Optimizing Bridge Cable Forces and Deck Positions with Digital Twins

Creative Commons License

Türer A.

Istanbul Bridge Conference (IBridge 2024) September 06 – 08 , İstanbul, Türkiye, 6 - 08 Eylül 2024, ss.1-10, (Tam Metin Bildiri)

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: İstanbul
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.1-10
  • Açık Arşiv Koleksiyonu: AVESİS Açık Erişim Koleksiyonu
  • Orta Doğu Teknik Üniversitesi Adresli: Evet

Özet

Cabled bridges such as cable stayed, suspension, and cabled arched are integral parts of modern infrastructure. The later stages of bridge construction is always a general hassle to achieve the desired cable forces and deck geometry under dead load. The predictions by the structural analysis, which may have been linear or nonlinear in nature, may not always turn out to be very close to the actual values on site due to some possible manufacturing or workmanship margins of error. Design engineers may also incorporate construction-stage analyses and soil structure interactions; however, despite careful planning and advanced analyses, the completed bridge may exhibit uneven cable forces and uneven deck deformations. While cable length adjustments may be possible within certain limits, the nonlinear and interdependent effects of cable lengths on cable forces and deck geometry often lead to unexpected complications. An optimal solution strategy using digital twins for achieving the desired bridge deck geometry and dead load cable forces simultaneously is developed. A vertical cabled hypothetical arch bridge is used to test the proposed approach, which is based on the principle that each cable’s length change affects the forces in all cables and vertical position of all deck points. A staged alteration and measurement procedure is proposed, where each cable's force and deck vertical positions are measured after each incremental changes in all cable’s constructed lengths. The solution involves iterative adjustments and measurements, with optimization and search algorithms to the changing responses and sometimes iteratively re-solving the digital twin to minimize deviations from the target cable forces and deck deflections. A FEM is not needed for this approach but actual field measurements are necessary. The results indicate that convergence achieved after a few rounds of modifications and measurements.