Deep neural network‐based regional seismic loss assessment considering correlation between EDP residuals of building structures
Bibliographic record
Abstract
Abstract Regional seismic loss assessment is essential for developing an emergency response plan in the event of an earthquake, which can reduce casualties and socioeconomic losses in an urban community. The uncertainties and correlations of structures’ engineering demand parameters (EDP) should be adequately considered to evaluate the community‐level seismic risk. Recently, the authors proposed an incremental dynamic analysis‐based method and regression‐based models to estimate the variances and correlations of residuals in EDP termed “EDP residuals.” The quantified uncertainties of the EDP residuals facilitate the accurate evaluation of the regional seismic performance. Still, the computational cost required in the estimation process makes its application a challenge. This study proposes two frameworks for regional seismic loss assessment based on deep neural networks (DNNs) to extend the applicability of EDP residual estimation and improve its accuracy. The first framework estimates the EDP residuals of buildings by combining the EDP residuals of various single‐degree‐of‐freedom (SDOF) systems through the modal combination rules. Three DNN models are constructed to predict the EDP residuals of SDOF systems. The second framework predicts the EDP residuals of buildings directly using two DNN models. The proposed frameworks are verified by numerical examples of regional seismic loss assessment, for which time history‐based “exact” solutions exist. The supporting source code, data, and trained models are available for download at https://github.com/TyongKim/EDP_residual . Highlights The importance of considering EDP residual correlation in seismic system reliability analysis is demonstrated. Two DNN‐based frameworks are developed to estimate EDP residuals of building structures. Modal combination rule is employed to utilize EDP residuals of SDOF systems representing structural modes. DNN models are constructed to predict EDP residuals of SDOF and MDOF systems. Accuracy and applicability of DNN‐based frameworks are successfully demonstrated by example of regional seismic loss assessment.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".