Search

Your search keyword '"Hajjar, Jerome F."' showing total 318 results
Start Over Author "Hajjar, Jerome F."
318 results on '"Hajjar, Jerome F."'

1. Vibration Testing of an All-Steel Modular Floor Assembly

Author

Avci, Onur, Rifai, Sahabeddin, Abla, Feras, Opie, Ben, Eatherton, Matthew, Schafer, Benjamin, Easterling, W. Samuel, Hajjar, Jerome F., Mouras, Joshua, Klemencic, Ron, Zimmerman, Kristin B., Series Editor, Whelan, Matthew, editor, Harvey, P. Scott, editor, and Moreu, Fernando, editor

Published
2025
Full Text
View/download PDF

2. Physics-Informed Machine Learning for Seismic Response Prediction OF Nonlinear Steel Moment Resisting Frame Structures

Author

Bond, R. Bailey, Ren, Pu, Hajjar, Jerome F., and Sun, Hao

Subjects
Physics - Applied Physics, Computer Science - Machine Learning
Abstract

There is growing interest in using machine learning (ML) methods for structural metamodeling due to the substantial computational cost of traditional simulations. Purely data-driven strategies often face limitations in model robustness, interpretability, and dependency on extensive data. To address these challenges, this paper introduces a novel physics-informed machine learning (PiML) method that integrates scientific principles and physical laws into deep neural networks to model seismic responses of nonlinear structures. The approach constrains the ML model's solution space within known physical bounds through three main features: dimensionality reduction via combined model order reduction and wavelet analysis, long short-term memory (LSTM) networks, and Newton's second law. Dimensionality reduction addresses structural systems' redundancy and boosts efficiency while extracting essential features through wavelet analysis. LSTM networks capture temporal dependencies for accurate time-series predictions. Manipulating the equation of motion helps learn system nonlinearities and confines solutions within physically interpretable results. These attributes allow for model training with sparse data, enhancing accuracy, interpretability, and robustness. Furthermore, a dataset of archetype steel moment resistant frames under seismic loading, available in the DesignSafe-CI Database [1], is considered for evaluation. The resulting metamodel handles complex data better than existing physics-guided LSTM models and outperforms other non-physics data-driven networks., Comment: 34 pages, 12 figures

Published
2024

3. An Unsupervised Machine Learning Approach for Ground-Motion Spectra Clustering and Selection

Author

Bond, R. Bailey, Ren, Pu, Hajjar, Jerome F., and Sun, Hao

Subjects
Physics - Geophysics, Computer Science - Artificial Intelligence, Computer Science - Machine Learning
Abstract

Clustering analysis of sequence data continues to address many applications in engineering design, aided with the rapid growth of machine learning in applied science. This paper presents an unsupervised machine learning algorithm to extract defining characteristics of earthquake ground-motion spectra, also called latent features, to aid in ground-motion selection (GMS). In this context, a latent feature is a low-dimensional machine-discovered spectral characteristic learned through nonlinear relationships of a neural network autoencoder. Machine discovered latent features can be combined with traditionally defined intensity measures and clustering can be performed to select a representative subgroup from a large ground-motion suite. The objective of efficient GMS is to choose characteristic records representative of what the structure will probabilistically experience in its lifetime. Three examples are presented to validate this approach, including the use of synthetic and field recorded ground-motion datasets. The presented deep embedding clustering of ground-motion spectra has three main advantages: 1. defining characteristics the represent the sparse spectral content of ground-motions are discovered efficiently through training of the autoencoder, 2. domain knowledge is incorporated into the machine learning framework with conditional variables in the deep embedding scheme, and 3. method exhibits excellent performance when compared to a benchmark seismic hazard analysis., Comment: 24 pages, 16 Figures

