摘要
针对多尺度加权排列熵(MWPE)的粗粒化处理易造成信息丢失,无法全面提取齿轮箱故障信息的缺陷,提出了基于改进的精细复合多尺度反向加权排列熵(IRCMRWPE)和支持向量机(SVM)的齿轮箱故障诊断方法。引入精细复合多尺度反向加权排列熵,将其粗粒化处理中的平均值替换为最小值来表征数据段信息,以缓解传统粗粒化处理的不足并强化故障特征的质量;将IRCMRWPE作为表征齿轮箱故障状态的特征向量,使用SVM分类器对故障特征进行识别。基于齿轮箱实验数据对所提方法进行了评估,结果表明该方法可准确地识别齿轮箱的不同故障型,相比于基于MRWPE、MRPE和MWPE的方法,能够更有效的提取齿轮箱的故障特征,增强故障识别的可靠性。
Aiming at the defect that the coarse-grained processing of multi-scale weighted permutation entropy(MWPE) is easy to cause information loss and can not fully extract gearbox fault information,a gearbox fault diagnosis method based on improved refined composite multi-scale reverse weighted permutation entropy(IRCMRWPE) and support vector machine(SVM) is proposed.The refined composite multi-scale reverse weighted permutation entropy is introduced to replace the average value in the coarse-grained processing with the minimum value to represent the data segment information,so as to alleviate the shortcomings of the traditional coarse-grained processing and strengthen the quality of fault features.IRCMRWPE is used as the feature vector to represent the fault state of gearbox,and SVM classifier is used to identify the fault features.The proposed method is evaluated based on the experimental data of gearbox.The results show that the method can accurately identify different fault types of the gearbox,and compared with the methods based on MRWPE,MRPE and MWPE,it can extract the fault features of gearbox more effectively and enhance the reliability of fault recognition.
作者
周云成
潘威
檀青松
ZHOU Yun-cheng;PAN Wei;TAN Qing-song(Inner Mongolia Vocational and Technical College of Communications,Chifeng 024500,China;School of Mechanical Engineering,Hefei University of Technology,Hefei 230009,China;不详)
出处
《组合机床与自动化加工技术》
北大核心
2023年第5期138-142,147,共6页
Modular Machine Tool & Automatic Manufacturing Technique
基金
国家自然科学基金(52075142)
内蒙古教育厅教育科学规划课题(NZJGH2021129)。
关键词
齿轮箱
故障诊断
反向加权排列熵
支持向量机
gearbox
fault diagnosis
reverse weighted permutation entropy
support vector machine
