Multi-scale modeling of 3D multi-axial compressive failures in unidirectional and woven composites
Abstract
The compressive failure of unidirectional (UD) and woven fiber composites under multi-axial stress states is a complex multi-scale and multi-mechanism phenomenon. So far, there does not exist a single continuum scale constitutive model that can reliably predict these failures under general triaxial stress states. Additionally, there exists significant ambiguity in parameter identification fordamage models for compressive failures in general. These gaps in predictive capabilities and knowledge motivate the proposed research. The technical goals are (i) to develop multi-mechanism, multi-scale, constitutive models for compressive failure of UD and wovencomposites for multi-axial compressive failures and (ii) to develop a standardized process of model parameter identification from distinct experiments. The multiaxial and multi-mechanism failures in UD and woven composites will be modeled here by the development of two multi-scale microplane constitutive models: viz the cylindrical microplane model for UD composites and the microplane triad model for woven composites. Both these models have an inherent dual scale nature and so are particularly well-suited to capture the aforementioned compressive failure mechanisms. These will be formulated here to capture the subscale constituent level stress-strain and inelastic behaviors during various compressive failure mechanisms. These include the pressure sensitive matrix micro-cracking damage, 3D oriented fiber kinking and subsequent rotation, fiber shear fractures, micro-crack sliding and friction, and transverse compressive failures. The underpinning hypothesis is that by formulating various damage mechanisms at the constituent level and capturing the normal and shear interactions on microcracks of various orientations, would yield the appropriate multi-axial failure behavior when homogenized to continuum scale. In addition, a semi-discreet approach will also be pursued here, which blends the microplane model for the matrixresponse with semi-discreet beam elements for fiber behavior, with the goal of thoroughly representing the multi-axial failure of composites. The models will be formulated and implemented in the commercial finite element software Abaqus as usersubroutines. The models will be systematically calibrated and validated using various tests available in literature and from prior PI efforts under ONR funding. The calibration process will aim to identify minimum required experiments necessary, determine gaps inexisting test data, as well as develop a standard practice for the calibration process. The various model predictions that will be evaluated include the peak stress, the brittleness or ductility of the failure behavior, the damage zone sizes and kink band widths,energy dissipation, and post-peak plateau stress capacity. The project is expected to yield two versatile and powerful multi-scale constitutive models, that can be applied to compressive failures of UD and woven composites under general multi-axial stresses loads. The pioneering semi-discreet approach, if successful, would signify a notable leap forward in constitutive modeling, offering a refined and adaptable means to predict micro-mechanical effects in the context of multi-axial compressive failure in composite structures. In addition, we expect to build a firm understanding of the minimum required experiments for complete and unambiguous calibration of such models. Collectively, we expect to reduce the ambiguity and improve the reliability of computer-aided designs of composite structures under compressive loads in various naval, aerospace, and other applications.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jan 13, 2025
- Source ID
- N000142512106
Entities
People
- Kedar Kirane
Organizations
- Office of Naval Research
- Research Foundation for the State University of New York
- United States Navy