基于MOGOA-VMD-LSSVM的轴承故障诊断方法研究  

Fault diagnosis of hoist main bearing based on MOGOA-VMD-LSSVM

作　　者：张辉 宋泓炎 范华超 赵连明 江帆[2] 鲁宗虎 ZHANG Hui;SONG Hongyan;FAN Huachao;ZHAO Lianming;JIANG Fan;LU Zonghu(CHN Energy Group Xinjiang Energy Co.,Ltd.,Urumqi 830063,China;School of Mechanical and Electrical Engineering,China University of Mining and Technology,Xuzhou 221008,China;Xinjiang Industrial Cloud Big Data Innovation Center Co.,Ltd.,Urumqi 830022,China)

机构地区：[1]国家能源集团新疆能源有限责任公司,新疆乌鲁木齐830063 [2]中国矿业大学机电工程学院,江苏徐州221008 [3]新疆工业云大数据创新中心有限公司,新疆乌鲁木齐830022

出　　处：《煤炭工程》2025年第2期149-155,共7页Coal Engineering

基　　金：国家自然科学基金项目(52374163,51605478);江苏省科技成果转化专项资金项目(BA2022075);江苏省高校青蓝工程优秀青年骨干教师项目(苏教师函〔2021〕11号);江苏省高校优势学科建设工程资助项目(PAPD)。

摘　　要：针对煤基活性炭生产设备轴承故障类型难以准确诊断的问题,提出了一种多目标蝗虫优化算法(MOGOA)优化变分模态分解(VMD)与最小二乘支持向量机(LSSVM)的煤基活性炭生产设备轴承故障诊断方法。首先,针对传统蝗虫优化算法(GOA)参数敏感、易于陷入局部最优的问题,引入多目标蝗虫优化算法,通过引入基于排列熵与峭度倒数归一化的复合适应度函数,优化VMD的惩罚因子和分解层数。其次,使用优化VMD分解提取的轴承振动信号并筛选出敏感变分模态分量(IMF)进行重构。最后,通过MOGOA优化LSSVM模型,形成MOGOA-LSSVM故障诊断模型。与GOA-LSSVM方法对比,本研究所提方法故障诊断准确率提高了5%,运行时间缩短了9.72 s,验证了该方法在故障诊断方面的优势。Aiming at the difficulty in precise fault diagnosis of coal-based activated carbon production equipment,we proposed a multi-objective locust optimization algorithm(MOGOA)optimized variational mode decomposition(VMD)and least squares support vector machine(LSSVM)for the accurate diagnosis of bearing fault types in coal based activated carbon production equipment.Firstly,to address the problem of parameter sensitivity and susceptibility to local optima in traditional locust optimization algorithms(GOA),a multi-objective locust optimization algorithm was introduced.By introducing a composite fitness function based on permutation entropy and kurtosis reciprocal normalization,the penalty factor and decomposition level of VMD were optimized.Secondly,optimized VMD decomposition was used to extract bearing vibration signals and select sensitive variational mode components(IMF)for reconstruction.Finally,by optimizing the LSSVM model through MOGOA,a MOGOA-LSSVM fault diagnosis model was formed.Compared with the GOA-LSSVM method,the fault diagnosis accuracy of the proposed method was 5%higher,and the running time was 9.72 seconds shorter,verifying the advantages of the proposed method in fault diagnosis.

关 键 词：煤基活性炭设备 轴承 多目标蝗虫优化算法 VMD LSSVM 

分 类 号：TD534[矿业工程—矿山机电]