Damage Identification of Bridge Structures Considering Temperature Variations-Based SVM and MFO

Civil structures are affected by some environmental factors such as traffic, ambient temperature, and noises. The change of structural dynamic characteristics arising from these factors may cover up those coming from structural damage, which makes the evaluation of structural health conditions based on vibration data more difficult. To overcome this difficulty, a novel method based on the support vector machine (SVM) and moth-flame optimization (MFO) is proposed to identify the damage of structures considering temperature variations. First of all, SVM is adopted to determine temperature variations and possible damage locations using the first six natural frequencies, and MFO is exploited to locate and quantify the damage accurately through the objective function constructed with a frequency-based multiple damage location assurance criterion (FMDLAC) and modal strain energy-based index (MSEBI). The combination of MFO and SVM can promote the efficiency of damage identification and accurately analyze environmental effects, which is a creative method with good robustness to solve the issue of damage identification considering environmental factors. To verify the effectiveness of the proposed method, a numerical simply-supported beam example considering temperature variations, as well as random noise, is investigated, and the optimal parameters for the method are acquired. Finally, a practical engineering example, the I-40 Bridge, is adopted to confirm the feasibility of the method further. The results demonstrate that the proposed approach is of a good optimization performance and can identify the damage of large complex structures considering temperature variations, which is of great practical application value.

• Record URL:
  https://doi.org/10.1061/(ASCE)AS.1943-5525.0001225
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/08931321
• Supplemental Notes:
  • © 2020 American Society of Civil Engineers.
• Authors:
  • Huang, Minshui
  • Lei, Yongzhi
  • Li, Xifan
  • Gu, Jianfeng
• Publication Date: 2021-3

Language

• English

Media Info

• Media Type: Web
• Features: References;
• Pagination: 04020113
• Serial:
  • Journal of Aerospace Engineering
  • Volume: 34
  • Issue Number: 2
  • Publisher: American Society of Civil Engineers
  • ISSN: 0893-1321
  • EISSN: 1943-5525
  • Serial URL: http://ascelibrary.org/journal/jaeeez

Subject/Index Terms

• TRT Terms: Bridge engineering; Bridges; Detection and identification; Machine learning; Structural health monitoring; Temperature
• Subject Areas: Bridges and other structures; Highways; Maintenance and Preservation;

Filing Info

• Accession Number: 01759150
• Record Type: Publication
• Files: TRIS, ASCE
• Created Date: Nov 24 2020 3:01PM