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 30, Number 2, August 2022 , pages 209-219 DOI: https://doi.org/10.12989/sss.2022.30.2.209 |
|
|
|
A LSTM-based method for intelligent prediction on mechanical response of precast nodular piles |
||
Xiao-Xiao Chen, Chang-Sheng Zhan and Sheng-Liang Lu
|
||
| Abstract | ||
| The determination for bearing capacity of precast nodular piles is conventionally time-consuming and high-cost by using numerous experiments and empirical methods. This study proposes an intelligent method to evaluate the bearing capacity and shaft resistance of the nodular piles with high efficiency based on long short-term memory (LSTM) approach. A series of field tests are first designed to measure the axial force, shaft resistance and displacement of the combined nodular piles under different loadings, in comparison with the single pre-stressed high-strength concrete piles. The test results confirm that the combined nodular piles could provide larger ultimate bearing capacity (more than 100%) than the single pre-stressed highstrength concrete piles. Both the LSTM-based method and empirical methods are used to calculate the shift resistance of the combined nodular piles. The results show that the LSTM-based method has a high-precision estimation on shaft resistance, not only for the ultimate load but also for the working load. | ||
| Key Words | ||
| combined nodular piles; field test; LSTM-based method; shaft resistance | ||
| Address | ||
| (1) Xiao-Xiao Chen: The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, 325016, China; (2) Chang-Sheng Zhan: Wenzhou Ecological Park Development and Construction Investment Group Co., Ltd, Wenzhou, 325207, China; (3) Sheng-Liang Lu: School of Civil Engineering and Architecture, Wenzhou Polytechnic, Wenzhou, 325000, China. | ||