Model-Based Mechanical Property and Structural Failure Prediction of Pseudo Ductile Hybrid Composite

Lightweight fiber reinforced composites are widely used in engineering structures, which often fail catastrophically due to the uncertainty of external loads and their brittle nature. The development of pseudo ductile hybrid composites was the proposed solution to create minimal ductility in fiber reinforced composites so that equipment downtime, cost, and loss of lives can be minimized in their structural application. However, the development of pseudo ductile hybrid composites does not guarantee that pseudo ductile hybrid composite is prone to failure. As a result, different models, including Halpin-Tsai, Hashin and Shtrikman, Weibull, and log-normal models, were developed to predict degradation of mechanical properties and structural failure so that prior recognition of failure can be achieved. The current structural health monitoring research trend shows the development of hybrid mechanical property and structural failure prediction models spalling the drawback of data-driven and physics-based models. Physics-based models require detail understanding of the root cause of failure in terms of mathematical or physical model to predict failure progression whereas data-driven models rely on historical data or sensor data collected from machineries or structures. While hybrid models combine the strengths of both physics-based and data-driven models providing manageable uncertainty and more accurate prediction. This paper aims to review model-based mechanical property and structural failure prediction strategies with regard to pseudo ductile hybrid composites highlighting future research directions and challenges, and offering insights beneficial to the research and industrial communities.