Spark平台环境下基于Aco-k means算法的滚轴故障检测算法研究  

FAULT PREDICTION METHOD OF ROLLING BEARING BASED ON ACO-K MEANS ALGORITHM IN SPARK PLATFORM

作　　者：刘兴建 原振文 Liu Xingjian;Yuan Zhenwen(Department of Computer Application Technology,Guangdong Business and Technology University,Zhaoqing 526040,Guangdong,China;Artillery Academy,National University of Defense Technology,Changsha 410111,Hunan,China)

机构地区：[1]广东工商职业技术大学计算机应用技术系,广东肇庆526040 [2]国防科技大学炮兵学院,湖南长沙410111

出　　处：《计算机应用与软件》2021年第1期256-261,共6页Computer Applications and Software

基　　金：广东省教育厅高校特色创新类项目(自然科学)(2017GKTSCX110)。

摘　　要：针对现有滚轴故障预测方法预测精度差、效率低的不足,提出一种基于Aco-k means算法的滚轴故障预测方法。在Spark平台环境下,利用小波包变换提取滚轴故障信号的时频域特征,并对滚动轴承故障特征进行聚类分析;引入Aco仿生算法确定故障特征数据聚类中心及故障样本数据间的欧氏距离,在全局范围内寻优并确定滚轴故障的类别与严重程度。仿真结果证明,该方法在滚轴故障预测精度方面具有明显优势,其故障特征提取效率也高于传统算法。In order to overcome the shortcomings of poor prediction accuracy and low efficiency of existing rolling shaft fault prediction methods,a rolling shaft fault prediction method based on Aco-k means algorithm is proposed.In the environment of Spark platform,the time-frequency domain features of roller fault signals were extracted based on wavelet packet transform,and the fault features of rolling bearings were clustered.Aco bionic algorithm was introduced to determine the Euclidean distance between the fault feature data clustering center and the fault sample data,so as to optimize the whole situation and determine the roller fault accurately,classification and severity of impairment.The simulation results show that the proposed method has obvious advantages in the accuracy of fault prediction,and the efficiency of fault feature extraction is also higher than the traditional algorithm.

关 键 词：大数据 Spark平台 Aco-k means算法 全局寻优 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置] TP3[自动化与计算机技术—计算机科学与技术]