850 工程科学学报,第43卷,第6期 点和拾振点位置,利用式(23)进行频响函数的计 以及频响函数值均与实验结果较为接近,从而证 算,首先,利用反推辨识,对前5阶频响函数值依 明了提出的用复模量非均匀分布的虚拟材料模拟 次分配0.3、0.2、0.1、0.1、0.3的权重,设置种群数 螺栓影响区进而实施半解析建模可实现较高的仿 量、变异概率、交叉概率、迭代次数为50、0.05、 真计算精度 0.9和50,不断迭代模型中虚拟材料的耗能模量 值,使模型的频响函数曲线尽可能接近实测曲线, 参考文献 完成迭代后,虚拟材料的耗能模量为1.625×10°Pa [1]Reid J D,Hiser N R.Detailed modeling of bolted joints with 接着用此耗能模量值获得最终的频响函数曲线并 slippage.Finite Elem Anal Des,2005,41(6):547 [2] 与实测比对,见图4.从图中也可看出,仿真与实 Sawa S,Ishimura M,Sekiguchi Y,et al.3-D FEM stress analysis and mechanical characteristics in bolted joints under external 测的频响函数也有较好的接近 tensile loadings//ASME 2015 International Mechanical Engineering Congress and Exposition.Houston,2015: Simulation Measurement IMECE2015-50572,V02BT02A036 [3] Luyt P C B,Theron N J,Pietra F.Non-linear finite element modelling and analysis of the effect of gasket creep-relaxation on 10- circular bolted flange connections.IntJ Press Vessels Pip,2017, 15052 106 [4] Zhang D C,Wang G,Huang F H,et al.Load-transferring 01002003004005006007008009001000 Frequency/Hz mechanism and calculation theory along engaged threads of high- strength bolts under axial tension.J Constr Steel Res,2020,172 图4实测与仿真频响函数对比 106153 Fig.4 Comparison of the frequency response functions obtained based on the measured and simulated data [5] Luan Y,Guan Z Q,Cheng G D,et al.A simplified nonlinear dynamic model for the analysis of pipe structures with bolted 4 结论 flange joints.J Sound Vib,2012,331(2):325 [6] Meisami F,Moavenian M,Afsharfard A.Nonlinear behavior of (1)本文提出用复模量非均匀分布的虚拟材 single bolted flange joints:A novel analytical model.Eng Struct, 料来模拟螺栓搭接部分的刚度及阻尼特性,并给 2018,173:908 出详细的建模流程和方法.实践表明本文采用假 [7] Xiang J W,Zhao S W,Li D C,et al.An improved spring method 定的正弦、抛物线和线性等非均匀变化的虚拟材 for calculating the load distribution in multi-bolt composite joints Composites Part B,2017,117:1 料模拟螺栓搭接部分,相比于均匀分布能够更加 [8]Deaner B J,Allen M S,Starr M J,et al.Application of viscous and 精确地模拟螺栓连接结构的动力学特性 Iwan modal damping models to experimental measurements from (2)采用虚拟材料模拟螺栓搭接部分,虚拟材 bolted structures.J Vib Acoust,2015,137(2):021012 料的材料参数用复模量表示,可直接生成复数形 [9]Li L,Cai A J,Ruan X G,et al.Stiffness modeling and 式的刚度矩阵,省却了常规建模中生成结合部阻 characteristic analysis of bolted joints.JVib Eng,2017,30(1):1 尼矩阵的步骤,在保证模型精确性的基础上,简化 (李玲,蔡安江,阮晓光,等.栓接结合部刚度建模与特性分析 了螺栓搭接部分的建模过程.提出利用反推法辨 振动工程学报,2017,30(1):1) [10]Sun W,Li X Z,Han Q K.Nonlinear joint parameter identification 识虚拟材料的储能模量和耗能模量,其中储能模 for bolted beam structure.J Vib Eng,2013,26(2):185 量参数确定为其分布函数的最大值ED,耗能模量 (孙伟,李星占,韩清凯螺栓联接梁结构结合部非线性特性参 取相同值.通过所提出的反推辨识流程可较为精 数辨识振动工程学报,2013,26(2):185) 确的确定虚拟材料的参数 [11]Iranzad M,Ahmadian H.Identification of nonlinear bolted lap (3)为了更好地实施所研发的建模理念以及 joint models.Comput Struct,2012,96-97:1 反推辨识确认虚拟材料的复模量参数,自行研发 [12]Wang D,Fan X H.Nonlinear dynamic modeling for joint 了半解析程序.重点描述了复模量非均匀分布的 interfaces by combining equivalent linear mechanics with multi 虚拟材料引入螺栓连接结构半解析模型的过程, objective optimization.Acta Mech Solida Sin,2020.33(4):564 [13]Zha Y J,Zhang J F,Yu D W,et al.Modeling method for bolted 并推导出了快速求解半解析模型任意锤击点与拾 joint interfaces based on transversely isotropic virtual materials// 振点处频响函数的公式.最终的研究表明:用所创 2018 IEEElASME International Conference on Advanced 建的半解析模型计算获得的固有频率、模态振型 Intelligent Mechatronics (AIM).Auckland,2018:1118点和拾振点位置,利用式(23)进行频响函数的计 算. 