司福州供电公司 
张振宇 3. 国网福建省电力有限公司电力科学研究院 
高 源 3. 国网福建省电力有限公司电力科学研究院 
罗 翔 3. 国网福建省电力有限公司电力科学研究院 
高 伟 4. 福州大学电气工程与自动化学院 
AuthorInstitution
Wang Yongming 1. State Grid Fujian Electric Power Co. , Ltd. 
Chen Yuxing 1. State Grid Fujian Electric Power Co. , Ltd. 
Yin Zili 1. State Grid Fujian Electric Power Co. , Ltd. 
Li Kuanhong 2. State Grid Fujian Electric Power Co. , Ltd. Fuzhou Power Supply Company 
Zhang Zhenyu 3. State Grid Fujian Electric Power Co. , Ltd. Electric Power Research Institute 
Gao Yuan 3. State Grid Fujian Electric Power Co. , Ltd. Electric Power Research Institute 
Luo Xiang 3. State Grid Fujian Electric Power Co. , Ltd. Electric Power Research Institute 
Gao Wei 4. College of Electrical Engineering and Automation, Fuzhou University 
摘要点击次数: 269
全文下载次数: 511
中文摘要:
      通常,配电开关分合闸操作产生的振动信号中蕴含有体现机械状态的重要信息。 提出一种基于振动信号分析的新型配 电开关故障诊断方法。 首先对振动信号求取乔-威廉斯分布获得二维时频矩阵,然后对时频矩阵作分块奇异值分解,用于表征 不同机械状态的时频特性,最后结合极限学习机算法对 4 类实测振动信号的特征向量进行训练和测试。 所提方法的优点是有 效提取了配电开关振动信号时频域的特征,并且可以在较少样本的情况下训练诊断模型。 基于实测数据的实验表明,该方法具 有较高的识别精度和较快的收敛速度。
英文摘要:
      In general, vibration signals generated by the switching operation of a power distribution switch contains important information to reflect its mechanical status. A novel type of fault diagnosis method for a power distribution switch based on vibration signals analyses is proposed in this study. Firstly, the Choi-Williams distribution ( CWD) for the vibration signal is calculated to obtain a twodimensional time-frequency matrix. Then, the block singular value decomposition (BSVD) is performed on the two-dimensional timefrequency matrix, which is used to characterize the time-frequency characteristics of different mechanical states. Finally, the extreme learning machine (ELM) classification algorithm is adopted to train and test the feature vectors of four mechanical states of measured vibration signals. The advantages of the proposed method are that the time domain and frequency domain characteristics of vibration signals inside the power distribution switch are effectively extracted, and the diagnostic model can be trained without many samples. Experiments based on measured data show that the proposed method has a higher recognition accuracy with a faster convergent speed. Keywords:distribution switch; mechanical state recognition; Choi-Williams distribution (CWD); block singular value decomposition
查看全文  查看/发表评论  下载PDF阅读器