Skip to main content

ORIGINAL RESEARCH article

Front. Phys.
Sec. Interdisciplinary Physics
Volume 12 - 2024 | doi: 10.3389/fphy.2024.1475622
This article is part of the Research Topic Wave Propagation in Complex Environments, Volume II View all 7 articles

Double array system identification research based on LSTM neural network

Provisionally accepted
Chunhua Gao Chunhua Gao Mingyang Wang Mingyang Wang *Yifei Sima Yifei Sima Zihan Yuan Zihan Yuan
  • Xinyang Normal University, Xinyang, China

The final, formatted version of the article will be published soon.

    The earthquake simulation shaking table array is an important experimental equipment with a wide range of applications in the field of earthquake engineering. To efficiently address the complex nonlinear problems associated with earthquake simulation shaking array systems, this paper proposes the identification of the earthquake simulation shaking array system using the Long Short-Term Memory (LSTM) algorithm. A dual array system model with flexible specimen connections is established, and this system is identified using the LSTM neural network. The LSTM neural network was validated for identifying the dual array closed-loop system of the earthquake simulation shaking table by using three natural waves and one artificial wave. The results demonstrated that the similarity between the predicted output and the theoretical output of the network identified by LSTM exceeded 0.999. This indicates that the algorithm can accurately reproduce the characteristics of the shaking table itself and shows good performance in time series prediction and data mining. References for earthquake simulation shaking array system experiments are provided.

    Keywords: system identification, Dual array, LSTM Neural Network, Shaking table, deep learning

    Received: 04 Aug 2024; Accepted: 10 Dec 2024.

    Copyright: © 2024 Gao, Wang, Sima and Yuan. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

    * Correspondence: Mingyang Wang, Xinyang Normal University, Xinyang, China

    Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.