摘要
针对风力机齿轮箱振动信号非线性和非平稳性的特征,提出基于模糊熵(Fuzzy Entropy,FE)和灰狼算法优化(Grey Wolf Optimizer,GWO)的支持向量机(GWO Support Vector Machine,GWO-SVM)的故障诊断方法。通过集合经验模态分解算法(Ensemble Empirical Mode Decomposition,EEMD)对振动信号进行分解得到若干本征模态函数(Intrinsic Mode Function,IMF)分量;求取各状态IMF分量的模糊熵并构建特征向量;将各特征向量输入GWO-SVM模型进行故障识别及分类。结果表明:齿轮箱振动信号不同状态下的模糊熵有一定区分度,通过GWO-SVM能对其进行精确识别和分类,且GWO-SVM相对于粒子群优化(Particle Swarm Optimization,PSO)SVM模型和遗传算法(Genetic Algorithm,GA)优化SVM模型具有更短的运行时间和更高准确率,平均准确率高达92.5%。
Aiming at the nonlinear and instability characteristics of the wind turbine gearbox bearing fault signal,a method based on the Fuzzy Entropy and Grey Wolf Optimizer Support Vector Machine(GWO-SVM)for the fault diagnosis of gearbox was proposed in this paper.Firstly,EEMD was used to decompose the vibration signal into the several intrinsic mode functions(IMFs).Secondly,calculated the IMFs’fuzzy entropies in each state and constructed feature vectors.Finally,the vectors were adopted as the input parameters for the GWO-SVM to diagnose the fault.The results prove that the fuzzy entropy of gearbox vibration signals in different states has a certain degree of discrimination,it can be identified and classified by GWO-SVM accurately.Meanwhile,GWO-SVM is compared with PSO-SVM and GA-SVM,it has shorter time and higher accuracy,the mean accuracy can up to 92.5%.
作者
胡璇
李春
叶柯华
HU Xuan;LI Chun;YE KeHua(School of Energy and Power Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China;Shanghai Key Laboratory of Multiphase Flow and Heat Transfer in Power Engineering,Shanghai 200093,China)
出处
《机械强度》
CAS
CSCD
北大核心
2021年第5期1026-1034,共9页
Journal of Mechanical Strength
基金
国家自然科学基金项目(51976131,51676131)
上海市“科技创新行动计划”地方院校能力建设项目(19060502200)资助。
关键词
风力机齿轮箱
故障诊断
集合经验模态分解
灰狼算法优化
支持向量机
模糊熵
Wind turbine gearbox
Fault diagnosis
Ensemble empirical mode decomposition
Grey wolf optimizer
Support vector machine
Fuzzy entropy