融合ISSA和TA-CapNets的矿井滚动轴承故障诊断方法  

Mine Rolling Bearing Fault Diagnosis Method Based on ISSA and TA-CapNets

作　　者：屈波 张兰峰 王惠伟 闫明 周超逸 QU Bo;ZHANG Lanfeng;WANG Huiwei;YAN Ming;ZHOU Yichao(Guoneng Shendong Coal Intelligent Technology Center,Yulin 719315,China;Shaanxi Yijiexin Information Technology Co.,Ltd.,Xi′an 710065,China)

机构地区：[1]国能神东煤炭智能技术中心,陕西榆林719315 [2]陕西亿杰鑫信息技术有限公司,陕西西安710065

出　　处：《金属矿山》2025年第4期226-232,共7页Metal Mine

基　　金：中国神华能源股份有限公司神东煤炭分公司科技创新项目(编号:E210100270)。

摘　　要：滚动轴承作为矿井设备的核心部件,其运行状态直接关系到矿山生产安全和经济效益。为提升矿井滚动轴承故障诊断的性能,提出了一种融合改进麻雀搜索算法(Improved Sparrow Search Algorithm,ISSA)和时频自适应胶囊网络(Time-Frequency Adaptive Capsule Networks,TA-CapNets)的新型诊断方法。首先,通过采集矿井滚动轴承的运行数据,提取出反映轴承健康状况的特征;然后利用ISSA对特征进行优化选择,该算法通过模拟麻雀觅食行为,提高了全局搜索能力和收敛速度;再将优化后的特征输入TA-CapNets中,能够自适应地学习时频特征,有效捕捉轴承故障的动态变化。通过TA-CapNets的输出,结合故障模式识别,实现了对轴承故障的准确诊断。试验结果表明:该方法在故障诊断的准确性和实时性方面均优于传统方法,具有一定的实用价值和推广前景。As the core component of mine equipment,the running state of rolling bearing is directly related to the safety and economic benefits of mine production.In order to improve the fault diagnosis performance of mine rolling bearing,a new diagnosis method combining Improved Sparrow Search Algorithm(ISSA)and Time-Frequency Adaptive Capsule Network(TACapNets)was proposed.Firstly,by collecting the running data of mine rolling bearing,the characteristics of bearing health status are extracted.Then,ISSA is used to optimize the selection of features.The algorithm improves the global search ability and convergence speed by simulating the foraging behavior of sparrows.Finally,the optimized features are input into TA-CapNets,which can adaptively learn the time-frequency features and effectively capture the dynamic changes of bearing faults.The accurate diagnosis of bearing faults is realized by TA-CapNets output and fault pattern recognition.The experimental results show that this method is superior to the traditional method in both accuracy and real-time fault diagnosis,and has certain practical value and popularization prospect.

关 键 词：矿井滚动轴承 故障诊断 改进麻雀搜索算法 时频自适应胶囊网络 

分 类 号：TD67[矿业工程—矿山机电]