Wen-Hwa Wu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Wen-Hwa Wu has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
Chien-Chou Chen
Department of Civil and Construction Engineering, National Yunlin University of Science and Technology, Yunlin, Taiwan
14 shared publicationsGwolong Lai
Department of Civil and Construction Engineering, National Yunlin University of Science and Technology, Yunlin, Taiwan
12 shared publications- Chien-Chou Chen
National Yunlin University of Science and Technology
6 shared publications - Gwolong Lai
Department of Civil and Construction Engineering National Yunlin University of Science and Technology Yunlin 640 Taiwan yuntech.edu.tw
4 shared publications Shang-Li Lin
3 shared publicationsSheng-Wei Wang
Intelligent Electronic Systems Division, National Chip Implementation Center, National Applied Research Laboratories, Hsinchu City, Taiwan
3 shared publications
Latest Publications
- 15 September 2026
- Published by Elsevier in Measurement
- Vol. 292
Non-contact stay-cable tension estimation from gravitational geometry: comparative field evaluation of total-station and mobile LiDAR measurements
- 15 September 2026
- Published by Elsevier in Measurement
- Vol. 292
Non-contact stay-cable tension estimation from gravitational geometry: comparative field evaluation of total-station and mobile LiDAR measurements
- 26 May 2026
- Published by SAGE Publications in Structural Health Monitoring
Automated real-time cable tension warning framework using long-term monitoring data and a convolutional neural network
Stay cables are critical load-bearing components in cable-stayed bridges, and their tension variations serve as direct indicators of structural safety and serviceability. While vibration-based monitoring has enabled continuous tension estimation, the reliable transformation of such measurements into real-time warning information remains challenging. This study presents an automated real-time cable tension warning framework that integrates ambient vibration measurements with a convolutional neural network (CNN). A statistical detection algorithm was first developed by defining thresholds from year-long tension histories and applying a three-segment strategy to suppress scattered values. Using long-term monitoring data from representative cables, this method successfully identified abnormal tension variations and quantified the minimum detectable changes and detection times. The proposed framework emphasizes warning logic and long-term operational performance by explicitly evaluating false-alarm behavior, detectability limits, and detection latency under unfavorable conditions. To overcome the inherent limitations of rule-based thresholds, the anomaly detection task was reformulated as a CNN-based image classification problem. By encoding tension data into one-dimensional images and categorizing them into low, normal, and high states, the CNN consistently achieved 100% accuracy in distinguishing normal from abnormal conditions. The contributions of this work are threefold: (1) development of a statistical threshold-based detection algorithm validated with 1 year of field data; (2) reformulation of anomaly detection as a CNN-based classification task; and (3) demonstration of perfect classification accuracy through extensive testing. These results establish a practical, adaptive framework for automated cable tension warning, extending conventional monitoring into proactive decision-making support for bridge management.
- 26 May 2026
- Published by SAGE Publications in Structural Health Monitoring
Automated real-time cable tension warning framework using long-term monitoring data and a convolutional neural network
Stay cables are critical load-bearing components in cable-stayed bridges, and their tension variations serve as direct indicators of structural safety and serviceability. While vibration-based monitoring has enabled continuous tension estimation, the reliable transformation of such measurements into real-time warning information remains challenging. This study presents an automated real-time cable tension warning framework that integrates ambient vibration measurements with a convolutional neural network (CNN). A statistical detection algorithm was first developed by defining thresholds from year-long tension histories and applying a three-segment strategy to suppress scattered values. Using long-term monitoring data from representative cables, this method successfully identified abnormal tension variations and quantified the minimum detectable changes and detection times. The proposed framework emphasizes warning logic and long-term operational performance by explicitly evaluating false-alarm behavior, detectability limits, and detection latency under unfavorable conditions. To overcome the inherent limitations of rule-based thresholds, the anomaly detection task was reformulated as a CNN-based image classification problem. By encoding tension data into one-dimensional images and categorizing them into low, normal, and high states, the CNN consistently achieved 100% accuracy in distinguishing normal from abnormal conditions. The contributions of this work are threefold: (1) development of a statistical threshold-based detection algorithm validated with 1 year of field data; (2) reformulation of anomaly detection as a CNN-based classification task; and (3) demonstration of perfect classification accuracy through extensive testing. These results establish a practical, adaptive framework for automated cable tension warning, extending conventional monitoring into proactive decision-making support for bridge management.
