EI、Scopus 收录
中文核心期刊

具有随机参数的空间结构热致振动分析

THERMAL VIBRATION ANALYSIS OF SPATIAL STRUCTURES CONSIDERING RANDOM PARAMETERS

  • 摘要: 针对热环境下考虑多个随机参数的空间结构动力学模型,本文提出了一种基于混沌多项式展开(PCE)的随机柔性多体系统动力学计算方法,研究了结构热致振动的动态响应,分析了不同随机参数对热致振动的影响。首先,利用绝对节点坐标法(ANCF)对涉及热效应的空间简化模型进行建模,通过傅里叶温度单元法分析结构温度响应,推导出空间热-结构耦合模型。其次,利用PCE法构建原系统的代理模型,通过单项求容积法则(MCR)选取样本点,结合随机响应面法(SRSM)计算出PCE系数,得到热致振动响应的均值和标准差,并与蒙特卡洛模拟(MCS)结果进行对比,发现本文提出的算法在满足精度要求的情况下,所需样本点仅为MCS方法的1.5%,计算效率比MCS方法提高了76倍。最后,分析了多个随机参数包括弹性模量、梁的长度、太阳辐射吸收率、表面辐射系数、太阳辐射强度和太阳辐射角度对结构热致振动响应的影响,结果表明梁的长度和弹性模量的随机离散程度对热致振动响应的影响最为明显。

     

    Abstract: In this paper, a stochastic dynamical analysis method based on the polynomial chaos expansion (PCE) is proposed to study the thermal vibration of flexible multi-body systems with multiple random parameters. The influences of different random parameters on the thermal vibration is explored. Firstly, to derive the spatial thermal structure coupling model, the absolute node coordinate formulation (ANCF) is used to model the spatial simplified model involving thermal effect, and the Fourier temperature element method is used to analyze the structural temperature response. Secondly, the surrogate model of the stochastic dynamical system is constructed through PCE method, and sample points are selected by the monomial cubature rules (MCR). Then, the PCE coefficients are obtained by the stochastic response surface method (SRSM). The mean and standard deviation of thermal vibration response are obtained, which are compared with Monte Carlo simulation (MCS) results to verify the accuracy and efficiency of the proposed algorithm. It is found that compared with the MCS, the sample points of the proposed method reduce to only 1.5% and the calculation efficiency is improved by 76 times. Finally, the influences of several random parameters, including elastic modulus, beam length, solar radiation absorption rate, surface radiation coefficient, solar radiation intensity and solar radiation angle on the thermal vibration response of the structure are studied. The results show that the random dispersion degree of beam length and elastic modulus have the most obvious influence on the thermal vibration response.

     

/

返回文章
返回