迁移学习在机械设备故障诊断领域的进展研究
A Study on the Progress of Transfer Learning in the Field of Mechanical Equipment Fault Diagnosis
陈驻民 1韦继程2
作者信息
- 1. 上海第二工业大学智能制造与控制工程学院,上海 201209
- 2. 上海第二工业大学计算机与信息工程学院,上海 201209
- 折叠
摘要
迁移学习是一种新兴的机器学习方法,通过运用已学习的知识对不同但相关领域问题进行求解,能够较为有效的解决模型泛化能力弱、样本数据不足等问题.针对迁移学习在机械设备故障诊断领域的应用方法进行了综述,总结了三类关于迁移学习的诊断预测方法,并对迁移学习在故障诊断领域的未来研究方向进行了探讨.
Abstract
Migration learning is an emerging machine learning method,which can solve the problems of different but related domains by applying the learned knowledge,and can solve the problems of weak model generalisation ability and insufficient sample data more effectively.This paper provides an overview of the application of transfer learning in the field of mechanical equipment fault diagnosis,summarises three types of diagnostic prediction methods on transfer learning,and discusses the future research direction of transfer learning in the field of fault diagnosis.
关键词
迁移学习/故障诊断/参数微调/特征对齐/生成对抗网络Key words
migration learning/fault diagnosis/parameter fine-tuning/feature alignment/generative adversarial network引用本文复制引用
出版年
2024
NETL
NSTL
万方数据