Materials-by-design to develop high performance composite materials is often computational intractable due to the tremendous design space. Here, a deep operator network (DeepONet) is presented to bridge the gap between the material design space and mechanical behaviors. The mechanical response such as stress or strain can be predicted directly from material makeup efficiently, and a good accuracy is observed on unseen data even with a small amount of training data. Furthermore, the proposed approach can predict mechanical response of complex materials regardless of geometry, constitutive relations, and boundary conditions. Combined with optimization algorithms, the network offers an efficient tool to solve inverse design problems of composite materials.

In recent years, hydraulic fracturing techniques have been widely used in extracting unconventional reservoir resources. During multicluster fracturing process, the stress shadowing effect can lead to a nonuniform distribution of fracturing fluid. In this paper, a two-dimensional multiple fracture propagation model is developed based on phase field method considering the distribution of fracturing fluid within each fracture during fracturing process. The distribution of fracturing fluid injected into each fracture is calculated through solving governing equations of perforation friction as well as wellbore friction. Numerical results demonstrate that the natural fractures in the reservoir have a significant influence on the distribution of injected fluid and the morphology of fracture network. In addition, higher injection rates and higher fluid viscosity stimulations are in favor of inducing uniform distribution of injected fluid into each cluster. The presented numerical model provides an effective tool to extend our understandings in generating fracture networks.