RFOA优化EEMD阈值和SampEn的水电机组振动信号重构与特征提取  

Vibration Feature Extraction for Hydropower Units Based on RFOA Optimized Ensemble Empirical Mode Decomposition Threshold and Sample Entropy

作　　者：董利江 朱霄珣[3] 刘伟 杨春旭 林翔 高晓霞 吕朝阳 胡乔良 苏海鹏 DONG Li-jiang;ZHU Xiao-xun;LIU Wei;YANG Chun-xu;LIN Xiang;GAO Xiao-xia;LV Zhao-yang;HU Qiao-liang;SU Hai-peng(Urumqi Electric Power Construction and Commissioning Institute of Xinjiang Xinneng Group,Urumqi 830011,China;Institute of Energy Technology,Electric Power Research Institute of State Grid Xinjiang Electric Power,Urumqi 830011,China;Power Engineering Department,North China Electric Power University,Baoding 071003,China)

机构地区：[1]新疆新能集团有限责任公司乌鲁木齐电力建设调试所,新疆乌鲁木齐830011 [2]国网新疆电力有限公司电力科学研究院能源技术研究所,新疆乌鲁木齐830011 [3]华北电力大学动力工程系,河北保定071003

出　　处：《水电能源科学》2023年第11期178-182,共5页Water Resources and Power

基　　金：重庆市科学基金重点项目(035679);2002年高等学校博士学科点专项科研项目(20020183061)。

摘　　要：针对EEMD在水电机组振动信号降噪处理中的不足,提出一种基于改进果蝇算法(RFOA)优化EEMD噪声IMF分量阈值的降噪算法。通过EEMD算法将噪声信号分解,得到IMF分量,进而通过相关系数法确定噪声信号与有效信号,利用RFOA确定噪声信号IMF分量阈值;将求得的IMF分量的样本熵当作特征向量输入GRNN算法中,进行振动模式识别。研究结果表明,与小波阈值法、EEMD-GA方法相比,所提算法降噪时信噪比最高,降噪效果最佳。Aiming at the shortcomings of hydropower unit vibration signal denoising using ensemble empirical mode decomposition(EEMD),a denoising algorithm based on an improved fruit fly optimization algorithm(RFOA) for optimizing the EEMD noise IMF component threshold was proposed.Firstly,the noise signal was decomposed using the EEMD algorithm to obtain the IMF components,and then the correlation coefficient method was used to determine the noise signal and the effective signal.Then,the RFOA was used to determine the noise signal IMF component threshold.Finally,the sample entropy of the obtained IMF components was used as a feature vector input of the GRNN algorithm for vibration mode recognition.Compared with the wavelet threshold method and the EEMD-GA method,the results show that the proposed algorithm has the highest signal-to-noise ratio and the best denoising effect.

关 键 词：振动信号提取 集合经验模态分解 样本熵 特征提取 广义回归神经网络模型 

分 类 号：TV734.1[水利工程—水利水电工程]