一种面向混凝土传送机轴承的轻量化联邦故障诊断方法
By using a lightweight federated learning framework and spectral energy constraints, the problems of high computational resource consumption, data privacy leakage, and insufficient model generalization ability in the fault diagnosis of concrete conveyor bearings are solved, achieving high-precision, lightweight, and interpretable fault diagnosis for edge devices.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ZHEJIANG UNIV OF TECH
- Filing Date
- 2026-03-16
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies for fault diagnosis of concrete conveyor bearings suffer from high computational resource consumption, significant data privacy risks, insufficient model generalization ability, and poor interpretability. In particular, they are difficult to deploy with high precision and lightweight design on edge devices at construction sites.
A lightweight federated learning framework is adopted, which enables fault diagnosis of edge devices through collaborative training between the central server and edge nodes, combined with knowledge distillation and spectral energy concentration constraints.
High-precision, lightweight fault diagnosis was achieved on edge devices, improving the interpretability and generalization ability of the model, reducing the consumption of computing and communication resources, and ensuring data privacy.
Smart Images
Figure CN121882169B_ABST