Carbon fiber fabric reinforced composite laminates are widely used in the automotive and aerospace components, which are prone to suffering low velocity impacts. In this paper, helicoidal layups of fabrics inspired by the Bouligand type structure of the dactyl clubs of mantis shrimp are proposed to improve the impact resistance of carbon fiber fabric reinforced composite laminates. Low velocity impact tests and finite element simulation are carried out to investigate the effect of the rotation angle of helicoidal layups on the impact damage behaviors of composite laminates, including impact force response, energy absorption characteristics and damage mechanism. Results show that the simulation results of impact force–time response, absorbed energy–time response, and damage characteristics show good agreements with the experimental results. With the increase in impact energy, the maximum value of impact force, the absorbed energy and the energy absorption ratio for all specimens are all increased. Under all impact energies, the impact damage of specimens with helicoidal layups are lower than that of specimen QI1 (rotation angle of 0°), indicating that the helical layup of woven carbon fabric can sufficiently enhance the impact resistance of the composite material. Furthermore, the impact resistance of specimen HL2 (rotation angle of 12.8°) is the best, because it demonstrates the lowest impact damage and highest impact force under all energies. This work provides a bionic design guideline for the high impact performance of carbon fiber fabric reinforced composite laminate.

Cross-condition bearing fault diagnosis suffers from severe performance degradation due to domain shift across different operating conditions, especially when only a few labeled target-domain samples are available. To address this challenge, this paper proposes a progressive attention-guided two-stage transfer learning framework for few-shot bearing fault diagnosis across different fixed operating points. First, the raw time-domain vibration signals are fused with frequency-domain representations extracted by short-time Fourier transform (STFT) to enhance fault feature representation. Then, a progressive attention-guided feature learning strategy is developed by integrating dual efficient channel attention (ECA) modules into a deep one-dimensional convolutional neural network (1D-CNN), enabling the network to adaptively emphasize fault-sensitive features while suppressing redundant information. Subsequently, a two-stage transfer learning strategy is designed, consisting of transferable feature learning from the source domain and few-shot adaptation to the target domain. During target-domain adaptation, key feature extraction layers are frozen, and a sample-balanced optimization mechanism is introduced to alleviate the dominance of source-domain samples during joint training. Experimental results on the Case Western Reserve University (CWRU) bearing dataset demonstrate that the proposed method achieves an average accuracy of 99.96% across three cross-condition transfer tasks. Furthermore, experiments conducted on a self-built shaft system dataset show that the proposed method achieves an average accuracy of 87.11% under three representative transfer scenarios. The results verify that the proposed framework effectively mitigates domain shift and enables accurate bearing fault diagnosis with limited labeled target-domain samples.