Model Evolution Mechanism for Incremental Fault Diagnosis
Model Evolution Mechanism for Incremental Fault Diagnosis
复制标题
增量故障诊断的模型演化机制
DOI：
10.1109/tim.2022.3200695
复制
发表时间：
2022
影响因子：
5.6
通讯作者：
Dongyuan Wang
中科院分区：
文献类型：
--
作者：
Liuen Guan;Fei Qiao;Xiaodong Zhai;Dongyuan Wang
故障诊断作为设备运行维护的关键问题，受到了广泛的关注，其中深度学习常被应用于故障诊断模型的构建。然而，工业系统运行状态的变化会导致模型性能的下降。因此，对模型的动态更新进行探索是十分必要和迫切的。建立了增量式故障诊断框架，提出了基于自适应知识提取和典型样本选择的模型进化机制。首先，当新样本涌入时，利用知识蒸馏（KD）挖掘新老样本之间的潜在相关性，并对模型参数优化进行自适应约束，以实现模型的更新。其次，遗传算法（GA）被应用于选择性地保留代表性的样本子集，以提高模型的处理动态数据的能力。此外，为了进一步验证MEMAR的有效性，测试了多阶段增量学习后的故障诊断模型的判别性能。实验结果表明，MEMAR算法可以减少旧样本的存储空间和训练时间。此外，它使基于深度学习的故障诊断模型对新样本和旧样本都具有良好的诊断性能。
As the key issue of equipment operation and maintenance, fault diagnosis has attracted extensive attention, in which deep learning is often applied to the construction of fault diagnosis models. However, the changing operation state of industrial systems will lead to the decline of model performance. Therefore, it is necessary and urgent to explore the dynamic update of models. This work establishes an incremental fault diagnosis (IFD) framework, and further proposes a model evolution mechanism based on adaptive knowledge distillation and representative exemplar selection (MEMAR). Firstly, when new samples pour in, knowledge distillation (KD) is used to mine the potential correlation between old and new samples, with adaptive constraints on parameter optimization for model updating. Secondly, genetic algorithm (GA) is applied to selectively retain representative exemplar subsets, so as to enhance the model2019;s ability to deal with dynamic data. Moreover, in order to further verify the effectiveness of MEMAR, the discrimination performance of the fault diagnosis model after multi-stage incremental learning is tested. The experimental results show that the proposed MEMAR can reduce the storage space of old samples and training time. Besides, it enables the fault diagnosis model based on deep learning to have excellent diagnostic performance for new and old samples.
登录
查看更多内容
DOI：
10.1007/978-0-387-39940-9_3525
发表时间：
2018-09
期刊：
2006 15th IEEE International Conference on High Performance Distributed Computing
影响因子：
--
作者：
Susmita Bandyopadhyay
通讯作者：
Susmita Bandyopadhyay
用于化学故障诊断的具有动态增量学习的不平衡改进深度神经网络
影响因子：
7.7
作者：
Zhixin Hu;Peng Jiang
通讯作者：
Peng Jiang
DOI：
10.1109/vts.2019.8758599
发表时间：
2019-04
期刊：
2019 IEEE 37th VLSI Test Symposium (VTS)
影响因子：
--
作者：
Mengyun Liu;Fangming Ye;Xin Li;K. Chakrabarty;Xinli Gu
通讯作者：
Mengyun Liu;Fangming Ye;Xin Li;K. Chakrabarty;Xinli Gu
具有增量学习能力的基于广泛卷积神经网络的工业过程故障诊断
影响因子：
7.7
作者：
Yu, Wanke;Zhao, Chunhui
通讯作者：
Zhao, Chunhui
DOI：
10.1109/icphm.2018.8448775
发表时间：
2018-06
期刊：
2018 IEEE International Conference on Prognostics and Health Management (ICPHM)
影响因子：
--
作者：
Wentao Mao;Siyu Tian;Xihui Liang;Jianliang He
通讯作者：
Wentao Mao;Siyu Tian;Xihui Liang;Jianliang He