Rolling bearings, as a type of precision mechanical component, are widely used in modern industrial machinery and equipment. It is of great significance to diagnose bearing faults using reasonable methods during bearing operation. However, in the actual complex and ever-changing environment, the collection of vibration signals often faces challenges such as limited sample sizes, noise interference, and operating condition variations, resulting in low fault diagnosis accuracy. To address the problem of small-sample rolling bearing fault diagnosis under noise interference and variable operating conditions, this paper proposed a meta-learning denoising model based on prototype domain enhancement (Meta-DAE). Firstly, a small-sample fault dataset based on time-frequency diagrams was constructed, and a deep convolutional generative adversarial network was introduced for data preprocessing to generate a pseudo-sample set with a similar distribution. Then, the fault sample set was input into Meta-DAE for adaptive feature extraction. Meta-DAE adopts a prototype domain enhancement strategy to make prototype points of the same category more closely clustered in the embedding space. At the same time, an encoder with noise reduction performance was constructed, and a target function based on prototype domain enhancement and denoising was designed. By fine-tuning the model under small-sample conditions, the noise robustness and classification accuracy of the model were improved. Experimental results of small-sample fault diagnosis under noise interference and variable operating conditions show that, compared to other models, the proposed model demonstrates strong noise robustness. Under -8 dB strong noise interference, the model achieves a classification accuracy improvement of 35.78% to 57.25% using only 10 samples for fine-tuning.

This paper introduces a novel color image encryption algorithm based on a five-dimensional continuous memristor hyperchaotic system (5D-MHS), combined with a two-dimensional Salomon map and an optimized Arnold transform. Firstly, convert the test image to a 2D pixel matrix then processed in blocks, and each block of the pixel matrix is permuted with chaotic sequences generated by 5D-MHS and 2D Salomon map. Then, the permuted image is permuted for three rounds with the optimized Arnold algorithm. Finally, one of the chaotic sequences generated by 5D-MHS is employed to diffuse the permuted image to obtain the final ciphertext image. In this paper, several pseudo-random sequences are generated and mixed in the permutation stage to achieve higher security. The algorithm achieves a key space of 2⁴⁷², the information entropy of the ciphertext image for the color image is 7.9998, number of pixels change rate (NPCR) and unified average changing intensity (UACI) reached 99.6131% and 33.4361%, respectively, and the correlation between pixels is close to 0. The simulation results show that the encryption algorithm is efficient and the key system is secure.

In order to improve the accuracy of the open-circuit fault diagnosis of the double-fed asynchronous wind turbine converter, a fault diagnosis method based on the global adaptive whale optimization algorithm to optimize the extreme learning machine is proposed. Firstly, establish a grid-connected model of doubly-fed induction generator(DFIG), and collect the three-phase line voltage signal under the fault state of the grid-side converter. Secondly, fast Fourier transform is performed on the collected voltage signal, and then the frequency amplitude of the different harmonic components of the three-phase line voltage and the DC component are reconstructed into a feature vector. In order to remove some redundant features, use the neighborhood to maintain the projection pair, the feature vector is dimensionally reduced. Finally, an extreme learning machine optimized by the global adaptive whale optimization algorithm(GAWOA-ELM) is used to diagnose the faults of the converter. Different methods are used to diagnose and analyze converter faults under different signal-to-noise ratios, verifying the effectiveness and robustness of the method proposed in this paper.