Search

With the development of precision sensing instruments and data storage devices, the fusion of multi-sensor data in gearbox fault diagnosis has attracted much attention. However, existing methods have difficulty in capturing the local temporal dependencies of multi-sensor monitoring information, and the inescapable noise severely decreases the accuracy of multi-sensor information fusion diagnosis. To address these issues, this paper proposes a fault diagnosis method based on dynamic graph convolutional neural networks and hard threshold denoising. Firstly, considering that the relationships between monitoring data from different sensors change over time, a dynamic graph structure is adopted to model the temporal dependencies of multi-sensor data, and, further, a graph convolutional neural network is constructed to achieve the interaction and feature extraction of temporal information from multi-sensor data. Secondly, to avoid the influence of noise in practical engineering, a hard threshold denoising strategy is designed, and a learnable hard threshold denoising layer is embedded into the graph neural network. Experimental fault datasets from two typical gearbox fault test benches under environmental noise are used to verify the effectiveness of the proposed method in gearbox fault diagnosis. The experimental results show that the proposed DDGCN method achieves an average diagnostic accuracy of up to 99.7% under different levels of environmental noise, demonstrating good noise resistance.

Fault diagnosis and health condition monitoring have always been critical issues in the engineering research community [...]

Advanced sensing, fault diagnosis, and structural health management are important parts of the maintenance strategy of modern industries [...]

When considering the transition probability matrix of ordinal patterns, transition permutation entropy (TPE) can effectively extract fault features by quantifying the irregularity and complexity of signals. However, TPE can only characterize the complexity of the vibration signals at a single scale. Therefore, a multiscale transition permutation entropy (MTPE) technique has been proposed. However, the original multiscale method still has some inherent defects in the coarse-grained process, such as considerably shortening the length of time series at large scale, which leads to a low entropy evaluation accuracy. In order to solve these problems, a composite multiscale transition permutation entropy (CMTPE) method was proposed in order to improve the incomplete coarse-grained analysis of MTPE by avoiding the loss of some key information in the original fault signals, and to improve the performance of feature extraction, robustness to noise, and accuracy of entropy estimation. A fault diagnosis strategy based on CMTPE and an extreme learning machine (ELM) was proposed. Both simulation and experimental signals verified the advantages of the proposed CMTPE method. The results show that, compared with other comparison strategies, this strategy has better robustness, and can carry out feature recognition and bearing fault diagnosis more accurately and with improved stability.

Hand detection is a crucial pre-processing procedure for many human hand related computer vision tasks, such as hand pose estimation, hand gesture recognition, human activity analysis, and so on. However, reliably detecting multiple hands from cluttering scenes remains to be a challenging task because of complex appearance diversities of dexterous human hands (e.g., different hand shapes, skin colors, illuminations, orientations, and scales, etc.) in color images. To tackle this problem, an accurate hand detection method is proposed to reliably detect multiple hands from a single color image using a hybrid detection/reconstruction convolutional neural networks (CNN) framework, in which regions of hands are detected and appearances of hands are reconstructed in parallel by sharing features extracted from a region proposal layer, and the proposed model is trained in an end-to-end manner. Furthermore, it is observed that the generative adversarial network (GAN) could further boost the detection performance by generating more realistic hand appearances. The experimental results show that the proposed approach outperforms the state-of-the-art on public challenging hand detection benchmarks.