Artificial Intelligence, Machine Learning, and Deep Learning in Structural Engineering: A Scientometrics Review of Trends and Best Practices

Statistische Auswertungen von Brückeneinsturzdaten zeigen, dass Massivbrücken signifikant seltener einstürzen als Brücken aus Stahl oder Holz. Da weltweit die Hauptursache für Brückeneinstürze Hochwasser und damit verbundene fluviale Prozesse wie Auskolkung, Muren etc. und Anpralle sind, liegt die Vermutung nahe, dass die hohe Eigenlast der Massivbrücken zu einem insgesamt robusteren Verhalten bei diesen Ereignissen führt. Im Rahmen dieses Beitrags soll geprüft werden, ob die IABSE‐Einsturzdatenbank diese These bestätigt und Hinweise auf weitere Ursachen identifiziert werden können. Dazu wird die IABSE‐Einsturzdatenbank mittels Verfahren der Künstlichen Intelligenz und des maschinellen Lernens (KI/ML) untersucht. Die KI/ML‐Analyse bestätigt die bisherige These allerdings nicht. Im Weiteren werden Ursachen für die Ablehnung der These, wie z. B. die Repräsentativität der Daten, diskutiert. Es wird eine Erweiterung der Datenbank für Ereignisse mit großen Einsturzzahlen empfohlen.

show abstract

“…B. [15–20]. Gleichwohl wurden in [21] die Grenzen der Anwendung von KI/ML im Bereich des Ingenieurwesens aufgezeigt.…”

Section: Einführung In Die Ki/ml‐verfahrenunclassified

“… Vereinfachte Darstellung des Random Forest Classifier nach [18] Simplified Visualization of the Random Forrest Classifier according to [18] …”

Section: Vorgehensweiseunclassified

KI/ML‐gestützte Auswertung und Interpretation der IABSE‐Brückeneinsturzdatenbank

Proske

Güner²,

Hingorani

et al. 2023

Beton und Stahlbetonbau

View full text Add to dashboard Cite

show abstract

“…With the rise of artificial intelligence (AI) in recent years, especially the advancement of deep learning developed from neural network, AI technology has been broadly used in the field of engineering structures [7][8][9][10], such as damage identification [11][12][13][14], model optimization [15][16], and performance prediction [17][18][19][20]. However, the neural networks remain a black box that cannot be explicitly described in a physical sense, and this purely data-driven method heavily relies on vast amounts of high-fidelity data.…”

Section: Main Textmentioning

confidence: 99%

Multi-Physics-informed deep learning approach for computational structural mechanics

Kong,

He,

et al. 2023

Preprint

View full text Add to dashboard Cite

Finite element method (FEM) has been popular for decades for structure analysis by solving partial differential equations (PDEs). Recently, physics-informed deep learning has emerged as a promising approach for solving PDEs and shows great advantages over FEM in the computational efficacy. However, it is still a challenge for the structural mechanics computation since it involves solving higher-order PDEs as the governing equations are fourth-order nonlinear PDEs. Therefore, we develop a novel multi-physics-informed neural network (m-PINN) framework via combining multiple neural networks (NNs) where each NN representing physics information such as geometrical compatibility, constitutive relation, and static equilibrium. The verification result demonstrates that m-PINN has the same accuracy as FEM for the computation of beam and shell structures. The proposed method has the potential to be a new paradigm in the structural mechanics computation as an alternative of FEM, and could serve as an intelligent computational module in digital twin system.

show abstract

“…In order to achieve the three steps above defined for damage identification, instead of using classical approaches based on modal tracking, [2][3][4] many researchers have turned to machine learning. 5 artificial neural networks (ANNs) undoubtedly deserve to be highlighted among the machine learning algorithms. ANNs are extremely versatile and can be applied to SHM problems through various architectures, such as fully connected networks, 6,7 autoencoders, 8,9 and, more recently, convolutional neural networks (CNNs).…”

Section: Introductionmentioning

confidence: 99%

Damage identification using convolutional neural networks from instantaneous displacement measurements via image processing

Resende,

Finotti,

Barbosa

et al. 2023

Structural Health Monitoring

View full text Add to dashboard Cite

This work investigates the effectiveness of using convolutional neural networks (CNNs) and instantaneous displacement measurements for damage identification in beams. The study involves subjecting laboratory beams to eight distinct damage scenarios and capturing the vertical positions of 60 points along the beam length during free-vibration tests using a high-speed camera. The data obtained was subsequently used to train a CNN in a supervised manner to estimate the level of damage at each point. Results showed that the CNN models were able to correctly localize and quantify the damage levels when trained on data from all damage scenarios. The soundness of the proposed methodology was demonstrated in a robustness assessment, where all eight damage scenarios were correctly identified even when two of them were excluded from the training dataset.

show abstract

Artificial Intelligence, Machine Learning, and Deep Learning in Structural Engineering: A Scientometrics Review of Trends and Best Practices

Cited by 91 publications

References 216 publications

KI/ML‐gestützte Auswertung und Interpretation der IABSE‐Brückeneinsturzdatenbank

KI/ML‐gestützte Auswertung und Interpretation der IABSE‐Brückeneinsturzdatenbank

Multi-Physics-informed deep learning approach for computational structural mechanics

Damage identification using convolutional neural networks from instantaneous displacement measurements via image processing

Contact Info

Product

Resources

About