Research on Foundation Pit Engineering Multi-source Data Fusion Method Based on GA-LSTM - A Case Study

DOI

10.2991/978-94-6463-726-7_55How to use a DOI?

Keywords

Deep foundation pit engineering; Horizontal deformation; Data fusion model; Genetic algorithm; Neural network prediction model

Abstract

To improve the deformation prediction accuracy of foundation pit engineering, a centralized and distributed fusion model for multi-source monitoring data is proposed based on the correlation analysis of multi-source monitoring data. Furthermore, genetic algorithm (GA) is used to optimize Long Short-Term Memory (LSTM) model parameters, including the number of LSTM layers, the number of hidden layers, the number of units in the hidden layer, and dropout. A multi-source data fusion deformation prediction model for foundation pit engineering based on GA-LSTM is constructed. Taking the deep foundation pit engineering monitoring data in Beijing Convention and Exhibition Center as an example, the effectiveness of the multi-source heterogeneous data fusion model and the LSTM neural network optimized by genetic algorithm is studied and verified. The results show that the multi-source fusion model can make reasonable use of the complementarity between the multi-source heterogeneous data, eliminate the fuzziness, uncertainty and randomness of the multi-source data. The MAE, RMSE, and R² of the GA-LSTM-distributed prediction model for foundation pit deformation prediction are 0.13, 0.13, and 0.97, respectively, with an average relative error of only 5.9%. The distributed data fusion based on GA-LSTM exhibits higher prediction accuracy and stability.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Hao Guo
AU  - Zhixue Wu
AU  - Peng Zuo
AU  - Ke Liu
AU  - Lei Zhang
AU  - Jiefei An
PY  - 2025
DA  - 2025/06/13
TI  - Research on Foundation Pit Engineering Multi-source Data Fusion Method Based on GA-LSTM - A Case Study
BT  - Proceedings of the 2024 6th International Conference on Hydraulic, Civil and Construction Engineering (HCCE 2024)
PB  - Atlantis Press
SP  - 549
EP  - 559
SN  - 2589-4943
UR  - https://doi.org/10.2991/978-94-6463-726-7_55
DO  - 10.2991/978-94-6463-726-7_55
ID  - Guo2025
ER  -