Convolutional LSTM-Attention Based Encoder-Decoder Neural Network for Prediction of Chaotic Vibrations of Multi-Dimensional Dynamic Systems

被引:0
作者
Wang, Luyao [1 ,2 ]
Dai, Liming [1 ,2 ]
Zhao, Haixing [1 ]
Fang, Pan [3 ]
机构
[1] Qinghai Normal Univ, Coll Comp, Xining 810008, Qinghai, Peoples R China
[2] Univ Regina, Ind Syst Engn, Regina, SK S4S 0A2, Canada
[3] Southwest Petr Univ, Sch Mech Engn, Chengdu 610500, Peoples R China
基金
加拿大自然科学与工程研究理事会;
关键词
Convolutional-LSTM; multi-dimensional dynamic system; chaotic vibrations; encoder-decoder; OSCILLATOR; STABILITY; MODEL;
D O I
10.1142/S0219455424502274
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
The present research proposes a convolutional long-short term memory (ConvLSTM) with an attention mechanism (AM) model, termed as ConvLSTM-AM, to conduct prediction of chaotic vibrations of multi-dimensional dynamic systems. The proposed data-driven model is based on an encoder-decoder architecture where the lengths of inputs and outputs are variable. Different from other conventional benchmarks which consider the temporal correlation solely to deal with the chaotic sequences, this research work takes the spatial information into account. In this sense, the ConvLSTM is adopted as an encoder to acquire useful chaotic spatiotemporal patterns and retain long-term successive dependencies. LSTM and AM in this research are taken as the main structures of the procedure in decoding, in which the LSTM is used as the further temporal processor and AM is stacked on the top to exploit more salient information of the historical data. Among that, a residual connection between the outputs of LSTM and the information of attention is considered in AM to prevent gradient vanishment. Two datasets of chaotic vibrations of multi-dimensional systems are employed to adequately illustrate the effectiveness and feasibility of the proposed model. Besides, five conventional benchmarks are built to demonstrate the advantages of the proposed model in terms of both training and generalization performance. As found in the research, the training time is reduced with lower testing loss in comparing with the other five counterparts, as the spatial information introduced expedites the training convergence. The present research provides a useful guidance for predicting and analysing chaotic vibrations of multi-dimensional dynamic systems.
引用
收藏
页数:29
相关论文
共 55 条
  • [11] Control of an extending nonlinear elastic cable with an active vibration control strategy
    Dai, L.
    Sun, L.
    Chen, C.
    [J]. COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION, 2014, 19 (10) : 3901 - 3912
  • [12] Dai L., 1998, PROC 15 CANADIAN C A, V1, P390
  • [13] Dai L., 2008, Nonlinear Dynamics of Piecewise Constant Systems and Implementation of Piecewise Constant Arguments
  • [14] DAI L, 2008, J COMPUT NONLIN DYN, P1
  • [15] An algorithm for diagnosing nonlinear characteristics of dynamic systems with the integrated periodicity ratio and lyapunov exponent methods
    Dai, Liming
    Xia, Dandan
    Chen, Changping
    [J]. COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION, 2019, 73 : 92 - 109
  • [16] Accuracy and Reliability of Piecewise-Constant Method in Studying the Responses of Nonlinear Dynamic Systems
    Dai, Liming
    Wang, Xiaojie
    Chen, Changping
    [J]. JOURNAL OF COMPUTATIONAL AND NONLINEAR DYNAMICS, 2015, 10 (02):
  • [17] An active control strategy for vibration control of an axially translating beam
    Dai, Liming
    Chen, Changping
    Sun, Lin
    [J]. JOURNAL OF VIBRATION AND CONTROL, 2015, 21 (06) : 1188 - 1200
  • [18] Chaos for a microelectromechanical oscillator governed by the nonlinear Mathieu equation
    DeMartini, Barry E.
    Butterfield, Holly E.
    Moehlis, Jeff
    Turner, Kimberly L.
    [J]. JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, 2007, 16 (06) : 1314 - 1323
  • [19] Jaya-Based Long Short-Term Memory Neural Network for Structural Damage Identification with Consideration of Measurement Uncertainties
    Ding, Zhenghao
    Hou, Rongrong
    Xia, Yong
    [J]. INTERNATIONAL JOURNAL OF STRUCTURAL STABILITY AND DYNAMICS, 2022, 22 (14)
  • [20] Temporal self-attention-based Conv-LSTM network for multivariate time series prediction
    Fu, En
    Zhang, Yinong
    Yang, Fan
    Wang, Shuying
    [J]. NEUROCOMPUTING, 2022, 501 : 162 - 173