基于AVMD与Teager能量算子的风电机组故障诊断方法  

Wind Turbine Fault Diagnosis Method Based on AVMD and Teager Energy Operator

作　　者：时培明[1] 伊思颖 张慧超 范雅斐 韩东颖[2] SHI Peiming;YI Siying;ZHANG Huichao;FAN Yanfei;HAN Dongying(School of Electrical Engineering,Yanshan University Qinhuangdao,066004,China;School of Vehicle and Energy,Yanshan University Qinhuangdao,066004,China)

机构地区：[1]燕山大学电气工程学院,秦皇岛066004 [2]燕山大学车辆与能源学院,秦皇岛066004

出　　处：《振动.测试与诊断》2025年第2期390-397,418,共9页Journal of Vibration,Measurement & Diagnosis

基　　金：河北省自然科学基金资助项目(F2024203035);湖北省工程研究中心开放课题资助项目(IDICP-KF-2024-10)。

摘　　要：为解决变分模态分解(variational mode decomposition,简称VMD)在噪声情况下提取风电机组故障特征时因参数设置的人为经验不足而带来的误差问题及耗费时间的问题,提出一种基于自适应变分模态分解(adaptive variational mode decomposition,简称AVMD)算法的风电机组故障诊断方法。首先,将包络熵-峭度-互信息准则(envelope entropy,kurtosis and mutual information,简称EKM)作为黏菌算法(slime mold algorithm,简称SMA)的适应度函数来寻找最优解,并按照最优解对故障信号进行分解;其次,计算每个固有模态函数分量(inherent modal function,简称IMF)的峭度和与原信号的互信息,选择具有故障特征的分量进行重构;最后,通过Teager能量算子解调来识别风电机组故障特征频率。仿真信号和实际风电机组故障信号表明,所提方法能够找到故障频率及其倍频,验证了其在风电机组故障诊断领域中的有效性。In order to solve the problem of error caused by lack of human experience in parameter setting and the problem of wasting a lot of time due to continuous attempts when variational mode decomposition(VMD) is used to analyze the fault characteristics of wind turbine generator units under noise background,an adaptive variational mode decomposition(AVMD)-based fault diagnosis method is proposed.The procedure includes:The criterion of envelope entropy,kurtosis and mutual information(EKM) is used as the fitness function of the slime mold algorithm(SMA) to find the optimal solution;The fault signal is decomposed according to the optimal solution.Next,the kurtosis of each intrinsic modal function(IMF) and the mutual information with the original signal are calculated,and the components with fault characteristics are selected for reconstruction.Finally,the characteristic frequency of wind turbine fault is identified by Teager energy operator demodulation.The method is verified by the simulation signal and the actual fault signal of wind turbine,which can find the fault frequency and double frequency well,and has certain effectiveness in the field of fault diagnosis of wind turbine.

关 键 词：自适应变分模态分解 黏菌算法 包络熵-峭度-互信息准则 TEAGER能量算子 

分 类 号：TH133.3[机械工程—机械制造及自动化] TS737.1[轻工技术与工程—制浆造纸工程]