Published
2022
Full Text
View/download PDF

4. Physics-Guided Machine Learning for Structural Metamodeling and Fragility Analysis

Author

Bond, Robert Bailey, Ren, Pu, Sun, Hao, Hajjar, Jerome F., di Prisco, Marco, Series Editor, Chen, Sheng-Hong, Series Editor, Vayas, Ioannis, Series Editor, Kumar Shukla, Sanjay, Series Editor, Sharma, Anuj, Series Editor, Kumar, Nagesh, Series Editor, Wang, Chien Ming, Series Editor, Cui, Zhen-Dong, Series Editor, Mazzolani, Federico M., editor, Piluso, Vincenzo, editor, Nastri, Elide, editor, and Formisano, Antonio, editor

Published
2024
Full Text
View/download PDF

22. Climate Hazards and Critical Infrastructures Resilience

Author

Bhatia, Udit, Traylor, Allison, Moskos, Catherine, Blumenfeld, Laura, Bressler, Lindsey, Hall, Tyler, Heiss, Rachael, Clark, Kevin D., Deng, Nan, Kumar, Devashish, Kodra, Evan, Flynn, Stephen E., Koutsopoulos, Haris N., Hajjar, Jerome F., Ganguly, Auroop R., Shekhar, Shashi, editor, Xiong, Hui, editor, and Zhou, Xun, editor

Published
2017
Full Text
View/download PDF

26. An unsupervised machine learning approach for ground‐motion spectra clustering and selection.

Author

Bond, Robert Bailey, Ren, Pu, Hajjar, Jerome F., and Sun, Hao

Subjects
MACHINE learning, GROUND motion, DEEP learning, APPLIED sciences, ENGINEERING design, EARTHQUAKE hazard analysis, EARTHQUAKE resistant design
Abstract

Clustering analysis of sequence data continues to address many applications in engineering design, aided with the rapid growth of machine learning in applied science. This paper presents an unsupervised machine learning algorithm to extract defining characteristics of earthquake ground‐motion spectra, also called latent features, to aid in ground‐motion selection (GMS). In this context, a latent feature is a low‐dimensional machine‐discovered spectral characteristic learned through nonlinear relationships of a neural network autoencoder. Machine discovered latent features can be combined with traditionally defined intensity measures and clustering can be performed to select a representative subgroup from a large ground‐motion suite. The objective of efficient GMS is to choose characteristic records representative of what the structure will probabilistically experience in its lifetime. Three examples are presented to validate this approach, including the use of synthetic and field recorded ground‐motion datasets. The presented deep embedding clustering of ground‐motion spectra has three main advantages: (1) defining characteristics that represent the sparse spectral content of ground motions are discovered efficiently through training of the autoencoder, (2) domain knowledge is incorporated into the machine learning framework with conditional variables in the deep embedding scheme, and (3) the method results in a ground‐motion subgroup that is more representative of the original ground‐motion suite compared to traditional GMS techniques. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

30. Future directions for broader impacts at the NSF

Author

Jacobson, Sheldon H., Hajjar, Jerome F., Johnson, Andrew, Moreno-Centeno, Erick, and Shen, Siqian

Subjects
United States. National Science Foundation -- Services, Research grants, Grants-in-aid -- Social aspects -- Economic aspects, Business, general, Science and technology, Social sciences
Abstract

On May 26-27, 2016, a group of 30 researchers and administrators in the areas of mechanical, industrial and civil engineering met to discuss a game plan to uncover how investments [...]

Published
2018

35. An Unsupervised Machine Learning Approach for Ground Motion Clustering and Selection

Author

Bond, R. Bailey, Ren, Pu, Hajjar, Jerome F., and Sun, Hao

Subjects
FOS: Computer and information sciences, Physics - Geophysics, Computer Science - Machine Learning, Artificial Intelligence (cs.AI), Computer Science - Artificial Intelligence, FOS: Physical sciences, Geophysics (physics.geo-ph), Machine Learning (cs.LG)
Abstract

