Search

The rod-fastened rotor is a critical component of heavy-duty gas turbines, and therefore, it is imperative to investigate its failure mechanisms and develop intelligent diagnostic techniques. The issue of effective cross-domain diagnosis has garnered significant interest considering the changing operating conditions of rod-fastened rotors. To mitigate the effects of limited data samples or high collection costs in the target domain, we introduce a fault diagnosis model named the task-generalization-based graph convolutional network (TG-GCN), which leverages adversarial perturbations within a meta-learning framework to enhance the robustness of prior knowledge incorporated into the graph convolutional network. Initially, the method involves generating adversarial training samples in proximity to the distribution of the source-domain data by optimizing the worst case formulation to enhance generalization to task distribution. Next, the support set and the query set are simultaneously fed into the feature extraction network, yielding corresponding feature vectors. Subsequently, similarities between feature vectors and the degree of dispersion are computed in the task space, serving as connection weights for constructing the graph network model aimed at executing fault diagnosis. The proposed TG-GCN framework employs the introduction of data augmentation during the iteration process, which helps to strengthen the vulnerable components of the network by connecting them with the worst case perturbations. Moreover, we employed a rod-fastened rotor experimental system to elucidate the underlying failure mechanisms and corroborate the superiority of the TG-GCN.

The functional hierarchical structures of the triply periodic minimal surface are receiving much attention in tissue engineering applications due to their lightweight and multifunctionality. However, current functionally graded structure design methods are not friendly to heterogeneous structures containing different orientations and different unit types and often face the problems of insufficient connection in the hybrid regions and low local stiffness. In this paper, an improved gradient structure design method was proposed, which solves the problem of insufficient connection between substructures by constructing hybrid region transition functions. Three improved heterogeneous structures were constructed using Primitive and Gyroid lattices and compared with the unimproved heterogeneous structure. Their mechanical properties, deformation mechanism, and energy absorption capacity were examined by finite element analysis and experiments. The results showed that the proposed design method can effectively solve the problems of insufficient connection and poor bearing capacity in the hybrid region between substructures. This method can not only ensure the full connection of the hybrid regions but also flexibly adjust the mechanical properties and energy absorption capacity as well as effectively expand the application range of the energy absorption. Overall, these findings provide valuable guidelines for designing gradient structures with disordered and hybrid features.

• A discretized nonlinear bolted rotor system was obtained by discretization of the normalized excitation frequency to construct the specific mapping structures for semi-analytical solutions. • The global bifurcation trees are visualized semi-analytically, the motion cascade, stability, and bifurcation type of P-1→P-2, P-3→P-6→P-9 were discussed in detail. • The independent branches of P-3(m + 1), (m = 0, 1, 2) motions were presented, the independence of the bifurcation tree in partial higher order periodic motions was discovered. • The semi-analytical bifurcation trees provide all possible motions that the nonlinear rotor system may exhibited, which explain the bistable/multi-stable phenomenon of the nonlinear rotor system theoretically. • Compared to the numerical solutions, the semi-analytical ones provide the complete bifurcation paths and independent periodic motions of the nonlinear bolted rotor system, which help ones to understand the complex nonlinear motion mechanisms and stability comprehensively. The bolted structures have the advantages of high specific strength and easy maintenance, which are widely used in rotating machines such as gas turbines and aero-engines. Meanwhile, the complex nonlinear motions and bifurcations are easily induced by the nonlinear stiffness caused by the bolted structure in rotor systems, but these characteristics cannot be fully revealed only by numerical results. In this research, an analytical model of a bolted rotor system considering contact nonlinearity is established. The global bifurcation trees and semi-analytical periodic motions of the system are presented by the implicit discrete mapping method, which are also compared to the numerical solutions for validation. Multi-stable phenomena of independent periodic motions P-3(m +1), (m = 0, 1, 2) are discovered, the bifurcation jumping phenomena between the higher-order and lower-order motions are revealed, the occurrence of unstable motions is indicated in Neimark bifurcation judged by eigenvalues. In addition, the independence of the higher-order periodic motions in bifurcation tree is proved. Finally, the periodic motions of the nonlinear rotor system are numerically predicted by harmonic amplitudes and phase characteristics. These findings provide clear and complete solutions for understanding and presenting the nonlinear dynamic characteristics of bolted rotor systems considering the contact effect, which have great potential in guiding the dynamic design and parameter adjustment of practical rotating machines. [ABSTRACT FROM AUTHOR]

The design of scaffolds for tissue engineering has to consider two trade-off properties: mechanical and mass-transport properties. This is particularly true for additively manufactured scaffolds with the structures of minimal surfaces, and notably, the influence of the surface curvature of the structure on the mechanical and mass-transport properties remains unclear. This work presents our study on the scaffolds designed with the structure of triply periodic minimal surfaces (TPMS), with a focus on discovering the influence of surface curvature on the mechanical response and the mass-transport property or permeability of the scaffolds. Based on the entropy weight fuzzy comprehensive evaluation method, a model representative of both mechanical and permeable properties of scaffolds was developed; scanning electron microscopy (SEM) and finite element analysis (FEA) were also used to reveal the influence mechanism of curvature on structural fracture and deformation behavior. AlSi10Mg samples of scaffolds designed with different surface curvatures were manufactured using selective laser melting (SLM), and their mechanical and permeable properties were examined and characterized by both experiments and simulations. Our results illustrate that at the same porosity, the more concentrated the curvature distribution of the same type of unit, the better trade-off mechanical and mass-transport properties the scaffolds have. Particularly, at the porosity of 55%, the compressive elastic modulus and permeability of the Dte structure are increased by 2.03 times and 1.95 times compared with the Diamond unit, respectively. The fusion structure can greatly improve permeability performance at the cost of mechanical properties. Our results also show that porosity has the greatest influence on mechanical and permeable properties, followed by the surface curvature. The study illustrates that the surface curvature has a significant influence on the mechanical and permeable properties of scaffolds, and that the developed scaffold performance evaluation scheme is an effective means for the optimization and evaluation of scaffold performance.