Discrete Adaptive Mesh Refinement for Equivalence Modelling of Planar Beamlike Periodic Truss with Rigid Joints
Xinyuan Li, Ti Chen, Dongping Jin
Abstract: A novel approach to build a reduce-order model of planar beamlike periodic truss with both high accuracy and high computational efficiency is presented. This approach merges the classical equivalence method and the discrete adaptive mesh refinement (AMR). In the classical equivalent method, a continuous displacement field of cells is assumed by using the Taylor expansion. Based on the kinetic and potential energy equivalence, one cell of truss is reduced to a one-dimensional continuum beam element considering the identical displacement fields. The technique of discrete AMR is raised which aims to reduce the degree of freedoms of system and combines multiple continuum beam elements together based on a new mesh criterion. Then, a new reduced-order dynamic model in which the elements is much less than the cell number of truss is obtained. The analysis results indicate that the discrete AMR improves both the computational efficiency and accuracy by further reducing the dimension of system and overcoming the shortcoming of over-constraint in the classical equivalence method.
文章链接：https://www.sciencedirect.com/science/article/pii/S0263823125000618
