PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
May 11, 2026Computer-Aided Civil and Infrastructure Engineering1 citationsOpen Access

Generative Digital Twin for Deformation Field Reconstruction in Rockfill Dams

Generative digital twin for deformation field reconstruction of high rockfill dams based on diffusion models

View Full Paper
Ask AI
Bookmark
Share

Authors

ZAZhitao AiGMGang MaTQTongming Qu

Discussion

Loading...

Member takes

Overview

Randomized trial develops a generative digital twin model to enhance deformation monitoring in rockfill dams, indicating increased accuracy and efficiency.

Key Points

  • This study aims to enhance the deformation perception of rockfill dams using a generative digital twin model based on diffusion techniques.
  • Developed a generative digital twin model for monitoring deformation in rockfill dams using Denoising Diffusion Probabilistic Model.
  • Introduced an active sampling strategy to improve training efficiency and diversity.
  • Employed a monitoring-consistency gradient guidance strategy to embed sparse data into the reverse diffusion process.
  • Achieved a point-wise relative reconstruction accuracy of 90.7% compared to monitoring data.
  • Model inference time for generating a single deformation field is approximately 30 seconds.
  • Demonstrated viability of the model in the tallest rockfill dam, Lianghekou Dam.

Cite This Study

Ai et al. (2026) studied this question.

synapsesocial.com/papers/6a0171983a9f334c28271b49https://doi.org/10.1016/j.cacaie.2026.100090
View Full Paper
Ask AI
Bookmark
Share