人工智能在复合材料研究中的应用

复合材料以其轻质高强高模、可设计性强等优点成为结构轻量化的重要用材. 然而, 随着复合材料组分、结构以及性能需求的日益复杂化, 以实验观测、理论建模和数值模拟为主体的传统研究范式, 在复合材料力学性能分析、设计和制造等方面遇到了新的科学问题与技术瓶颈. 其中, 实验观测不足、理论模型缺乏、数值分析受限、结果验证困难等问题在一定程度上制约了先进复合材料在面向未来工程领域中应用的发展. 人工智能方法以数据驱动的模型替代传统研究中的数学力学模型, 直接由高维高通量数据建立变量间的复杂关系, 捕捉传统力学研究方法难以发现的规律, 在复杂系统的分析、预测、优化方面拥有与生俱来的优势. 而通过人工智能赋能来寻求复合材料中传统研究方法所面临难题的新的解决方案, 目前已成为复合材料研究领域的发展趋势. 本文综述并评价了人工智能方法在复合材料性能预测、优化设计、制造检测及健康监测等方面的研究进展, 并对未来发展方向进行了探讨和展望. 

Application of artificial intelligence in composite materials

Composite materials have become the major materials of light-weight structure due to their light weight, high strength, high modulus, and strong designability. However, as the component, structure and requirements of capability become increasingly complex, traditional research methods based on experiment, theoretical modeling and numerical simulation meet lots of new problems in the properties prediction, design optimization, manufacturing and processing of composite materials. Insufficient experimental observation, lacking theoretical model, constrained numerical simulation and difficult conclusion validation have seriously restricted the development of advanced composite materials in the future-oriented engineering. Instead of the mathematical models used by mechanics, data-driven models are used in the Artificial Intelligence. It directly establishes the complex relationship between variables from high-dimensional and high-throughput data, then captures the laws that are difficult to be discovered by traditional mechanical method, shows natural advantages in simulation, prediction, optimization in complex systems. It has become the development trend in the field of composite materials to find new solutions to the problems faced by traditional methods through Artificial Intelligence. In this paper, the status of properties prediction, design optimization, manufacturing and health monitoring is introduced. The future development direction of this field is discussed.