Rolling bearings, as critical components of rotating machinery, directly affect the reliability and efficiency of the system. Due to extended operation under high load, harsh environmental conditions, and continuous use, bearings become more susceptible to failure, leading to a higher likelihood of malfunction. To prevent sudden failures, reduce downtime, and optimize maintenance strategies, early and accurate diagnosis of rolling bearing faults is essential. Although existing methods have achieved certain success in processing acoustic and vibration signals, they still face challenges such as insufficient feature fusion, inflexible weight allocation, lack of effective feature selection mechanisms, and low computational efficiency. To address these challenges, we propose a dynamic weighted multimodal fault diagnosis model based on the fusion of acoustic and vibration information. This model aims to enhance feature fusion, dynamically adapt to signal characteristics, optimize feature selection, and reduce computational complexity. The model incorporates an adaptive fusion method based on a multi-branch convolutional structure, enabling unified processing of both acoustic and vibration signals. At the same time, a cross-modal dynamic weighted fusion mechanism is employed, allowing the real-time adjustment of weight distribution based on signal characteristics. By utilizing an attention mechanism for dynamic feature selection and weighting, the robustness of classification is further improved. Additionally, when processing acoustic signals, a depthwise separable convolutional network is used, effectively reducing computational complexity. Experimental results demonstrate that our method significantly outperforms other algorithms in terms of convergence speed and final performance. Additionally, the accuracy curve during training showed minimal fluctuation, reflecting higher robustness. The model achieved over 99% diagnostic accuracy under all signal-to-noise ratio (SNR) conditions, showcasing exceptional robustness and noise resistance in both noisy and high-SNR environments. Furthermore, its superiority across different data scales, especially in small-sample learning and stability, highlights its strong generalization capability.

Predicting the remaining useful life (RUL) of wind turbine generator rolling bearings can effectively prevent damage to the transmission chain and significant economic losses resulting from sudden failures. However, the working conditions of generator bearings are variable, and the collected run-to-failure data combine multiple working conditions, which significantly impacts the accuracy of model predictions. To solve the problem, a local enhancement temporal convolutional network with multistage degenerate distribution matching based on domain adaptation (MDA-LETCN) is proposed, extracting degradation features of wind turbine generator bearings and predicting their remaining service life in composite working conditions. This method first utilizes the local enhancement temporal convolutional network (LETCN) to extract time series features and used the K-means method for unsupervised division of the degradation status of rolling bearings. Secondly, the multistage degradation stage distribution matching (MDSDM) module is proposed to learn domain-invariant temporal features at different stages of bearing degradation under composite working conditions. Finally, the model is transferred to the target bearing using some health data that are easily available from the target bearing to solve the problem of individual differences in the degradation of generator bearings in different wind turbines. Comparative experiments were conducted using actual wind farm data, and the results showed that MDA-LETCN has high prediction accuracy.

The clearance between the inner ring (IR) of rolling bearing and rotor is directly proportional to the temperature rise. The increased friction between the IR and rotor, as well as changes in the frequency of rolling bearing defects, and the accuracy of bearing health monitoring decreases. Existing research cannot effectively solve this problem. In response to the above issues, in this paper, the clearance between the bearing IR and the rotor and the internal clearance of bearing were considered, and the clearance was considered as the motion boundary of the rotor inside the IR, dynamic models of the bearing IR, outer ring (OR), and rolling element defects were established, and the trend of rolling bearing defects frequencies with temperature was analyzed, then the rationality of the dynamic models were verified through experimental methods. The research results show that the friction between the rotor and the IR increases with increasing temperature, and the speed of the IR decreases accordingly. The characteristic frequencies of defects in the IR, OR, and rolling element of rolling bearings have decreased by approximately 6.42%, 8.07%, and 6.93% compared to the theoretical values without considering temperature, respectively. This indicates that when identifying the frequency of rolling bearing defects at different temperatures, the frequency deviation caused by the increase in clearance between the IR and the rotor should be considered. The research results may serve as an important basis for fault diagnosis and condition monitoring of rolling bearings.

A planar Antipodal Vivaldi Antenna (AVA) with a director for ultra-wideband (UWB) applications is proposed. The director is a triangular structure. It guides electromagnetic waves emitted from the aperture of the antenna to the edge of the conductive director with a small reflection and good transmission characteristics compared to the traditional AVAs. The proposed antenna is designed and simulated successfully. And the simulated results show that it possesses a wide bandwidth from 3.1 to 11 GHz for S₁₁ <; - 10 dB; and the peak gain of the antenna is apparently increased in the range of 8.5-11 GHz, which validates the feasibility of this novel design.