An imprecise multiscale uncertainty quantification framework for fiber reinforced composites

The study focuses on the reliability and global sensitivity analysis of fiber-reinforced composite radome structures, considering uncertainty from a multiscale perspective. Macroparameters are estimated based on micro- parameters using the multiscale analysis method for composites, and a reliability analysis model of the composite structure at the macrolevel is constructed. The material performance mechanism is explored in depth, both "from bottom to top" and "from top to bottom", to reveal its inherent laws. Due to insufficient variable distribution information, an imprecise probabilistic model is introduced to characterize the uncertainty effect in multiscale composite analysis. A nested optimization calculation method is applied to obtain reliability and sensitivity results. To ensure both calculation accuracy and efficiency, the regression and classification problems encountered in the proposed framework are addressed using two support vector machine models. The reliability and sensitivity analysis under the imprecise probabilistic framework can help engineers identify significant influential factors, thereby guiding the design of composite radome structures.