FRP工程结构多尺度不确定性分析研究进展

纤维复合材料(FRP)结构在桥梁工程中的应用日益增多，但其采用复杂的制造工艺导致FRP材料属性及其构成结构性能存在较大的离散性，同时不确定性参数的多层级、非均质和种类多等特点使得其结构可靠性较难准确量化。为推动FRP结构不确定性分析方法和应用研究的进一步发展，从FRP结构不确定性来源、材料力学性能概率预测方法、结构可靠性分析方法3个方面梳理国内外FRP结构不确定性分析方法的研究进展，并探讨其不足和发展趋势。不确定性来源方面，从复合材料制造缺陷分类出发，详细阐述典型制造缺陷的成因、对FRP力学性能的影响及其引起的不确定性。材料力学性能概率预测方法方面，综述了基于均匀化理论的摄动随机有限元法和谱随机有限元法等概率均匀化方法的适用范围，并介绍了上述方法在细观尺度不确定性对宏观力学属性的影响程度和敏感性分析中的应用。结构可靠性分析方面，分别综述了用于FRP结构静力分析、动力分析和屈曲稳定分析的多尺度不确定性分析方法以及可靠性分析方法的研究。研究结果表明：FRP结构显著的几何多层级和复杂的制造工艺不可避免地产生各种缺陷而造成力学性能下降和离散性，采用试验、细观力学模型和均匀化方法并结合随机有限元...

Research Advances on Multiscale Uncertainty Quantification for Fiber Reinforced Polymer Composite Structures

The application of fiber-reinforced polymer (FRP) composites in bridge structures has garnered an increasing interest. However, the complicated manufacturing processes and complexity of microstructures of FRP composite structures lead to larger variations in the performance of FRP composites compared with those of structures made of conventional materials. In addition, reliability and safety of FRP composite structures are difficult to determine. This study provides a comprehensive review of the sources of uncertainty, probabilistic prediction methods of material mechanical properties, and structural reliability to explore the state-of-the-art uncertainty quantification for FRP composite structures and research trends. In terms of the sources of uncertainty, starting from the classification of composite material manufacturing defects, causes of typical manufacturing defects, their influences on the mechanical properties of composite materials, and caused uncertainties are elaborated. In terms of the probabilistic prediction methods of material mechanical properties, the application scope of the perturbation-based stochastic finite element and spectral stochastic finite element methods for probabilistic homogenization are reviewed. Additionally, the application of these methods to determine the influences of microscope uncertainties on the macroscale mechanical properties is discussed. In terms of the structural reliability analysis, the existing methods and research results of the static, dynamic, buckling stability, and reliability analyses of FRP composite structures are reviewed. Studies have demonstrated that significant geometric multiscale levels and complex manufacturing processes of FRP composite materials inevitably produce various defects, and cause mechanical performance degradation and dispersion. The principal tools to analyze multiscale uncertainties include implementing experiments, micromechanical models, and homogenization methods, combined with stochastic finite element methods, etc., to propagate uncertainties from constituent materials and plylevel properties to structural responses. Experiments can only measure composite materials with specific component materials, whereas theoretical models are typically stablished under certain assumptions. Neither existing experimental nor theoretical methods can comprehensively describe uncertainties in the mechanical properties and structural responses induced by realistic manufacturing defects. Therefore, the development of uncertainty quantification methods based on the computational homogenization method is a future direction for the multiscale uncertainty analysis of FRP structures.