基于数字孪生与工况自适应的低速重载设备智能运维方法研究及应用
摘要
动的智能运维方法。该方法采用“多源数据自适应融合-动态工况智能划分-分工况故障精准诊断-数字孪生交叉
验证-诊断模型增量自进化”的递进式技术路线。在轧机上的应用结果表明:系统诊断准确率为95.8%,比采用单一
模型提高了33.3个百分点;设备非计划停机率由7‰降低到2‰,提前预报故障平均时间约300小时,为开展预防性
维修工作提供了数据支撑,有效节省了维修费用并提高了设备利用率。
关键词
全文:
PDF参考
[1]Lei Y, Jia F, Lin J, et al. An intelligent fault diagnosis
method using unsupervised representation learning and deep
belief network [J]. Mech Syst Signal Process, 2016, 81:1-14.
[2]Randall R B, Antoni J. Rolling element bearing
diagnostics—A tutorial [J]. Mech Syst Signal Process, 2011,
25(2):485-520.
[3]汤宝平,陈法法,章国稳。基于多传感器信息
融合的设备状态监测与故障诊断综述[J].仪器仪表学报,
2013,34(5):971-979.
[4]Zhang Y, Li X, Wang H, et al. Lightweight fault
diagnosis model for rotating machinery based on attention
mechanism and knowledge distillation [J]. IEEE Trans Ind
Inform, 2020, 16(11):7194-7203.
[5]王树宏,刘杰,张宏,等.数字孪生驱动的轧机
剩余寿命预测方法[J].机械工程学报,2021,57(10):
123-132.
[6]Liu C, Zhang Y, Li X, et al. Digital twin-driven
predictive maintenance for low-speed heavy-load equipment
[J]. IEEE Trans Ind Inform, 2022, 18(6):4567-4576.
[7]Li J, Liu Y, Zhang Z, et al. Federated learningbased collaborative fault diagnosis for low-speed heavy-load
equipment [J]. IEEE Trans Cybern, 2024, 54(3):1890-1901.
[8]Wang S, Chen Z, Li H, et al. Digital twin with
multi-physics coupling for remaining useful life prediction
of heavy-duty bearings [J]. Mech Syst Signal Process, 2024,
210:111856.
Refbacks
- 当前没有refback。