- 22 May 2026
- Published by Elsevier in Engineering Structures
- Vol. 362
Modal coupling and vibration-based identification of individual tensions in interlinked parallel stay cables
- 22 May 2026
- Published by Elsevier in Engineering Structures
- Vol. 362
Modal coupling and vibration-based identification of individual tensions in interlinked parallel stay cables
- 11 March 2026
- Published by Elsevier in Measurement
- Vol. 273
Digital image-based estimation of stay cable tension through gravitational sag measurement
- 11 March 2026
- Published by Elsevier in Measurement
- Vol. 273
Digital image-based estimation of stay cable tension through gravitational sag measurement
- 1 October 2025
- Published by Springer Nature
- p. 832-842
Development of a Convenient Tension Estimation Method for Linked Parallel Stay Cables Based on Multiple Vibration Measurements
- 1 October 2025
- Published by Springer Nature
- p. 832-842
Development of a Convenient Tension Estimation Method for Linked Parallel Stay Cables Based on Multiple Vibration Measurements
- Open Access
- 11 June 2025
- Published by Elsevier in Results in Engineering
- Vol. 27
Experimental investigations and field applications of a tension estimation method for two linked suspenders using only local vibration measurements
- 11 June 2025
- Published by Elsevier in Results in Engineering
- Vol. 27
Open AccessExperimental investigations and field applications of a tension estimation method for two linked suspenders using only local vibration measurements
- Open Access
- 14 October 2023
- Published by Wiley in Structural Control and Health Monitoring
- Vol. 2023, 1-21
Extensive Field Validations and Corresponding Numerical Investigations for a Cable Tension Estimation Method Based on Local Vibration Measurements
To assess the applicability of the tension estimation method using local vibration measurements in a thorough manner, this research work is devoted to systematically investigate the appropriate covering ranges of measurements for different cables. Four cables of three cable-stayed bridges are deliberately chosen to cover a wide range of the cable slenderness parameter. Numerical analyses with finite element models and field validations with real measurements are conducted for these stay cables to demonstrate that the covering range of local measurements can be undoubtedly reduced with the increase of cable slenderness. For a relatively short cable with the slenderness parameter at the order of 4000, the adoption of 1/3 coverage is sufficient to keep a high-level accuracy with at most 1% of error. Besides, 1/4, 1/6, and 1/7 coverages are found adequate to maintain the same level of accuracy for longer cables with the slenderness parameter at the orders of 30000, 55000, and 80000, respectively. With the solid validations presented in the current study, the covering range of the cable for this simplified method employing local vibration measurements can be confidently reduced to greatly alleviate the expense and hard work of sensor installation near the high end.
- 14 October 2023
- Published by Wiley in Structural Control and Health Monitoring
- Vol. 2023, 1-21
Open AccessExtensive Field Validations and Corresponding Numerical Investigations for a Cable Tension Estimation Method Based on Local Vibration Measurements
To assess the applicability of the tension estimation method using local vibration measurements in a thorough manner, this research work is devoted to systematically investigate the appropriate covering ranges of measurements for different cables. Four cables of three cable-stayed bridges are deliberately chosen to cover a wide range of the cable slenderness parameter. Numerical analyses with finite element models and field validations with real measurements are conducted for these stay cables to demonstrate that the covering range of local measurements can be undoubtedly reduced with the increase of cable slenderness. For a relatively short cable with the slenderness parameter at the order of 4000, the adoption of 1/3 coverage is sufficient to keep a high-level accuracy with at most 1% of error. Besides, 1/4, 1/6, and 1/7 coverages are found adequate to maintain the same level of accuracy for longer cables with the slenderness parameter at the orders of 30000, 55000, and 80000, respectively. With the solid validations presented in the current study, the covering range of the cable for this simplified method employing local vibration measurements can be confidently reduced to greatly alleviate the expense and hard work of sensor installation near the high end.