Machine learning (ML) and Deep learning (DL) are increasingly pivotal in the design of advanced metamaterials, seamlessly integrated with material or topology optimization. Their intrinsic capability to predict and interconnect material properties across vast design spaces, often computationally prohibitive for conventional methods, has led to groundbreaking possibilities. This paper introduces an innovative machine learning approach for the optimization of acoustic metamaterials, focusing on Multiresonant Layered Acoustic Metamaterial (MLAM), designed for targeted noise attenuation at low frequencies (below 1000 Hz). This method leverages ML to create a continuous model of the Representative Volume Element (RVE) effective properties essential for evaluating sound transmission loss (STL), and subsequently used to optimize the overall topology configuration for maximum sound attenuation using a Genetic Algorithm (GA). The significance of this methodology lies in its ability to deliver rapid results without compromising accuracy, significantly reducing the computational overhead of complete topology optimization by several orders of magnitude. To demonstrate the versatility and scalability of this approach, it is extended to a more intricate RVE model, characterized by a higher number of parameters, and is optimized using the same strategy. In addition, to underscore the potential of ML techniques in synergy with traditional topology optimization, a comparative analysis is conducted, comparing the outcomes of the proposed method with those obtained through direct numerical simulation (DNS) of the corresponding full 3D MLAM model. This comparative analysis highlights the transformative potential of this combination, particularly when addressing complex topological challenges with significant computational demands, ushering in a new era of metamaterial and component design.

This study presents the results of fracture toughness tests conducted on specimens obtained by additive manufacturing techniques, specifically using low-force stereolithography. The samples were manufactured from a transparent 3D printing material for biocompatible applications, the so-called BioMed Durable Resin, which is a Formlabs-patented polymer material that simulates the strength and rigidity of polyethylene. The selected toughness tests in this context were performed following the ASTM D5045-99 guidelines. All tests were conducted under controlled laboratory conditions at 23 °C and 50% relative humidity, ensuring adherence to the standard and the replicability of the experimental results. To investigate the influence of printing plane orientation, specimens were produced at three printing orientation angles (0, 45, and 90 degrees). These angles were selected to provide a comprehensive evaluation of the anisotropy effects in the material. They cover both extreme orientations (0° and 90°) and include an intermediate value (45°), allowing us to assess variations in mechanical behavior across a representative range of printing orientations, consistent with prior research in the field. The experimental tests yielded data on the crack resistance and energy release rate for each angle of orientation. There are various implications of the findings, beyond materials engineering, for applications in biomedicine. Indeed, this same approach opens the door to new research methods for manufacturing certified biocompatible materials from such durable resins. Finally, complementary issues such as related medical applications have been slightly addressed for future work, since biomedicine innovation clusters can contribute to accelerating growth in this crucial field for productive sector activity and the local business environment.

Metaconcrete (MC) is a class of engineered cementitious composites that integrates locally resonant inclusions to filter stress waves. While the dynamic benefits are well established, the effect of resonator content and geometry on static compressive resistance remains unclear. This study develops the first two-dimensional mesoscale finite-element model that explicitly represents steel cores, rubber coatings, and interfacial transition zones to predict the quasi-static behavior of MC. The model is validated against benchmark experiments, reproducing the 56% loss of compressive strength recorded for a 10.6% resonator volume fraction with an error of less than 1%. A parametric analysis covering resonator ratios from 1.5% to 31.8%, diameters from 16.8 mm to 37.4 mm, and coating thicknesses from 1.0 mm to 4.2 mm shows that (i) strength decays exponentially with volumetric content, approaching an asymptote at ≈20% of plain concrete strength; (ii) larger cores with thinner coatings minimize stiffness loss (<10%) while limiting strength reduction to 15–20%; and (iii) material properties of the resonator have a secondary influence (<6%). Two closed-form expressions for estimating MC strength and Young’s modulus (R² = 0.83 and 0.94, respectively) are proposed to assist with the preliminary design. The model and correlations lay the groundwork for optimizing MC that balances vibration mitigation with structural capacity.