This work presents the results of an experimental investigation of the mechanical properties of polyetheretherketone (PEEK) specimens additively manufactured (AM) by using fused filament fabrication with different printing parameters and subjected to postprocessing heat treatment. Standard and compact tension samples were manufactured with a different infill angle using 0.4 mm and 0.6 mm nozzle diameters. Some of the samples were subjected to heat treatment at 220 °C after manufacturing. Tensile tests were conducted to determine the values of elastic modulus, tensile strength, as well as mode-I fracture toughness and critical strain energy release rate. Tensile properties of single-thread and as-delivered filaments were also studied. It was concluded that heat treatment significantly improved the elastic properties, tensile strength and fracture toughness of the AM PEEK samples: the fracture resistance increased by 33 to 45% depending on the stacking order, while the tensile strength increased by some 45–65%, with the elasticity modulus grown by up to 20%. Strain fields induced in specimens by crack propagation were captured with a digital image correlation technique and compared with results of numerical simulations implemented with the extended finite-element method (XFEM). Conclusions on the optimal parameters of 3D printing of PEEK were made.

Generative adversarial neural networks with a variational autoencoder (VAE-GANs) are actively used in the field of materials design. The synthesis of random structures with nonrepeated geometry and predetermined mechanical properties is important for solving various practical problems. Geometric parameters of such artificially generated random structures can vary within certain limits compared to the training dataset, causing unpredicted fluctuations in their resulting mechanical response. This study investigates the statistical variability of mechanical and morphological characteristics of random 3D models reconstructed from 2D images using a VAE-GAN neural network. A combined multitool method employing different mathematical and statistical instruments for comparison of the reconstructed models with their corresponding originals is proposed. It includes the analysis of statistical distributions of elastic properties, morphometric parameters, and stress values. The neural network was trained on two datasets, containing models created based on Gaussian random fields. Statistical fluctuations of the mechanical and morphological parameters of the reconstructed models are analyzed. The deviation of the effective elastic modulus of the reconstructed models from that of the original ones was less than 5.7% on average. The difference between the median values of ligament thickness and distance between ligaments ranged from 3.6 to 6.5% and 2.6 to 5.2%, respectively. The median value of the surface area of the reconstructed geometries was 4.6–8.1% higher compared to the original models. It is thus shown that mechanical properties of the NN-generated structures retain the statistical variability of the corresponding originals, while the variability of the morphology is highly affected by the training set and does not depend on the configuration of the input 2D image.