MétaCan
Menu
Back to cohort
Record W4416343430 · doi:10.1061/jsendh.steng-14737

Seismic Response Estimation and Performance Assessment of Large-Scale Structures Using Deep-Learning Methods

2025· article· en· W4416343430 on OpenAlex
Jacob A. Morgan, Kalil Erazo, Oral Büyüköztürk

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

VenueJournal of Structural Engineering · 2025
Typearticle
Languageen
FieldEngineering
TopicStructural Health Monitoring Techniques
Canadian institutionsWSP (Canada)
Fundersnot available
KeywordsRobustness (evolution)Artificial neural networkNonlinear systemEarthquake shaking tableConvolutional neural networkContext (archaeology)Earthquake engineeringExperimental data

Abstract

fetched live from OpenAlex

Applications of machine learning (ML) in structural engineering and mechanics have increased substantially in recent years. Interest in ML stems from its potential to handle complex problems with increased accuracy and/or computational efficiency compared with traditional methods when a sufficient amount of data is available. This study explored the application of deep neural networks to seismic response and damage estimation for large-scale civil structures in the context of performance-based earthquake engineering. The objective was to experimentally validate the accuracy and efficiency of modified versions of the long short-term memory (LSTM) neural network and the physics-guided convolutional neural network (PhyCNN) in structures subjected to base excitations in linear and nonlinear response regimes. In particular, the robustness of the methods to predict the response and state of integrity of structures undergoing severe damage due to strong base motions when the methods are trained using limited low-amplitude linear input–output data was assessed. The performance and predicting capabilities of the methods were evaluated using experimental data from two full-scale case studies: a six-story building structure in San Bernardino, California, instrumented as part of the California Strong Motion Instrumentation Program (CSMIP), and a seven-story full-scale shear wall structure tested on the high-performance outdoor shake table at the University of California at San Diego. The results show that deep neural networks have limited robustness when applied to estimate response features outside the training domain, with a significant reduction in the prediction accuracy if the level of nonlinearity for prediction is higher than in the training data set.

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.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.287
Threshold uncertainty score0.752

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.001
Insufficient payload (model declined to judge)0.0000.000

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.

Opus teacher head0.011
GPT teacher head0.359
Teacher spread0.348 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it