Buy article PDF
The purchased file will be sent to you
via email after the payment is completed.
US$ 35
|
Smart Structures and Systems Volume 34, Number 5, November 2024 , pages 283-297 DOI: https://doi.org/10.12989/sss.2024.34.5.283 |
|
|
|
Smart Lyapunov LMI criterion for TMD RC structure systems via deep reinforcement learning algorithms |
||
ZY Chen, Huakun Wu, Ruei-Yuan Wang, Yahui Meng and Timothy Chen
|
||
| Abstract | ||
| This study presents a novel approach to enhancing the seismic performance of tuned mass damper (TMD) systems in reinforced concrete (RC) structures through the implementation of a Smart Lyapunov Linear Matrix Inequality (LMI) criterion, optimized via deep reinforcement learning (DRL) algorithms. Traditional methods for TMD design often rely on heuristic or empirical approaches, which may not adequately address the complexities of dynamic interactions in RC structures under varying seismic loads. By leveraging the capabilities of DRL, this research develops a framework that dynamically adjusts TMD parameters in real-time, ensuring optimal performance across a range of seismic scenarios. The proposed Smart Lyapunov LMI criterion provides a robust mathematical foundation for stability and performance assessment, allowing for the systematic evaluation of TMD effectiveness in mitigating structural vibrations. Through extensive numerical simulations, the integration of DRL algorithms demonstrates significant improvements in the adaptability and efficiency of TMD systems, outperforming conventional design methods. The results indicate that the proposed approach not only enhances the resilience of RC structures under seismic events but also contributes to the development of intelligent structural control systems. This research underscores the potential of combining advanced control theories with artificial intelligence techniques to address contemporary challenges in structural engineering, paving the way for more resilient and adaptive building designs in earthquake-prone regions. | ||
| Key Words | ||
| Deep Reinforcement Learning (DRL); fuzzy model; hybrid heuristic search algorithm; seismic performance, sustainable and resilience; vibration mitigation | ||
| Address | ||
| (1) ZY Chen, Ruei-Yuan Wang, Yahui Meng: School of Science, Guangdong University of Petrochemical Technology, Maoming, Guangdong, China; (2) Huakun Wu: School of Computer Science, Guangdong Polytechnic Normal University, Guangzhou, Guangdong, China; (3) Timothy Chen: Division of Engineering and Applied Science, Caltech, CA 91125, USA. | ||