基于ABC-BW优化CHMM的风机齿轮箱故障诊断  

Fault Diagnosis of Fan Gearboxes Based on ABC-BW Optimized CHMM

作　　者：李韵仪 沈艳霞 LI Yunyi;SHEN Yanxia(Engineering Research Center of Application of Things Internet Technology,Jiangnan University,Wuxi 214122,Jiangsu,China)

机构地区：[1]江南大学物联网技术应用教育部工程研究中心,江苏无锡214122

出　　处：《噪声与振动控制》2021年第4期80-85,233,共7页Noise and Vibration Control

基　　金：国家自然科学基金资助项目(61573167,61572237)。

摘　　要：提出一种将人工蜂群算法(Artificial Bee Colony Algorithm,ABC)与Baum-Welch算法结合的连续隐马尔科夫模型(Continuous Hidden Markov Model,CHMM)并将其应用于风力发电机齿轮箱故障诊断。首先利用小波包分解与重构提取信号频带能量作为特征向量;将正常及各故障状态的训练样本特征作为模型观测值输入CHMM模型与优化模型进行训练,最后将各个检验样本特征输入各状态模型中进行对比,得到待测样本在训练模型中的输出概率,将其作为故障诊断的评判依据。A continuous hidden Markov model(CHMM),which combines artificial bee colony(ABC)algorithm with Baum-Welch algorithm,is proposed and applied to the fault diagnosis of the fan gearbox of a wind turbine generator.Firstly,wavelet packet decomposition and reconstruction are used to extract signal band energy as the feature vector.The characteristics of training samples of different states are input into the CHMM model and the optimization model for training.Finally,the characteristics of each test sample are input into each state model for comparison.The output probabilities of the test samples in the training model are obtained and used as the evaluation bases for fault diagnosis.

关 键 词：故障诊断 风机齿轮箱 连续隐马尔科夫模型 小波包变换 人工蜂群算法 

分 类 号：TP315[自动化与计算机技术—计算机软件与理论]