Component-wise analysis of laminated structures by hierarchical refined models with mapping features and enhanced accuracy at layer to fiber-matrix scales

Abstract

ABSTRACTModern methodologies for the analysis of composite structures are demanded to satisfy the accuracy requirements at different scales, from micro to macro and possibly in a global/local sense. In this domain, the present paper proposes a hierarchical, component-wise approach for the linear static analysis of layered structures. By employing the Carrera Unified Formulation and a variational statement, finite element arrays of refined beam models are expressed in terms of fundamental nuclei. Legendre-based polynomials are utilized to implement, in a hierarchical form, higher-order beam kinematics. Also, curved cross-section geometries are formulated in a correct and consistent manner through mapping blending functions. Mapping and refined kinematics beam models are, thus, inherently combined to give the component-wise method, according to which each component of the structure (e.g., layer, fibers, and matrix) is modeled independently and with its geometric and mechanical characteristics. This approach is demonstrated to provide enhanced accuracy for the analysis of composite structures, from layer scale to fiber-matrix level, but still with low computational costs.