航空学报 > 2022, Vol. 43 Issue (9): 225952-225952   doi: 10.7527/S1000-6893.2021.25952

平稳随机载荷的信号特征提取与深度神经网络识别

杨特1, 杨智春1, 梁舒雅1, 康在飞1, 贾有1,2   

  1. 1. 西北工业大学 航空学院, 西安 710072;
    2. 太原科技大学 应用科学学院, 太原 030024
  • 收稿日期:2021-06-15 修回日期:2021-08-07 出版日期:2022-09-15 发布日期:2021-09-22
  • 通讯作者: 杨智春,E-mail:yangzc@nwpu.edu.cn E-mail:yangzc@nwpu.edu.cn
  • 基金资助:
    国家自然科学基金(12102353)

Feature extraction and identification of stationary random dynamic load using deep neural network

YANG Te1, YANG Zhichun1, LIANG Shuya1, KANG Zaifei1, JIA You1,2   

  1. 1. School of Aeronautics, Northwestern Polytechnical University, Xi 'an 710072, China;
    2. School of Applied Science, Taiyuan University of Science and Technology, Taiyuan 030024, China
  • Received:2021-06-15 Revised:2021-08-07 Online:2022-09-15 Published:2021-09-22
  • Supported by:
    National Natural Science Foundation of China (12102353)

摘要: 针对线性时不变结构的平稳随机载荷识别问题, 从结构的动力学响应求解原理出发, 利用小波变换对于信号特征的提取能力与长短期记忆神经网络(LSTM)对于序列问题的强大建模与映射能力, 提出了一种针对平稳随机载荷的特征信号识别方法, 通过对作用于三自由度振动系统数值模型上的平稳随机动载荷识别, 证明了方法的可行性。对一个受2点平稳随机载荷作用的加筋壁板结构模型进行动载荷识别实验, 结果表明, 用提出的方法识别的动载荷均方根相对误差均小于5%, 该动载荷识别方法具有良好的识别能力。

关键词: 平稳随机载荷, 小波变换, 振动信号特征提取, 深度神经网络, 动载荷识别

Abstract: A feature signal identification method for stationary random dynamic load is proposed based on the dynamic principle of structures. using Wavelet transform is used to extract the time-frequency characteristics of signals, and Long-Short Term Memory (LSTM) is employed to model and map sequence problems. The feasibility of the method is proved byidentification of stationary random dynamic loads acting on a three-degree-of-freedom vibration system. The dynamic load identification experiment is carried out on a stiffened panel structure model under two-point stationary random loads. The results show that the root mean square error of dynamic load identified by the proposed method is less than 5%, and the method has good identification ability.

Key words: stationary random dynamic load, wavelet transform, vibration signal feature extraction, deep neural network, dynamic load identification

中图分类号: