The purpose of this study is to demonstrate the capabilities of convolutional and generative adversarial networks in problems related to mechanics, in particular, in the design of monolithic slabs on a base. In this paper, for the first time, an approach based on the use of a voxel description of the object under study is proposed. In a number of cases at the design stage, the presence of technological holes of various shapes is envisaged, the slab surface may have a complex geometric shape. Determination of the stress-strain state in a closed form in such cases is very labor-intensive or unattainable. The paper highlights the promising potential of 3D Convolutional and 3D Generative adversarial neural networks in predicting the magnitudes of shrinkage stresses and displacements. The ultimate goal of the research is to create a slab design method that combines the advantages of theoretical models, finite element methods, and biosimilar technologies.
Conference topics reveals:
1.      Possibility of convergence of mechanics and neurotechnology.
2.      The possibility of using “soft computing” with the application of deep learning in design-related tasks.
3.      Advantages of convolutional neural networks (CNN), generative neural networks (GNN) in predicting forced displacements and stress-state condition (SSC) in slabs on the base when there is a deficit (or absence) of initial data on SSC in the zone of technological holes.
4.      The method of slab database creation, the method of data coding for neural networks training with the subsequent integration with already available and planned data of subsequent stages.

Keywords: Convolutional neural networks, generative adversarial neural networks, neurons, slabs on base, voxels, shrinkage.