基于解析小波的时间序列数据增强的故障诊断方法及系统
By using a time-series data augmentation method based on analytic wavelets, the generated augmented dataset is used to train a convolutional neural network, which solves the problem of insufficient data in mechanical fault diagnosis and enables high-precision fault detection of key components such as rolling bearings.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SEVNCE ROBOTICS CO LTD
- Filing Date
- 2025-07-15
- Publication Date
- 2026-07-17
AI Technical Summary
In existing technologies, mechanical fault diagnosis models lack sufficient fault data, resulting in poor training performance of deep learning models. In particular, it is difficult to achieve high accuracy and robustness in fault diagnosis of key components such as rolling bearings.
A time series data augmentation method based on analytic wavelet is adopted. The original and augmented scale maps are generated by generalized Morse analytic wavelet transform. The parameter β is optimized by combining the Heisenberg uncertainty principle. The generated augmented dataset is used to train a convolutional neural network to improve data richness and feature extraction accuracy.
The training dataset has been significantly expanded, improving the classification accuracy and generalization ability of the fault diagnosis model. It is applicable to fault diagnosis of various mechanical equipment, especially the accurate detection of rolling bearings.
Smart Images
Figure CN120950951B_ABST