This research direction serves as a core foundational pillar within the team's broader research framework on cement-based materials and structures. It focuses on multi-scale numerical simulation — spanning atomic-molecular, microscopic, mesoscopic, and macroscopic levels — and artificial intelligence (AI)-assisted design technologies. The overarching aim is to overcome the limitations of conventional experimental approaches, which are time-consuming, costly, and often unable to reveal underlying microscopic mechanisms. Through systematic investigation of the microstructural formation mechanisms and macroscopic performance evolution of civil engineering materials, this direction provides theoretical foundations and digital tools for the development of novel high-performance cement-based materials, the service life prediction of concrete structures under harsh environments, and performance enhancement strategies — ultimately driving the transition of civil engineering materials from "empirical trial-and-error" to "science-based design."

Core Research Areas

1. Multi-Scale Numerical Simulation of Cement-Based Materials

Atomic-scale molecular dynamics (MD) and Monte Carlo (MC) simulations are conducted to analyze the atomic-level structural characteristics of cement hydration products, including C-S-H gel and calcium hydroxide, and to elucidate the microscopic nature of their mechanical behavior, transport properties, and hydration reactions. At the mesoscale, stochastic aggregate models and microstructural evolution models for concrete are developed to simulate the full process of internal damage initiation, propagation, and coalescence. Quantitative relationships between microstructural parameters — such as porosity, pore size distribution, and degree of hydration — and macroscopic mechanical and durability properties are established, enabling accurate prediction of material performance.

2. AI-Assisted Intelligent Design of Civil Engineering Materials

Machine learning and deep learning algorithms are integrated with materials genomics concepts to build predictive models for key performance indicators of cement-based materials, including strength, durability, and workability. Intelligent design platforms are developed for novel materials such as solid waste-based cementitious binders and ultra-ductile cement-based composites, enabling rapid formulation optimization and inverse design. By combining experimental data with numerical simulation results, this approach substantially shortens the material development cycle and reduces research and development costs.

3. Numerical Simulation of Concrete Performance Evolution under Harsh Environments

Damage evolution processes in concrete subjected to single and coupled deterioration mechanisms — including chloride ion ingress, sulfate attack, freeze-thaw cycling, and carbonation — are simulated. Full service life prediction models for concrete that account for environment–material–structure interactions are established, providing support for the durability design of major infrastructure such as marine engineering structures, rail transit systems, and structures in high-altitude cold regions. The evolution of interfacial bond behavior between repair materials and existing concrete is also investigated, providing a theoretical basis for the optimization of structural repair and strengthening schemes.

4. Computational Mechanics of Ultra-Ductile Cement-Based Materials

Mesoscale mechanical models for fiber-reinforced cement-based composites are developed to elucidate the interfacial interaction mechanisms between fibers and the matrix, as well as the toughening mechanisms involved. The mechanical responses and failure modes of ultra-ductile materials under complex loading conditions — including tension, bending, and impact — are simulated. The type, dosage, aspect ratio, and spatial distribution of fibers are optimized to guide the engineering application of high-performance ultra-ductile cement-based materials.

Representative Research Foundation and Achievements

• Principal Investigator of multiple related research projects, including NSFC General Program and Young Scientist Fund projects and Hunan Provincial Natural Science Foundation grants, conducting systematic fundamental research on multi-scale simulation and intelligent design of cement-based materials.

• Published more than 50 SCI papers in leading international journals such as Cement and Concrete Composites and Construction and Building Materials, with a total citation count exceeding 2,000; multiple papers have been selected as ESI Highly Cited Papers.

• Developed several software tools for performance prediction and intelligent design of cement-based materials; holder of 5 authorized national invention patents in this area.

• Research outcomes have been successfully applied to the development and performance optimization of high-speed railway track slab concrete, marine engineering concrete, and solid waste-based ecological building materials, yielding significant economic and social benefits.

Research Outlook

Future work will further deepen the integration of computational simulation and artificial intelligence technologies, expand the application of digital twin technology to the full life-cycle management of civil engineering materials and structures, and promote the digital and intelligent transformation of the building materials industry — providing core technical support for the high-quality development of infrastructure construction in China.