With the acceleration of the mining process, the goaf has become one of the main sources of danger in underground mines, seriously threatening the safe production of mines. To make an accurate prediction of the risk level of the goaf quickly, this paper optimizes the features of the goaf by correlation analysis and feature importance and constructs a combination of feature parameters for the risk level prediction of the goaf to solve the problem of redundancy of evaluation indexes. Multiple machine learning algorithms are applied to 121 sets of goaf data respectively, and the optimal algorithm and the best combination of feature parameters are obtained by evaluating the mining area with multiple indicators such as accuracy and kappa coefficient. The best combination of features parameters are ground-water, goaf layout, volume of goaf, goaf volume, span-height ratio, and mining disturbance, and the optimal algorithm is Extra Tree (ET), which needles the goaf risk level prediction problem with the accuracy of 94%. This model can be used to solve the problem of how to quickly and accurately predict the risk level of the goaf.
The number of express boxes worldwide exceeded 170 billion in 2021, and, from several regions in China, tested positive. Therefore, it is important to study the transmission of viruses through express boxes. In this paper, we establish a model of express box virus transmission based on comprehensive consideration of environmental factors, such as temperature, disinfection, humidity, virus release intensity, and volume of vehicle, to study the transmission of express box virus, and explore the spatial and geographic spread variation of express box viruses in China. Several important findings emerged from the study, including: (1) Disinfection can prolong the spread of viruses in the express box for ≥21 h; (2) For every 1 °C rise in temperature, the infected time can be prolonged by ≥1.2 h, and for every 10% rise in relative humidity, the virus transmission time can be prolonged by ≥1.32 h; (3) In an environment suitable for virus transmission, when loaded with 1000, 2000, 4000 express boxes, areas where the express delivery time exceeds 22.56, 18, 14.64 h will face the risk of all the boxes in the carriage being infected. These findings could help public health departments prevent the risk of virus transmission from express boxes.
Loosening of bolt connections in industrial pipeline systems poses significant risks to structural integrity and operational safety. However, conventional detection methods often suffer from low efficiency and poor robustness in complex environments. To address these challenges, this study proposes a lightweight and real-time bolt loosening detection framework based on active guided waves and multi-channel piezoelectric sensing, enhanced by advanced deep learning techniques. Specifically, an eight-channel piezoelectric sensor array is used to capture guided wave responses, which are transformed into two-dimensional (2D) representations via multi-scale feature fusion and local enhancement to facilitate deep learning. A total of 34 complex loosening scenarios—including single, adjacent, diagonal, and multi-bolt combinations—are experimentally simulated under diverse noise conditions to emulate real-world disturbances. An improved ResNet18 architecture is developed by integrating a multi-head attention mechanism for capturing long-range dependencies, along with a Squeeze-and-Excitation (SE) module for adaptive channel recalibration. Experimental results show that the proposed model achieves 99% detection accuracy under noisy conditions, with inference latency ranging from 7.2 to 10.1 ms and a throughput of 114–163 FPS (frames per second), fulfilling real-time requirements. Ablation studies confirm the effectiveness of the attention and SE components. Compared with deeper models such as ResNet50 and VGG16, the proposed method significantly reduces parameter count (13.02M) while maintaining competitive performance, enabling efficient edge deployment. Furthermore, few-shot learning experiments demonstrate that over 90% accuracy can be achieved with only five training samples in previously unseen working conditions. This research provides a robust, efficient, and scalable solution for intelligent structural health monitoring of pipeline bolt assemblies and offers valuable insights for fault diagnosis under complex industrial noise environments.
Predicting mining disaster risk levels is a critical component of intelligent mining systems. This study utilizes five common mining disaster datasets to predict various risk levels. By analyzing correlation coefficients and feature importance for each dataset, optimal evaluation indicators are identified. The Shapley Additive Explanations model is then applied to enhance interpretability. To address the presence of outliers and imbalanced data categories, the Mahalanobis Distance Discriminant Method and the Synthetic Minority Oversampling Technique algorithm based on Tomek Links are used for data preprocessing. Subsequently, Support Vector Machine, Random Forest, Extreme Gradient Boosting, one-dimensional Convolutional Neural Networks, and multi-Grained Cascade Forest algorithms are applied to the five mining disaster datasets. Comparative analysis reveals that the Deep Forest algorithm demonstrates superior performance and generalization in predicting stability levels of goaf, slope stability, rockburst intensity levels, pillar stability, and Hanging Wall stability, with prediction accuracies of 92.31%, 96.77%, 92.50%, 91.67%, and 95.00%, respectively. This research provides a systematic solution for mining disaster classification prediction, offering technical support and a scientific theoretical basis for intelligent mining development and mining safety operations.