首先,利用反推辨识,对前 5 阶频响函数值依 次分配 0.3、0.2、0.1、0.1、0.3 的权重,设置种群数 量、变异概率、交叉概率、迭代次数为 50、0.05、 0.9 和 50,不断迭代模型中虚拟材料的耗能模量 值,使模型的频响函数曲线尽可能接近实测曲线, 完成迭代后,虚拟材料的耗能模量为 1.625×109 Pa. 接着用此耗能模量值获得最终的频响函数曲线并 与实测比对,见图 4. 从图中也可看出,仿真与实 测的频响函数也有较好的接近. 0 100 200 300 400 500 600 700 800 900 1000 0 Simulation Measurement Amplitude/(m·s−1·N−1 ) 10−1 10−2 10−3 10−6 10−5 10−4 Frequency/Hz 图 4 实测与仿真频响函数对比 Fig.4 Comparison of the frequency response functions obtained based on the measured and simulated data 4 结论 (1)本文提出用复模量非均匀分布的虚拟材 料来模拟螺栓搭接部分的刚度及阻尼特性,并给 出详细的建模流程和方法. 实践表明本文采用假 定的正弦、抛物线和线性等非均匀变化的虚拟材 料模拟螺栓搭接部分,相比于均匀分布能够更加 精确地模拟螺栓连接结构的动力学特性. ED (2)采用虚拟材料模拟螺栓搭接部分,虚拟材 料的材料参数用复模量表示,可直接生成复数形 式的刚度矩阵,省却了常规建模中生成结合部阻 尼矩阵的步骤,在保证模型精确性的基础上,简化 了螺栓搭接部分的建模过程. 提出利用反推法辨 识虚拟材料的储能模量和耗能模量,其中储能模 量参数确定为其分布函数的最大值 ,耗能模量 取相同值. 通过所提出的反推辨识流程可较为精 确的确定虚拟材料的参数. (3)为了更好地实施所研发的建模理念以及 反推辨识确认虚拟材料的复模量参数,自行研发 了半解析程序. 重点描述了复模量非均匀分布的 虚拟材料引入螺栓连接结构半解析模型的过程, 并推导出了快速求解半解析模型任意锤击点与拾 振点处频响函数的公式. 最终的研究表明:用所创 建的半解析模型计算获得的固有频率、模态振型 以及频响函数值均与实验结果较为接近,从而证 明了提出的用复模量非均匀分布的虚拟材料模拟 螺栓影响区进而实施半解析建模可实现较高的仿 真计算精度. 参 考 文 献 Reid J D, Hiser N R. Detailed modeling of bolted joints with slippage. Finite Elem Anal Des, 2005, 41(6): 547 [1] Sawa S, Ishimura M, Sekiguchi Y, et al. 3-D FEM stress analysis and mechanical characteristics in bolted joints under external tensile loadings//ASME 2015 International Mechanical Engineering Congress and Exposition. Houston, 2015: IMECE2015-50572, V02BT02A036 [2] Luyt P C B, Theron N J, Pietra F. Non-linear finite element modelling and analysis of the effect of gasket creep-relaxation on circular bolted flange connections. Int J Press Vessels Pip, 2017, 150: 52 [3] Zhang D C, Wang G, Huang F H, et al. Load-transferring mechanism and calculation theory along engaged threads of highstrength bolts under axial tension. J Constr Steel Res, 2020, 172: 106153 [4] Luan Y, Guan Z Q, Cheng G D, et al. A simplified nonlinear dynamic model for the analysis of pipe structures with bolted flange joints. J Sound Vib, 2012, 331(2): 325 [5] Meisami F, Moavenian M, Afsharfard A. Nonlinear behavior of single bolted flange joints: A novel analytical model. Eng Struct, 2018, 173: 908 [6] Xiang J W, Zhao S W, Li D C, et al. An improved spring method for calculating the load distribution in multi-bolt composite joints. Composites Part B, 2017, 117: 1 [7] Deaner B J, Allen M S, Starr M J, et al. Application of viscous and Iwan modal damping models to experimental measurements from bolted structures. J Vib Acoust, 2015, 137(2): 021012 [8] Li L, Cai A J, Ruan X G, et al. Stiffness modeling and characteristic analysis of bolted joints. J Vib Eng, 2017, 30(1): 1 (李玲, 蔡安江, 阮晓光, 等. 栓接结合部刚度建模与特性分析. 振动工程学报, 2017, 30(1):1) [9] Sun W, Li X Z, Han Q K. Nonlinear joint parameter identification for bolted beam structure. J Vib Eng, 2013, 26(2): 185 (孙伟, 李星占, 韩清凯. 螺栓联接梁结构结合部非线性特性参 数辨识. 振动工程学报, 2013, 26(2):185) [10] Iranzad M, Ahmadian H. Identification of nonlinear bolted lap joint models. Comput Struct, 2012, 96-97: 1 [11] Wang D, Fan X H. Nonlinear dynamic modeling for joint interfaces by combining equivalent linear mechanics with multiobjective optimization. Acta Mech Solida Sin, 2020, 33(4): 564 [12] Zha Y J, Zhang J F, Yu D W, et al. Modeling method for bolted joint interfaces based on transversely isotropic virtual materials// 2018 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM). Auckland, 2018: 1118 [13] · 850 · 工程科学学报,第 43 卷,第 6 期