Ke Du
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Ke Du has a total of 47 co-authors. Here are the top co-authors by number of shared publications:
Huan Luo
8 shared publicationsJingjiang Sun
Professor, Key Laboratory of Earthquake Engineering and Engineering Vibration, Institute of Engineering Mechanics, China Earthquake Administration, Harbin 150080, China
7 shared publicationsZehua Huang
4 shared publications- Jiawei Gao
Key Laboratory of Earthquake Engineering and Engineering Vibration, Institute of Engineering Mechanics, China Earthquake Administration, Harbin, China
3 shared publications Ruizhi Wen
3 shared publications- Wen Bai
Institute of Engineering Mechanics, China Earthquake Administration
3 shared publications
Latest Publications
- Top Cited Papers
- 12 March 2026
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 173
Discovering interpretable structural dynamic response using physics-informed Kolmogorov-Arnold network
- 12 March 2026
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 173
Top Cited PapersDiscovering interpretable structural dynamic response using physics-informed Kolmogorov-Arnold network
- 11 March 2026
- Published by Taylor & Francis in Journal of Earthquake Engineering
- Vol. 30 (12) , 1-38
A Damage-Dependent Hysteresis Model with Dynamic Pinching for Efficient and Accurate Simulation of RC Beam-Column Joint Shear
- 11 March 2026
- Published by Taylor & Francis in Journal of Earthquake Engineering
- Vol. 30 (12) , 1-38
A Damage-Dependent Hysteresis Model with Dynamic Pinching for Efficient and Accurate Simulation of RC Beam-Column Joint Shear
- Top Cited Papers
- 24 February 2026
- Published by Springer Nature in Bulletin of Earthquake Engineering
- Vol. 24 (6) , 3795-3817
A fourier-enhanced physics-informed Kolmogorov–Arnold network for multi-frequency seismic response analysis of structures
- 24 February 2026
- Published by Springer Nature in Bulletin of Earthquake Engineering
- Vol. 24 (6) , 3795-3817
Top Cited PapersA fourier-enhanced physics-informed Kolmogorov–Arnold network for multi-frequency seismic response analysis of structures
- Top Cited Papers
- 24 December 2025
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 166
A label-free physics informed neural network with hard constraints and Fourier features spectrally-enhanced for multi-frequency seismic structural dynamic response
- 24 December 2025
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 166
Top Cited PapersA label-free physics informed neural network with hard constraints and Fourier features spectrally-enhanced for multi-frequency seismic structural dynamic response
- 28 November 2025
- Published by Wiley in The Structural Design of Tall and Special Buildings
- Vol. 34 (17)
Finite Element Model Updating of a High‐Rise Building Based on Measured Seismic Response Data from the 2021 Yangbi Earthquake
Finite element model updating (FEMU) is essential for improving the accuracy of structural simulations, particularly in seismic analysis of high‐rise buildings. This study focuses on updating the finite element model of a high‐rise structure based on seismic response data recorded during the M6.4 Yangbi earthquake in Dali, Yunnan Province, China, on May 21, 2021. A structural seismic response monitoring array was deployed in the building, capturing valuable data that reflects the building's dynamic characteristics under actual seismic loads. To enhance the accuracy of the model update, intelligent optimization algorithms were applied, including particle swarm optimization (PSO), ant colony optimization (ACO), fish swarm algorithm (FSA), and artificial bee colony (ABC). Additionally, a response surface model was constructed to reduce computational costs and improve efficiency. The study further explores the influence of key structural parameters—such as elastic modulus and moment of inertia—on model performance through sensitivity analysis and response surface modeling. The results demonstrate that the intelligent optimization algorithms successfully improved the accuracy of the finite element model, with average errors below 1.5% for all tested algorithms. Evolutionary algorithms, particularly PSO and ACO, showed superior performance in reducing discrepancies between simulated and observed data, enhancing the model's predictive capability. The findings underscore the significance of integrating seismic response data with intelligent optimization techniques for robust structural model updating, contributing to more reliable performance‐based earthquake engineering (PBEE) practices.
