数据缺失条件下基于ANFIS与k-means的轴承故障分析  

Bearing Failure Analysis Based on ANFIS and K-means under the Condition of Data Loss

作　　者：史海鹏 陈家兑[1] 吴永明[1] 王波[1] 陈琳升 SHI Hai-peng;CHEN Jia-dui;WU Yong-ming;WANG Bo;CHEN Lin-sheng(Key Laboratory of Advanced Manufacturing Technology,Ministry of Education,Guizhou University,Guiyang 550025,China)

机构地区：[1]贵州大学,现代制造技术教育部重点实验室,贵阳550025

出　　处：《组合机床与自动化加工技术》2020年第9期33-36,共4页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金资助项目(51505094);贵州省科学技术基金计划项目[(2016)1037];贵州省科技支撑计划项目[(2017)2029];贵州大学引进人才科研项目[贵大人基合字(2014)60号]。

摘　　要：针对轴承信号数据信息采集过程中轴承数据缺失导致轴承故障类型无法识别问题,提出了一种基于自适应神经模糊推理系统(ANFIS)和k-means聚类算法相结合的轴承故障分析模型和方法。首先,基于ANFIS算法建立数据预测模型,利用75%的轴承故障数据作为训练样本,对缺失信号数据进行预测,同时与采集到的信号数据整合形成完整的数据集,然后利用k-means算法进行轴承故障诊断,最后将ANFIS-k均值方法与传统k-means算法进行比较,实验结果表明本文提出的模型和方法分类效果更加准确。A bearing fault analysis model and method based on the combination of adaptive neural fuzzy inference system(ANFIS) and k-means clustering algorithm was proposed to solve the problem of bearing fault type identification caused by bearing data missing in the process of bearing signal data collection. First of all, establish data forecasting model based on ANFIS algorithm, with 75% of the bearing fault data as the training sample, to forecast the missing data signal, the signal data were collected at the same time, integration of form a complete data set, then the use of k-means algorithm for bearing fault diagnosis, the final will be ANFIS-k-means method comparing with the traditional k-means algorithm, the experimental results show that the proposed model and method of the classification result more accurate.

关 键 词：数据缺失 ANFIS 故障分析 K-MEANS算法 

分 类 号：TH162[机械工程—机械制造及自动化] TG506[金属学及工艺—金属切削加工及机床]