Caglar Oskay delivered an invited lecture at the Naval Research Laboratory
Lecture Title: Reduced-Order Multiscale Modeling of Heterogeneous Materials.
Location: Materials Science and Technology Division, US Naval Research Laboratory, Washington, DC.
Abstract:
Tremendous amount of effort has been devoted to the development of multiscale computational modeling and simulation strategies for predicting the mechanical and functional response of heterogeneous materials over the past few decades. The focus of this talk is on the coupled-hierarchical (aka. information passing) type of multiscale computational modeling. While this type of multiscale modeling has a plethora of benefits from the modeling perspective, high computational complexity associated with the “tyranny of scales” so far limited its impact beyond the academic realm.
We will present a reduced order multiscale modeling approach, named “eigendeformation- based reduced order homogenization method (EHM)” for computationally efficient and accurate multiscale analysis of materials with heterogeneous microstructures. EHM employs the idea of precomputing certain information on the material microstructure such as the influence functions, localization operators and coefficient tensors through RVE scale simulations, prior to the macroscale analysis. The model order is reduced by assuming that the inelastic response fields within the RVE are spatially piecewise constant within subdomains of the characteristic volume. The applications of the EHM method to polycrystalline metals as well as polymer matrix composite microstructures will be presented. We also present the results of a recent study on prediction of the static and fatigue failure response of carbon fiber reinforced composite materials. The study, administered by the AFRL and Lockheed Martin, is the assessment of the state of the art in predictive capabilities of the composite failure prediction methods and models as a part of a larger program on the assessment and quantification of applying damage tolerant design principles to aerospace composite structures. The proposed reduced order multiscale approach was found to be accurate in predicting the fatigue properties and damage propagation characteristics of complex layup and specimen configurations (notched and unnotched). Based on the blind prediction results, the model improvement strategies are devised and employed to further improve the predictive capability of the composite damage and failure model.