Clustering analysis of sequence data continues to address many applications in engineering design, aided with the rapid growth of machine learning in applied science. This paper presents an unsupervised machine learning algorithm to extract defining characteristics of earthquake ground-motion records, also called latent features, to aid in ground-motion clustering and selection. In this context, a latent feature is a low dimensional machine-discovered spectral characteristic learned through nonlinear relationships of a neural network autoencoder. Clustering can be performed on the latent features and used to select a representative archetypal subgroup from a large ground-motion suite. The objective of efficient ground-motion selection is to choose records representative of what the structure will probabilistically experience in its lifetime. Three examples are presented to validate this approach, including a synthetic spectral dataset and spectra from field recorded ground-motion records. Deep embedding clustering of ground motion spectra improves on the results of static feature extraction, utilizing characteristics that represent the sparse spectral content of ground motions., 24 pages, 15 Figures

Published
2022

39. Design models for predicting shear resistance of studs in solid concrete slabs based on symbolic regression with genetic programming

Author

Degtyarev, Vitaliy V., Hicks, Stephen J., and Hajjar, Jerome F.

Subjects
TA, TJ
Abstract

Accurate design models for predicting the shear resistance of headed studs in solid concrete slabs are essential for obtaining economical and safe steel-concrete composite structures. In this study, symbolic regression with genetic programming (GPSR) was applied to experimental data to formulate new descriptive equations for predicting the shear resistance of studs in solid slabs using both normal and lightweight concrete. The obtained GPSR-based nominal resistance equations demonstrated good agreement with the test results. The equations indicate that the stud shear resistance is insensitive to the secant modulus of elasticity of concrete, which has been included in many international standards following the pioneering work of Ollgaard et al. In contrast, it increases when the stud height-to-diameter ratio increases, which is not reflected by the design models in the current international standards. The nominal resistance equations were subsequently refined for use in design from reliability analyses to ensure that the target reliability index required by the Eurocodes was achieved. Resistance factors for the developed equations were also determined following US design practice. The stud shear resistance predicted by the proposed models was compared with the predictions from 13 existing models. The accuracy of the developed models exceeds the accuracy of the existing equations. The proposed models produce predictions that can be used with confidence in design, while providing significantly higher stud resistances for certain combinations of variables than those computed with the existing equations given by many standards.\ud

Published
2022

43. Climate Hazards and Critical Infrastructures Resilience

Author

Bhatia, Udit, primary, Traylor, Allison, additional, Moskos, Catherine, additional, Blumenfeld, Laura, additional, Bressler, Lindsey, additional, Hall, Tyler, additional, Heiss, Rachel, additional, Clark, Kevin D., additional, Deng, Nan, additional, Kumar, Devashish, additional, Kodra, Evan, additional, Flynn, Stephen E., additional, Koutsopoulos, Haris N., additional, Ganguly, Auroop, additional, and Hajjar, Jerome F., additional

Published
2016
Full Text
View/download PDF

50. Mixed finite element for three-dimensional nonlinear dynamic analysis of rectangular concrete-filled steel tube beam-columns

Author

Tort, Cenk and Hajjar, Jerome F.

Subjects
Finite element method -- Research, Concrete -- Mechanical properties, Dynamic testing -- Methods, Science and technology
Abstract

A beam finite-element formulation following Euler-Bernoulli beam theory is presented for geometrically and materially nonlinear analysis of rectangular concrete-filled steel tube (RCFT) beam-columns. The formulation is geared for conducting transient dynamic analysis of composite steel/concrete frame structures. The element stiffness and internal forces were derived through adopting a mixed finite-element approach based on the Hellinger-Reissner variational principle. The load transfer between the steel and concrete constitutive materials was provided through steel and concrete interface via friction and interlocking. Six extra translational degrees-of-freedom (DOFs) were added to the conventional 12 DOF beam element to quantify the differential displacement between the two media. The formulation was verified for a range of geometrically nonlinear test problems and geometrically and materially nonlinear RCFT experimental test specimens from the literature. Strong correlation and convergence characteristics were achieved compared to the published results. DOI: 10.1061/(ASCE)EM.1943-7889.0000179 CE Database subject headings: Beams; Finite element method; Slip; Beam columns; Tubes; Dynamic analysis. Author keywords: Beam finite element; Mixed finite-element formulation; Corotational formulation; Slip; Composite beam-column; Concrete-filled steel tube.

Published
2010

Catalog

Books, media, physical & digital resources
See catalog results