数据驱动的复合材料强韧仿生智能设计理论及增材制造方法

Composites have been widely used in aviation, aerospace and other industries. However, their potential advantages of high strength and toughness have not yet been fully explored. One of the main reasons is that the strengthening and toughening mechanism of composites is unclear. And when designing and manufacturing composites, it is quite difficult to consider both strength and toughness, which are conflicting and mutually exclusive mechanical properties. Consequently, the comprehensive mechanical performance of composites is severely jeopardized and this restricts their further applications. ..This project intends to investigate a data-driven intelligent design theory for bio-inspired design of strong and tough composites and their additive manufacturing method. After establishing the mapping database of the design variables and the corresponding mechanical properties of the composites, the common strengthening and toughening mechanism is mined based on machine learning, and is expressed as an optimal design model, which can be conveniently used by the composites designers. The sensitivity analysis of the design variables provides an effective way to uncover the influencing mechanism of the design variables to the mechanical properties of composites. This could be adopted to reduce the design space of the iterative optimization of the composites. Subsequently, efficient direction-controlled additive manufacturing is employed to create complex bio-inspired structures of composites. This research will uncover the strengthening and toughening mechanism and it will free the composites designers from the constraint of priori knowledge. The comprehensive mechanical performance of synthetic composites could be improved as the bio-inspired design and manufacturing of composites hence are truly mimic the strengthening and toughening mechanism, but not just the external shape and morphology of the biological composites. Therefore, the strong and tough bio-inspired composites could be used in more extreme conditions. It is of great theoretical significance and application value.

复合材料在航空航天等领域应用广泛，但其潜在强、韧特性未充分发挥，其中一个重要原因是，复合材料强韧机制不明，导致复合材料设计制造中难以兼顾强度和韧性，严重影响其综合力学性能及在工程中的进一步应用。..针对以上问题，本项目拟开展数据驱动的复合材料强韧仿生智能设计理论及增材制造方法研究。建立复合材料强韧仿生设计变量与力学性能的映射数据库，基于机器学习挖掘生物复合材料的共性强韧机制，并表达为便于设计人员使用的优化设计模型；对复合材料设计变量进行灵敏度分析，揭示设计变量对力学性能的影响规律，提高复合材料迭代优化设计效率；研究精准控向增材制造方法，实现复合材料复杂仿生结构的精准高效制造。通过上述研究，克服复合材料强韧仿生设计机理不明、依赖实验数据和经验模型等不足，实现复合材料仿生设计制造从“仿形”到“仿性”的转变，对提高复合材料的综合力学性能进而拓展其在工程中的应用范围，具有重要的理论意义和工程价值。

复合材料因其轻质、高强等诸多优点，广泛应用于航空航天领域。然而，现有复合材料潜在的强、韧特性尚未充分发掘，复合材料结构强韧机制不明使复合材料结构设计中难以兼顾强度和韧性，特别是韧性通常不足，限制了复合材料在工程中的进一步应用。..针对上述问题，本项目开展复合材料强韧仿生设计与增材制造方法研究。通过分析自然界多种生物材料强韧结构的强韧机制，进而构建复合材料仿生结构的参数化表达，基于有限元模拟仿真获得大量特征-性能数据，建立特征-性能的机器学习模型，在复合材料结构设计空间内对结构力学性能进行预测。通过综合两种或以上生物材料强韧结构，设计了多种混合强韧复合材料结构，进一步提升复合材料结构的力学性能。研究了基于网格化复杂曲面的多引导线自动铺丝路径规划方法，以及基于灵敏度分析的连续纤维增强复合材料3D打印参数优化通用框架，实现了复杂复合材料仿生结构的增材制造。..研究表明，缝线结构、非连续结构、螺旋结构以及多种强韧结构的组合，可以在保持复合材料强度的同时，有效地增加复合材料的韧性，通过机器学习等方法对强韧结构进行精细化设计，可以进一步提高复合材料结构的综合力学性能。自动铺丝技术可实现较大尺寸复合材料构件的自动化制造，而连续纤维3D打印技术可实现结构高度复杂的复合材料结构的制造，基于网格化复杂曲面的多引导线自动铺丝路径规划方法能够高效获得大尺寸复合材料构件的低缺陷铺丝路径，连续纤维增强复合材料3D打印参数优化通用框架则可获得适宜的3D打印参数，支撑复杂复合材料结构的高质量制造。..项目的研究为高强度、高韧性复合材料结构设计及自动化制造提供了方法和工艺支撑。项目研究过程中发表相关领域SCI论文14篇，授权国家发明专利5项，获得软件版权2项（其中1项实现了成果转化），培养硕士研究生5名，达到了项目研究预期。