Machine Learning Based Surrogate Model for Bioinspired Gradient Composites

Biological materials evolve extraordinary protective systems to survive the competitive environment, thus having outstanding mechanical properties and multifunctionality. For instance, bone and bamboo are both structural bio-composites with superior mechanical properties. In recent decades, understanding the properties of bio-inspired composites has been a popular subdiscipline of material science. The idea of bioinspiration is to learn from biological structural materials and apply novel structural design strategies for the development of composites with superior mechanical properties. In this research, the composite structures are inspired by the topology of bone and the density distribution of bamboo. In order to explore the extreme huge design space of structural materials, we developed a machine learning based surrogate model by using a combination of principal component analysis (PCA) and deep neural networks (DNN) for predicting the entire stress-strain behavior of the bone- and bamboo-inspired composite structures. Our results show that the surrogate model is accurate and efficient for investigating the design space and the proposed approach in this work can be extended to other composite structures to further accelerate material design and optimization.