Slope displacement prediction based on multisource domain transfer learning for insufficient sample data
Applied Geophysics - Trang 1-9
Tóm tắt
Accurate displacement prediction is critical for the early warning of landslides. The complexity of the coupling relationship between multiple influencing factors and displacement makes the accurate prediction of displacement difficult. Moreover, in engineering practice, insufficient monitoring data limit the performance of prediction models. To alleviate this problem, a displacement prediction method based on multisource domain transfer learning, which helps accurately predict data in the target domain through the knowledge of one or more source domains, is proposed. First, an optimized variational mode decomposition model based on the minimum sample entropy is used to decompose the cumulative displacement into the trend, periodic, and stochastic components. The trend component is predicted by an autoregressive model, and the periodic component is predicted by the long short-term memory. For the stochastic component, because it is affected by uncertainties, it is predicted by a combination of a Wasserstein generative adversarial network and multisource domain transfer learning for improved prediction accuracy. Considering a real mine slope as a case study, the proposed prediction method was validated. Therefore, this study provides new insights that can be applied to scenarios lacking sample data.
Tài liệu tham khảo
Arjovsky, M., Chintala, S., and Bottou, L., 2017, Wasserstein generative adversarial networks: Proceedings of the 34nd International Conference on Machine Learning, 214–223.
citation_journal_title=Applied Geophysics; citation_title=Seismic velocity inversion based on CNN-LSTM fusion deep neural network; citation_author=W Cao, XB Guo, F Tian; citation_volume=18; citation_issue=4; citation_publication_date=2021; citation_pages=499-514; citation_doi=10.1007/s11770-021-0913-3; citation_id=CR2
citation_journal_title=Statistics & Probability Letters; citation_title=A recursive algorithm for solving the spatial YuleWalker equations of causal spatial AR models; citation_author=BS Choi; citation_volume=33; citation_issue=3; citation_publication_date=1997; citation_pages=241-251; citation_doi=10.1016/S0167-7152(96)00133-2; citation_id=CR3
citation_journal_title=IEEE Transactions on Signal Processing; citation_title=Variational Mode Decomposition; citation_author=K Dragomiretskiy, D Zosso; citation_volume=62; citation_issue=3; citation_publication_date=2014; citation_pages=531-544; citation_doi=10.1109/TSP.2013.2288675; citation_id=CR4
citation_journal_title=Landslides; citation_title=Landslide displacement prediction based on variational mode decomposition and WA-GWO-BP model; citation_author=ZZ Guo, LX Chen, L Gui, D Du, KL Yin; citation_volume=17; citation_issue=3; citation_publication_date=2020; citation_pages=567-583; citation_doi=10.1007/s10346-019-01314-4; citation_id=CR5
citation_journal_title=Earth-Science Reviews; citation_title=Forecasting the time of failure of landslides at slopescale: A literature review; citation_author=E Intrieri, T Carla, G Gigli; citation_volume=193; citation_publication_date=2019; citation_pages=333-349; citation_doi=10.1016/j.earscirev.2019.03.019; citation_id=CR6
citation_journal_title=IEEE Transactions on Neural Networks and Learning Systems; citation_title=Multi-Source Contribution Learning for Domain Adaptation; citation_author=K Li, J Lu, H Zuo; citation_volume=33; citation_issue=10; citation_publication_date=2022; citation_pages=5293-5307; citation_doi=10.1109/TNNLS.2021.3069982; citation_id=CR7
citation_journal_title=Rock and Soil Mechanics; citation_title=Research on similarity evaluation method of landslide monitoring points based on motion-angle-difference; citation_author=Y Liu, S Feng, Z Qin; citation_volume=298; citation_issue=01; citation_publication_date=2019; citation_pages=295-303; citation_id=CR8
citation_journal_title=Engineering Geology; citation_title=Landslide displacement prediction based on multi-source data fusion and sensitivity states; citation_author=Y Liu, C Xu, B Huang, XW Ren, CQ Liu, BD Hu; citation_volume=271; citation_publication_date=2020; citation_pages=105608-105608; citation_doi=10.1016/j.enggeo.2020.105608; citation_id=CR9
citation_journal_title=Engineering Geology; citation_title=A multi-feature fusion transfer learning method for displacement prediction of rainfall reservoir-induced landslide with step-like deformation characteristics; citation_author=J Long, C Li, Y Liu; citation_volume=297; citation_publication_date=2022; citation_pages=106494; citation_doi=10.1016/j.enggeo.2021.106494; citation_id=CR10
citation_journal_title=Computers and Geosciences; citation_title=Displacement prediction of Baijiabao landslide based on empirical mode decomposition and long short-term memory neural network in Three Gorges area, China; citation_author=SL Xu, RQ Niu; citation_volume=111; citation_publication_date=2018; citation_pages=87-96; citation_doi=10.1016/j.cageo.2017.10.013; citation_id=CR11
citation_journal_title=IEEE Transactions on Knowledge and Data Engineering; citation_title=A Survey on Transfer Learning; citation_author=SJ Pan, Y Qiang; citation_volume=22; citation_issue=10; citation_publication_date=2010; citation_pages=1345-1359; citation_doi=10.1109/TKDE.2009.191; citation_id=CR12
citation_journal_title=Mechanical Systems and Signal Processing; citation_title=Multisource domain factorization network for cross-domain fault diagnosis of rotating machinery: An unsupervised multisource domain adaptation method; citation_author=YW Shi, AD Deng, X Ding, S Zhang, S Xu; citation_volume=164; citation_publication_date=2022; citation_pages=108219; citation_doi=10.1016/j.ymssp.2021.108219; citation_id=CR13
citation_journal_title=Applied Geophysics; citation_title=Research on fault recognition method combining 3D Res-UNet and knowledge distillation; citation_author=J Wang, JH Zhang, JL Zhang, FM Lu, RG Meng; citation_volume=18; citation_issue=02; citation_publication_date=2021; citation_pages=199-212+274; citation_doi=10.1007/s11770-021-0894-2; citation_id=CR14
citation_journal_title=Journal of Service Science and Management; citation_title=Prediction Method of Deep Horizontal Displacement of Slope Soil Based on Damped Holt-Winters Model; citation_author=K Yan, JY Wu, YQ Zhang, L Yang, YX Zhang; citation_volume=12; citation_issue=3; citation_publication_date=2019; citation_pages=391-406; citation_doi=10.4236/jssm.2019.123027; citation_id=CR15
Yang, J., Yan, R., and Hauptmann, A.G., 2007, Cross-domain video concept detection using adaptive svms: In Proceedings of the 15th ACM international conference on Multimedia, 188–197.