The Missing Dimension in Digital Twinning of Composite Structures: Residual Stress, Fatigue Prediction, and Uncertainty Quantification

Digital twinning is transforming composite manufacturing by developing dynamic virtual models of physical components that mirror real-time performance. This technology facilitates improved design optimization, process efficiency, and proactive maintenance strategies. However, accurately predicting residual stress and its impact on fatigue life remains a significant challenge in composites. Internal stresses, particularly during the curing process, directly influence structural integrity and fatigue performance, as evidenced by the failure of NASA’s X-33. While analytical models provide insights, they are computationally expensive and depend on complex, temperature-dependent properties. Experimental methods for measuring bulk residual stresses in polymer composites are also limited in accuracy and resolution.

This study critically evaluates both numerical and experimental methods for assessing residual stress and fatigue life in polymer composites. The contour method, originally applied to metals, shows promising results in polymer composites, offering new avenues for residual stress mapping. Additionally, it explores the adaptation of Taylor’s critical plane method for predicting fatigue behaviour under multiaxial loading conditions using quantified bulk residual stress. We propose an integrated framework combining simulated residual stress into fatigue life assessment to enhance the accuracy of digital twins in composite structures. By merging analytical, numerical, and experimental techniques, this framework enables real-time validation and uncertainty quantification, paving the way for more reliable and efficient design and manufacturing processes. This pioneering approach advances the predictive power of digital twins and ensures more robust, sustainable composite materials in engineering applications.

Keywords: Digital Twins, Residual Stress, Composite Manufacturing, Simulation, Fatigue Life, Experimental Characterization