The transfer learning method, based on unsupervised domain adaptation (UDA), has been broadly utilized in research on fault diagnosis under variable working conditions with certain results. However, traditional UDA methods pay more attention to extracting information for the class labels and domain labels of data, ignoring the influence of data structure information on the extracted features. Therefore, we propose a domain-adversarial multi-graph convolutional network (DAMGCN) for UDA. A multi-graph convolutional network (MGCN), integrating three graph convolutional layers (multi-receptive field graph convolutional (MRFConv) layer, local extreme value convolutional (LEConv) layer, and graph attention convolutional (GATConv) layer) was used to mine data structure information. The domain discriminators and classifiers were utilized to model domain labels and class labels, respectively, and align the data structure differences through the correlation alignment (CORAL) index. The classification and feature extraction ability of the DAMGCN was significantly enhanced compared with other UDA algorithms by two example validation results, which can effectively achieve rolling bearing cross-domain fault diagnosis.

A flexible eddy current probe can detect the defects in components with a complex structure due to its bendability and adaptation to the surface of the tested specimen for nondestructive testing. However, the traditional, flexible eddy current probe with circular and rectangular coils may neglect small cracks along certain special orientations. In this paper, the authors introduce the theory of fractal geometry to guide the design of flexible probes. An annular fractal curve is first proposed, and its system of iterative functions and geometric characteristics are then studied. Following this, six flexible differential excitation-based eddy current probes are designed based on the annular fractal curve. Simulations and experiments are used to compare the performance of the probes to identify the optimum one for inspecting defects. The results show that the M6K3 probe delivers the best performance in terms of the identification of cracks with lengths shorter than half the length of the coil of the probe.

The noise effects on a homeostatic regulation of sleep-wake cycles’ neuronal mathematical model determined by the hypocretin/orexin and the local glutamate interneurons spatiotemporal behaviors are studied within the neighborhood of stimulus threshold in this work; the neuronal noise added to the stimulus, the conductance, and the activation variable of the modulation function are investigated, respectively, based on a circadian input skewed in sine function. The computer simulation results suggested that the increased amplitude of external current input will lead to the fact that awakening time is advanced but the sleepy time remains the same; for the bigger conductance and modulation noise, the regulatory mechanism of the model sometimes will be collapsed and the coupled two neurons of the model show very irregular activities; the falling asleep or wake transform appears at nondeterminate time.