高级搜索
    孔凡, 王恒, 徐军, 董华. 基于多谐波平衡法的分数阶Duffing振子随机动力响应分析[J]. 应用基础与工程科学学报, 2023, 31(1): 103-112. DOI: 10.16058/j.issn.1005-0930.2023.01.008
    引用本文: 孔凡, 王恒, 徐军, 董华. 基于多谐波平衡法的分数阶Duffing振子随机动力响应分析[J]. 应用基础与工程科学学报, 2023, 31(1): 103-112. DOI: 10.16058/j.issn.1005-0930.2023.01.008
    KONG Fan, WANG Heng, XU Jun, DONG Hua. Multi-Harmonic Balance Method for Stochastic Response Determination of Duffing Oscillator Endowed with Fractional Derivative[J]. Journal of Basic Science and Engineering, 2023, 31(1): 103-112. DOI: 10.16058/j.issn.1005-0930.2023.01.008
    Citation: KONG Fan, WANG Heng, XU Jun, DONG Hua. Multi-Harmonic Balance Method for Stochastic Response Determination of Duffing Oscillator Endowed with Fractional Derivative[J]. Journal of Basic Science and Engineering, 2023, 31(1): 103-112. DOI: 10.16058/j.issn.1005-0930.2023.01.008

    基于多谐波平衡法的分数阶Duffing振子随机动力响应分析

    Multi-Harmonic Balance Method for Stochastic Response Determination of Duffing Oscillator Endowed with Fractional Derivative

    • 摘要: 计算分数阶非线性动力系统的随机动力响应一直是随机振动研究的难点.利用多谐波平衡法计算了随机动力激励下分数阶阻尼Duffing振子的确定性和随机动力响应.该方法将分数阶非线性运动微分方程转化为一组以响应Fourier级数为未知量的非线性代数方程,从而通过Newton迭代法求解该非线性代数方程组得到系统响应的Fourier级数;对获得的响应Fourier级数进行逆Fourier变换可获得响应时程;直接对激励功率谱抽样得到的样本激励Fourier级数,重复使用所提出方法,可获得样本响应Fourier级数或随机响应功率谱密度.其中,为避免系统非线性引起的频率混叠效应,提高计算精度并加快计算速度,将响应立方项的Fourier级数表示为响应Fourier级数的显式闭合形式.数值算例表明,所建议方法对非线强度不同的Duffing随机动力系统均具有良好的精度及适用性.

       

      Abstract: Determination of stochastic response of a nonlinear dynamic systme endowed with fractional-order derivative element has always been the difficulty of the random vibration research.A multi-harmonic balance method is presented for determining the stochastic response of a nonlinear Duffing oscillator endowed with fractional derivative elements and subject to stochastic excitation.Specifically, the fractional-order nonlinear differential equation is transferred into a set of nonlinear algebra equations in terms of unknown Fourier coefficients of response by employing the multi-harmonic balance method.Next, Newton's iterative approach is adopted for determining the unknown response Fourier coefficient.Thus, the response time history can be obtained by the inverse Fourier transform.Further, the Fourier coefficients of sample responses or the response power spectral density can be calculated by repeated use of the proposed method from the Fourier coefficients of excitations sampled from its power spectral density.A closed form solution of the Fourier coefficients of the cubic term is derived to avoid the so-called aliasing effect due to system non-linearity and enhance compuational accuracy and efficiency.Finally, a pertinent numerical example demonstrates the accuracy and reliability of the proposed method for determing response power spectral density of a Duffing oscillator with different levels of nonlinearity.

       

    /

    返回文章
    返回