Buy article PDF
The purchased file will be sent to you
via email after the payment is completed.
US$ 35
|
Smart Structures and Systems Volume 34, Number 5, November 2024 , pages 323-333 DOI: https://doi.org/10.12989/sss.2024.34.5.323 |
|
|
|
Prediction of construction alignment for large-span bridges based on mean value theorem expansion response surface and neural network surrogate model |
||
Xingwang Sheng, Xu Song, Weiqi Zheng, Huanzhong Sun and Yonghong Yang
|
||
| Abstract | ||
| As the span increases, the difficulty of bridge construction control continuously escalates. Accurate construction control effectively ensures that bridges maintain a reasonable stress state, proper alignment, and track smoothness. This work innovatively integrates the Mean Value Theorem Expansion Response Surface method with a Neural Network Surrogate Model to precisely identify key parameters during the construction process, achieving high-accuracy predictions of construction alignment for large-span bridges. Initially, the Response Surface-Monte Carlo method is used for the sensitivity analysis of the main construction parameters. Subsequently, a parameter identification model is established to identify and correct key parameters affecting alignment and to refine the finite element model. Based on the adjusted model, sample data are collected to create an alignment prediction network model, which predicts alignment deviations for subsequent beam segments in construction, achieving high-precision reliability assessment of bridge construction alignment. The applications of case project demonstrate that the proposed methods for structural parameter identification and alignment prediction significantly enhance the precision of alignment forecasts. Characterized by the simplicity and high accuracy of the proposed method, it can offer a novel, efficient approach for alignment control under complex construction conditions. | ||
| Key Words | ||
| alignment prediction; construction control; large-span bridge; neural network surrogate model; response surface method; sensitivity analysis | ||
| Address | ||
| (1) Xingwang Sheng, Xu Song, Weiqi Zheng, Huanzhong Sun, Yonghong Yang: School of Civil Engineering, Central South University, Changsha, Hunan 410075, China; (2) Weiqi Zheng: National Engineering Research Center for High-speed Railway Construction Technology, Changsha, Hunan 410075, China; (3) Huanzhong Sun: China Railway 14th Bureau Second Engineering Co., LTD., Tai'an, Shandong, China 271000, China (4) Yonghong Yang: Shanghai-Hangzhou Railway Passenger Dedicated Line Co., LTD., Shanghai, China 200040, China. | ||