Inspired by the circulatory systems present in a wide range of living organisms, microvascular composites are a new class of fiber-reinforced polymers that possess a network of embedded microchannels. The focus of this work is placed on circulating a coolant through these microchannels to allow for the use of composite components in high temperature conditions, well beyond their traditional use. The design of these materials naturally leads to a multidisciplinary optimization problem: from a thermal point of view, more efficient active cooling can be achieved by multiplying the number of microchannels in which flows the coolant. But from a structural point of view, every channel represents a small void that may lead to stress concentration and negatively affect the stiffness and strength of the composite. Although the multidisciplinary computational design method to be developed as part of this project focuses on high temperature applications, microvascular composites are being considered for a wide range of multidisciplinary applications, including new electrical, electromagnetic or sensing properties. The combined computational and experimental research project will provide a unique multidisciplinary training experience for two graduate students and one summer undergraduate research assistant.
The hierarchical computational design method to be developed in this project will lead to important advances in the efficient and accurate modeling of heterogeneous materials, and in the formulation of a robust gradient-based shape optimization approach that eludes issues associated with mesh distortion often associated with conventional finite element methods. At the heart of the modeling effort is an isogeometric interface-enriched generalized finite element method (IIGFEM) that allows for the thermal and structural modeling of microvascular composites with finite element discretizations that do not conform to the embedded microchannel network and the composite microstructure. The IIGFEM is then combined with a hierarchical, gradient-based shape optimization scheme to compute the optimal shape of the microchannel network based on a set of objective functions and constraints associated with the thermal performance and flow efficiency of the embedded network, and its impact on the structural properties of the microvascular composite. Building on state-of-the-art computational and experimental tools, this top-down design method, which relies on a hierarchy of thermo-mechanical models and length scales, offers an efficient approach to tackle the size and complexity of a design process characterized by multiple, conflicting, multi-disciplinary objective functions and constraints.
