Research on bearing fault diagnosis based on improved northern goshawk algorithm optimizing SVM
An improved northern goshawk optimization (INGO) algorithm was proposed to address the local optimization problem that swarm intelligence algorithms often encounter when optimizing support vector machine (SVM) models, and it was applied to fault diagnosis of rolling bearings. By introducing an adapt...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Office of Journal of Mechanical Strength
2025-05-01
|
| Series: | Jixie qiangdu |
| Subjects: | |
| Online Access: | http://www.jxqd.net.cn/thesisDetails#DOI：10.16579/j.issn.1001.9669.2025.05.010 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | An improved northern goshawk optimization (INGO) algorithm was proposed to address the local optimization problem that swarm intelligence algorithms often encounter when optimizing support vector machine (SVM) models, and it was applied to fault diagnosis of rolling bearings. By introducing an adaptive inertia weight factor based on the cosine variation and a Cauchy mutation strategy, the northern goshawk optimization (NGO) algorithm was improved, and an INGO-SVM fault diagnosis model was constructed using SVM. In order to evaluate the performance of the improved algorithm, firstly, benchmark testing functions were used for experiments, and the improved algorithm was compared with existing optimization algorithms such as NGO, particle swarm optimization (PSO), sparrow search algorithm (SSA), etc. The results show that the performance of the improved algorithm is improved to a certain extent. At the same time, the original diagnostic signals were feature extracted through wavelet packet decomposition and divided into 10 categories. The energy of each frequency band in the 3rd layer was used as the feature vector and input into the fault diagnosis model. Finally, the performance of the improved algorithm was compared with the other three algorithms in optimizing SVM parameters for fault classification. The results show that the improved algorithm can effectively and accurately achieve different fault classifications, with an accuracy rate of 99.39%, verifying the effectiveness and feasibility of this method. |
|---|---|
| ISSN: | 1001-9669 |