AI‐Powered Inverse Design of High‐Performance Unidirectional CFRP Composites: Breaking the Compressive‐Tensile Trade‐Off via Machine Learning

While unidirectional carbon fiber reinforced polymer (UD-CFRP) composites possess high specific strength and stiffness, their compressive strength falls significantly below their tensile strength, limiting their load-bearing capacity as primary structural components in engineering applications. To effectively develop UD-CFRPs with high compression-to-strength ratios, we devised a systematic AI-enabled framework integrating machine learning and genetic algorithms for inverse design. This work presents an ensemble model composed of multiple machine-learning models, enhancing the robustness of mechanical property predictions for UD-CFRPs. The proposed genetic algorithm designs the resin matrix parameters that meet the diverse performance requirements of UD-CFRPs. Our framework has been validated against experimental data, showcasing superior computational efficiency compared to traditional forward design methods. This research strategy not only circumvents the inefficiencies of trial-and-error experimentation but also provides the first case of AI-driven inverse design of composites, with the potential to extend its application to the design of various composites.