AI-assisted Design, 3d Printing, and Evaluation of Architecture Polymer-concrete Composites (APCC) With High Specific Flexural Strength and High Specific Toughness

Architectured polymer-concrete composite (APCC) is a promising structural material with high mechanical performance. Optimizing the design of APCC for high flexural strength, high toughness, and lightweight remains a challenge. This paper presents a machine learning-based optimization framework to design APCC with high specific flexural strength and high toughness. The proposed framework integrates sequential surrogate modeling, Latin hypercube sampling, Lion Pride Optimization, and finite element analysis to predict and optimize the flexural properties of APCC. The framework was implemented into designing APCC beams, which were fabricated via 3D printing and tested under flexural loads. Results show that the designed APCC achieved high specific flexural strength and high specific toughness. The architecture of APCC arrests crack propagation and promotes energy dissipation. Parametric studies were performed to evaluate the effect of the key design variables of APCC beams on the flexural properties. This research advances the knowledge and development of APCC.