基于分层聚合机制的有监督联邦迁移滚动轴承故障诊断方法及系统

By introducing a hierarchical aggregation mechanism and parameter transfer strategy into the federated learning framework, the problem of low model accuracy caused by non-independent and identically distributed data in traditional federated learning is solved, and high-precision fault diagnosis is achieved under small sample conditions.

CN120724237BActive Publication Date: 2026-07-17HARBIN INST OF TECH
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
HARBIN INST OF TECH
Filing Date
2025-06-13
Publication Date
2026-07-17

Smart Images

  • Figure CN120724237B_ABST
    Figure CN120724237B_ABST
Patent Text Reader

Abstract

本发明涉及一种基于分层聚合机制的有监督联邦迁移滚动轴承故障诊断方法及系统，包括：步骤1、将源域数据集和目标域数据集均划分成训练数据集和测试数据集两种类型进行均值为0、方差为1的标准化处理；步骤2、在客户端构建基于卷积神经网络的轴承故障诊断本地模型；步骤3、将与目标域诊断任务相关的轴承数据集作为源域数据集；步骤4、更新轴承故障诊断本地模型的特征提取模块和分类模块；步骤5、客户端基于分层聚合机制再次更新轴承故障诊断本地模型的相应参数；步骤6、重复步骤4和步骤5，直至轴承故障诊断本地模型收敛或达到最大迭代次数，输出各客户端的数据故障诊断结果。该方法为联邦客户端建立个性化的故障诊断模型提供了有效方案。
Need to check novelty before this filing date? Find Prior Art