Deepar-Based Ground Subsidence Prediction Method

DOI

10.2991/978-94-6463-005-3_77How to use a DOI?

Keywords

Ground subsidence; Soft-DTW; Kmedoids; DeepAR

Abstract

Ground subsidence near transmission lines is frequent, and there is a certain safety hazard for the normal operation of transmission lines. The existing deterministic models for ground settlement prediction are complicated to apply and require specified data parameters, which are difficult to obtain; the time series prediction based on single point historical observation data has problems such as lack of data volume. In this paper, we propose a model that uses smoothed formulation of DTW (Soft-DTW) to measure the similarity of ground settlement time series and combines Kmedoids for clustering, and then uses Autoregressive recurrent neural network (DeepAR) to build a prediction model for the clustered data. It achieves a unified prediction model for multiple observation points, and is simple to apply, reducing the requirement for the amount of historical data from a single observation point. The experimental results based on the subsidence data of Qujing Sentinel observation show that the accuracy of its subsidence prediction trend established by DeepAR has been improved more obviously after the classification of Kmedoids clustering method based on Soft-DTW.

Copyright

© 2023 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  - Tianyu Li
AU  - Feng Xiao
AU  - Jiaying Li
AU  - Jiaqing Zhang
PY  - 2022
DA  - 2022/11/10
TI  - Deepar-Based Ground Subsidence Prediction Method
BT  - Proceedings of the 2022 3rd International Conference on E-commerce and Internet Technology (ECIT 2022)
PB  - Atlantis Press
SP  - 760
EP  - 771
SN  - 2589-4943
UR  - https://doi.org/10.2991/978-94-6463-005-3_77
DO  - 10.2991/978-94-6463-005-3_77
ID  - Li2022
ER  -