Midspan Deflection Prediction of Long-Span Cable-Stayed Bridge Based on DIWPSO-SVM Algorithm
With the increasing emphasis on the safety and longevity of large-span cable-stayed bridges, the accurate prediction of midspan deflection has become a critical aspect of structural health monitoring (SHM). This study proposes a novel hybrid model, DIWPSO-SVM, which integrates dynamic inertia weight...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-05-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/10/5581 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | With the increasing emphasis on the safety and longevity of large-span cable-stayed bridges, the accurate prediction of midspan deflection has become a critical aspect of structural health monitoring (SHM). This study proposes a novel hybrid model, DIWPSO-SVM, which integrates dynamic inertia weight particle swarm optimization (DIWPSO) with support vector machines (SVMs) to enhance the prediction accuracy of midspan deflection. The model incorporates wavelet transform to decompose deflection signals into temperature and vehicle load effects, allowing for a more detailed analysis of their individual impacts. The DIWPSO algorithm dynamically adjusts the inertia weight to balance global exploration and local exploitation, optimizing SVM parameters for improved performance. The proposed model was validated using real-world data from a long-span cable-stayed bridge, demonstrating superior prediction accuracy compared to traditional SVM and PSO-SVM models. The DIWPSO-SVM model achieved an average prediction error of 1.43 mm and a root-mean-square error (RMSE) of 2.05, significantly outperforming the original SVM model, which had an average error of 5.29 mm and an RMSE of 5.62. These results highlight the effectiveness of the DIWPSO-SVM model in providing accurate and reliable midspan deflection predictions, offering a robust tool for bridge health monitoring and maintenance decision-making. |
|---|---|
| ISSN: | 2076-3417 |