EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Interval Prediction Model of Deformation Behavior for Dam Safety during Long-Term Operation Using Bootstrap-GBDT

Erfeng Zhao, Yi Li, Jingmei Zhang, Zhangyin Li

Structural Control and Health Monitoring · 12.05.2023 · 2023 · 1-14

DOI: 10.1155/2023/6929861 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Structural Control and Health Monitoring, erschienen am 12.05.2023. Verfasst von Erfeng Zhao und 3 weiteren Autor:innen. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Science and Technology Project of Zhejiang Water Resources Department.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate and reliable prediction on structural deformation is a powerful means to evaluate the safety state of dams during long-term operation. This work is a contribution to the quantification of the cognitive and stochastic uncertainties in prediction models for the dam performance using measured time series, through constructing the prediction interval (PI) by an innovative model combined with the gradient boosting decision tree (GBDT) and the bootstrap method. The constructed PI, improved by the kernel density estimation (KDA), consists of an upper bound and a lower bound of the interval to provide a confidence level for dam deformation prediction. The bootstrap method combined with multiple GBDT is utilized to estimate the variance of the model bias, mainly derived from the cognitive uncertainties. The variance of random noise can be furtherly estimated through training the combined model to fit the residuals so as to indicate the stochastic uncertainties. The effectiveness of the newly proposed model is validated employing measured data of the Jinping Ι arch Dam. The results indicate that the methodology can obtain high-quality PIs and accurate prediction values and thus can provide strong support for better appraising dam performance during long-term operation.

Förderung

  • National Natural Science Foundation of China
  • State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering
  • Science and Technology Project of Zhejiang Water Resources Department

Zitieren

APA

Zhao, E., Li, Y., Zhang, J., & Li, Z. (2023). Interval Prediction Model of Deformation Behavior for Dam Safety during Long-Term Operation Using Bootstrap-GBDT. Structural Control and Health Monitoring, 2023, 1-14. https://doi.org/10.1155/2023/6929861

Vancouver

Zhao E, Li Y, Zhang J, Li Z. Interval Prediction Model of Deformation Behavior for Dam Safety during Long-Term Operation Using Bootstrap-GBDT. Structural Control and Health Monitoring. 2023;2023:1-14. doi:10.1155/2023/6929861

BibTeX
@article{Zhao2023,
  author = {Zhao, Erfeng and Li, Yi and Zhang, Jingmei and Li, Zhangyin},
  title = {Interval Prediction Model of Deformation Behavior for Dam Safety during Long-Term Operation Using Bootstrap-GBDT},
  journal = {Structural Control and Health Monitoring},
  year = {2023},
  volume = {2023},
  pages = {1--14},
  doi = {10.1155/2023/6929861}
}
Kostenlos starten →