Predicting dynamic deformation of retaining structure by LSSVR-based time series method

Zhiwei Ji, Bing Wang, Su Ping Deng, Zhuhong You

Research output: Contribution to journalArticlepeer-review

48 Scopus citations

Abstract

In this paper, we propose a LSSVR-based time series method (LTSM) to predict dynamic lateral deformation of retaining structure and ground surface settlement in deep foundation pit engineering. After reconstructing phase space, time-varying lateral displacement of each observation point on retaining structure can be predicted using its historic unary time series data. And then, the ground settlement nearby the deep foundation pit can be predicted using all the observed values of lateral deformation which were collected at the same time from different depths on the retaining structure. The experimental results show that LTSM achieved a high accuracy when predicting the lateral deformation and ground settlement.

Original languageEnglish
Pages (from-to)165-172
Number of pages8
JournalNeurocomputing
Volume137
DOIs
StatePublished - 5 Aug 2014
Externally publishedYes

Keywords

  • Ground settlement
  • Lateral displacement
  • LSSVR
  • Phase space reconstruction
  • Time series

Fingerprint

Dive into the research topics of 'Predicting dynamic deformation of retaining structure by LSSVR-based time series method'. Together they form a unique fingerprint.

Cite this