In conditions of foggy weather, challenges such as low light, blurred imagery, and dense fog that obscures target objects are prevalent. Moreover, computing resources are limited on edge devices. To tackle these challenges, a novel real-time detection algorithm GMS-YOLO for pedestrians and vehicles is proposed based on YOLOv10, which overcomes the semantic bottleneck of the model and enhances its detection performance. A novel Ghost Multi-Scale Convolution (GMSConv) module is constructed, serving as the Ghost Multi-Scale feature extraction backbone network (GMS-Net). The Shape Consistent Intersection over Union (SCIoU) is introduced as the localization loss function, which takes into account the influence of the attributes of the regression box in the loss computation. Additionally, a Compensatory Consistency Matching Metric (CCMM) formula is designed to reduce the sensitivity of the original metric to IoU and regression scores. The GMS-YOLO algorithm has a lightweight structure, achieving FPS of 94 and 92 during the detection phase at the ’n’ and ’s’ sizes, respectively. Furthermore, we have deployed the model on Jetson Nano hardware, and the inference speed is also quite encouraging. We validated the effectiveness of the algorithm on the Foggy Cityscapes, RTTS, VOC2007-fog, and VOC2012-fog datasets. Experimental results indicate that GMS-YOLO outperforms the baseline model, with a mean Average Precision (mAP) improvement of 6.3% and 5.5% for the ’n’ and ’s’ scales, respectively. Consequently, the proposed GMS-YOLO algorithm not only demonstrates superior detection performance but also maintains a relatively low model complexity, significantly enhancing the efficiency and accuracy of object detection tasks in foggy environments. The source code for our algorithm is available at: https://github.com/Fwdchina/GMS-YOLO.

Due to the fault vibration signal of the rolling bearing is greatly interfered by the background noise, the fault features are easily submerged and result in a low fault diagnosis accuracy. A novel fault diagnosis method of rolling bearing is proposed based on improved VMD-adaptive wavelet threshold combined with noise reduction in this paper. Firstly, the modal components are obtained based on VMD decomposition; Secondly, the dual determination criteria of sample entropy and correlation coefficient are constructed to filter the components; Subsequently, an adaptive wavelet thresholding function is proposed, and quadratic noise reduction is applied to mixed IMFs, which in turn reconstructs each component to achieve joint noise reduction. Finally, based on traditional machine learning and deep learning diagnosis methods, the features of noise reduction signals are extracted to realize fault diagnosis. By verifying and analyzing the simulated signal with the measured signal, noise components, the expression of fault characteristics, and the accuracy of fault diagnosis are eliminated, enhanced, and improved.

Tool wear condition monitoring plays an important role in the maintaining machining accuracy and machining efficiency of complex surface parts. In this study, a new on-line tool wear monitoring method based on a self-developed data processing approach for the impeller milling was proposed. To achieve that, a new tool wear experimental platform was first built to collect both the spindle current signal and thermal deformation data in entire life cycle of cutter. Based on collected data, features of the time domain, frequency domain and time-frequency domain were extracted indiscriminately, and a 38 × 156 feature-sample set was subsequently established. To further reduce the dimensions of this feature-sample set and rise its characterization capability, the feature set was further processed using the sensitivity analysis and deep auto-encoder algorithm. Finally, 12 synthesized features were filtered out and then used to build the mapping model of signal synthesized features to different tool wear conditions by adopting the structural artificial neural network (ANN) integrated with back propagation (BP) algorithm. To verify the reliability of the proposed BP-ANN-integrated tool wear condition monitoring model, another comparison analysis of different data processing approaches was conducted. The comparison results showed that the proposed method for tool wear condition online monitoring had reliable performance and the recognition accuracy was 88.9%.

As a fundamental element in industrial fluid transportation, the main pump fulfills an irreplaceable function in critical infrastructure, including the energy, water conservancy, petrochemical, and sewage treatment industries. As the core component of key power equipment, its operating condition is intrinsically connected to the safety, stability, and reliability of the entire system. This paper provides a systematic review of the latest advances in fault mechanism analysis and diagnosis methods for main pump systems. First, the typical structural composition and functional characteristics of the main pump system are examined, and the occurrence mechanisms and evolution rules of typical faults, such as mechanical malfunctions and performance degradation caused by hydraulic imbalance, are discussed in detail. Second, the main technical approaches to fault diagnosis are summarized and reviewed, including diagnosis methods based on signal processing, modeling, data-driven techniques, and multi-source information fusion. The advantages, limitations, and application scopes of these approaches are comparatively analyzed. On this basis, the development trends in main pump fault diagnosis technology and the key challenges faced—such as strong noise, small sample size, and multiple fault coupling—are identified and discussed. Finally, future research prospects are put forward in view of the limitations of current research. This review aims to provide theoretical insights and technical support for advancing condition monitoring, fault diagnosis, and health management of main pump systems.