The neural network model to solve the pre-consolidation stress
- DOI
- 10.2991/gcmce-17.2017.57How to use a DOI?
- Keywords
- pre-consolidation stress, BP artificial neural network, e-p curves, MATLAB
- Abstract
Pre-consolidation stress is an indicator representing stress history of soil and an important parameter which reflects the deformation characteristics of soil. Finding an accurate and easy way to solve pre-consolidation stress is of great significant in the study of engineering construction. Based on the previous studies on solutions of the pre-consolidation stress and analysis of the artificial neural network theory, a new approach - the BP neural network model to solve pre-consolidation stress is proposed. Based on the platform of MATLAB, the BP neural network model is established and trained by random data samples from compression tests. By operating three different algorithms in computing analysis, the L-M algorithm is identified as the optimized to be applied in the model. Using the neural network model with training completion, the output port can export rapid and accurate inversions of the predicting pre-consoilidation stress. The forecasting errors are greatly reduced compared to the Casagrande method and the numerical plate method according to error analysis of these three methods. Therefore, the BP artificial neural network model to solve the pre-consolidation stress is proved to have a good feasibility and a promotional value in real engineering.
- Copyright
- © 2017, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Ran An AU - Lingwei Kong AU - Chengsheng Li PY - 2017/06 DA - 2017/06 TI - The neural network model to solve the pre-consolidation stress BT - Proceedings of the 2017 Global Conference on Mechanics and Civil Engineering (GCMCE 2017) PB - Atlantis Press SP - 322 EP - 329 SN - 2352-5401 UR - https://doi.org/10.2991/gcmce-17.2017.57 DO - 10.2991/gcmce-17.2017.57 ID - An2017/06 ER -