- 28 November 2025
- Published by Wiley in The Structural Design of Tall and Special Buildings
- Vol. 34 (17)
Finite Element Model Updating of a High‐Rise Building Based on Measured Seismic Response Data from the 2021 Yangbi Earthquake
Finite element model updating (FEMU) is essential for improving the accuracy of structural simulations, particularly in seismic analysis of high‐rise buildings. This study focuses on updating the finite element model of a high‐rise structure based on seismic response data recorded during the M6.4 Yangbi earthquake in Dali, Yunnan Province, China, on May 21, 2021. A structural seismic response monitoring array was deployed in the building, capturing valuable data that reflects the building's dynamic characteristics under actual seismic loads. To enhance the accuracy of the model update, intelligent optimization algorithms were applied, including particle swarm optimization (PSO), ant colony optimization (ACO), fish swarm algorithm (FSA), and artificial bee colony (ABC). Additionally, a response surface model was constructed to reduce computational costs and improve efficiency. The study further explores the influence of key structural parameters—such as elastic modulus and moment of inertia—on model performance through sensitivity analysis and response surface modeling. The results demonstrate that the intelligent optimization algorithms successfully improved the accuracy of the finite element model, with average errors below 1.5% for all tested algorithms. Evolutionary algorithms, particularly PSO and ACO, showed superior performance in reducing discrepancies between simulated and observed data, enhancing the model's predictive capability. The findings underscore the significance of integrating seismic response data with intelligent optimization techniques for robust structural model updating, contributing to more reliable performance‐based earthquake engineering (PBEE) practices.
- 19 September 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Structural Stability and Dynamics
Deep Learning-Based Prediction of Nonlinear Seismic Response of Spherical Shell Structures
To overcome the inherent trade-off between computational efficiency and accuracy in traditional numerical methods, this study develops a series of deep learning models for accurately predicting the nonlinear seismic responses of spherical shell structures. The proposed models are based on four different architectures: bidirectional long short-term memory networks (BiLSTM), bidirectional gated recurrent units (BiGRU), and their attention-enhanced variants (AttLSTM, AttGRU), systematically investigating the influence of different deep learning architectures on structural dynamic response prediction performance. Based on the NGA West2 database, an input dataset incorporating multi-dimensional seismic excitations was constructed, and benchmark dynamic response solutions were generated using a finite element model for the training of deep learning models. Model performance was evaluated using both mean squared error and coefficient of determination metrics, and validated on independent test sets as well as representative earthquake records. Results demonstrate that compared with the finite element method (FEM), the deep learning models achieved a significant improvement in computational efficiency for predicting the nonlinear seismic responses of spherical shell structures. Among them, the BiGRU model, benefiting from its simplified gating mechanism and higher parameter efficiency, exhibited the best performance in predicting seismic responses in the [ ]- and [ ]-directions, with its predictions closely matching the FEM benchmark solutions. Additionally, the BiLSTM model, due to its more complex memory unit structure, showed superior directional balance when handling multi-dimensional inputs. Notably, although the attention mechanism enhanced the models’ capability to capture local nonlinear features, under the current conditions of limited dataset size and long-time sequences, its effect on improving the base models was not significant and even exhibited some degree of performance suppression. This study provides a methodological basis for selecting appropriate deep learning models for seismic safety assessments of complex spatial structures and offers in-depth insights into the applicability of different architectures under specific engineering scenarios.
- 19 September 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Structural Stability and Dynamics
Deep Learning-Based Prediction of Nonlinear Seismic Response of Spherical Shell Structures
To overcome the inherent trade-off between computational efficiency and accuracy in traditional numerical methods, this study develops a series of deep learning models for accurately predicting the nonlinear seismic responses of spherical shell structures. The proposed models are based on four different architectures: bidirectional long short-term memory networks (BiLSTM), bidirectional gated recurrent units (BiGRU), and their attention-enhanced variants (AttLSTM, AttGRU), systematically investigating the influence of different deep learning architectures on structural dynamic response prediction performance. Based on the NGA West2 database, an input dataset incorporating multi-dimensional seismic excitations was constructed, and benchmark dynamic response solutions were generated using a finite element model for the training of deep learning models. Model performance was evaluated using both mean squared error and coefficient of determination metrics, and validated on independent test sets as well as representative earthquake records. Results demonstrate that compared with the finite element method (FEM), the deep learning models achieved a significant improvement in computational efficiency for predicting the nonlinear seismic responses of spherical shell structures. Among them, the BiGRU model, benefiting from its simplified gating mechanism and higher parameter efficiency, exhibited the best performance in predicting seismic responses in the [ ]- and [ ]-directions, with its predictions closely matching the FEM benchmark solutions. Additionally, the BiLSTM model, due to its more complex memory unit structure, showed superior directional balance when handling multi-dimensional inputs. Notably, although the attention mechanism enhanced the models’ capability to capture local nonlinear features, under the current conditions of limited dataset size and long-time sequences, its effect on improving the base models was not significant and even exhibited some degree of performance suppression. This study provides a methodological basis for selecting appropriate deep learning models for seismic safety assessments of complex spatial structures and offers in-depth insights into the applicability of different architectures under specific engineering scenarios.