Convolution neural network (CNN)-based fault diagnosis methods have been widely adopted to obtain representative features and used to classify fault modes due to their prominent feature extraction capability. However, a large number of labeled samples are required to support the algorithm of CNNs, and, in the case of a limited amount of labeled samples, this may lead to overfitting. In this article, a novel ResNet-based method is developed to achieve fault diagnoses for machines with very few samples. To be specific, data transformation combinations (DTCs) are designed based on mutual information. It is worth noting that the selected DTC, which can complete the training process of the 1-D ResNet quickly without increasing the amount of training data, can be randomly used for any batch training data. Meanwhile, a self-supervised learning method called 1-D SimCLR is adopted to obtain an effective feature encoder, which can be optimized with very few unlabeled samples. Then, a fault diagnosis model named DTC-SimCLR is constructed by combining the selected data transformation combination, the obtained feature encoder and a fully-connected layer-based classifier. In DTC-SimCLR, the parameters of the feature encoder are fixed, and the classifier is trained with very few labeled samples. Two machine fault datasets from a cutting tooth and a bearing are conducted to evaluate the performance of DTC-SimCLR. Testing results show that DTC-SimCLR has superior performance and diagnostic accuracy with very few samples.

Numerical simulations are employed to investigate the sound generated by flow over two traveling wavy foils in a side-by-side arrangement by an immersed-boundary-method-based hybrid approach. The effects of Strouhal number (St), phase difference, and lateral spacing (S) between the foils on the flow performance and the sound pressure field are examined. The results imply that the sound produced by a single foil is dominated by the lift dipole and that the low-amplitude-high-frequency (LA-HF) foil can achieve higher thrust and higher sound pressure compared to the high-amplitude-low-frequency (HA-LF) foil. For the two side-by-side foils (i.e., an in-phase and anti-phase foil system), the sound pressure fields exhibit distinct features. Specifically, a dipole-like pattern appears during in-phase motion, whereas a monopole-like pattern exists during anti-phase motion. Moreover, the magnitude of the sound pressure increases slightly with increasing S in the in-phase case. However, the sound pressure decreases rapidly when S < 0.7L (foil length) and then remains nearly unchanged when S > 0.7L in the anti-phase case. Furthermore, the anti-phase foil system could improve thrust by increasing power consumption and could generate lower sound pressure compared to the in-phase one due to the distinct differences in wake patterns. The present work is expected to improve the understanding of sound-generation mechanisms of fish-like locomotion and collective motion for relevant biomimetic underwater vehicles.