Stochastic dynamic stiffness for damped taut membranes. (May 2021)

• Record Type: Journal Article
• Authors: Liu, Xiang; Zhao, Xueyi; Adhikari, Sondipon; Liu, Xiao
• Is Part Of: Computers & structures. Volume 248(2021)
• Publication Date: 2021-05
• Subjects: Stochastic dynamic stiffness method -- Membranes -- Stochastic dynamic analysis -- KL expansion; Structural engineering -- Data processing -- Periodicals; Electronic data processing -- Structures, Theory of -- Periodicals; 624.171
• DOI: 10.1016/j.compstruc.2021.106483
• Languages: English
• ISSNs: 0045-7949

Abstract

Highlights: Stochastic dynamic stiffness (SDS) is developed for damped taut membranes. 2D random fields for density and prestresses are represented by KL expansion. Closed-form SDS matrix is derived based on exact shape function and KL expansion. SDS matrices are assembled to model uncertain membrane assemblies with general BCs. The method's high efficiency and accuracy is demonstrated for whole frequency range. Abstract: An analytical stochastic dynamic stiffness formulation is developed for the dynamic analysis of damped membrane structures with parametric uncertainties. First, the exact general solution of a biaxially taut membrane in the frequency domain is derived, which is used as the frequency-dependent shape function. Both the material properties and the tension fields of the membrane are modelled as 2D random fields with an exponential autocorrelation function in both x and y directions. Then, the random fields are decomposed by Karhunen-Loève (KL) expansion. After a formulation procedure like the finite element method, the stochastic stiffness, and mass elemental matrices are derived based on the frequency-dependent shape function and the KL expansion, subsequently forming the stochastic dynamic stiffness matrix. The developed stochastic dynamic stiffness elements can be assembled to model membrane assemblies with general boundary conditions considering uncertainties. The proposed method can be utilized as a feasible technique for the efficient and accurate stochastic dynamic analysis in the whole frequency domain. The current research paves the way for stochastic dynamic stiffness formulation for other two-dimensional structures like plates and shells.