Search

With the increasingly widespread application of deep-learning technology in the field of coal mines, the image recognition of mine water inrush has become a hot research topic. Underground environments are complex, and images have high noise and low brightness. Additionally mine water inrush is accidental, and few actual image samples are available. Therefore, this paper proposes an algorithm that recognizes mine water inrush images based on few-shot deep learning. According to the characteristics of images with coal wall water seepage, A bilinear neural network was used to extract the image features and enhance the network's fine-grained image recognition. First, features was extracted using a bilinear convolutional neural network. Second, the network was pre-trained based on cosine similarity. Finally, the network was fine-tuned on the predicted image. For single-line feature extraction, the method is compared with big data and few-shot learning. According to the experimental results, the recognition rate reaches 95.2% for few-shot learning based on bilinear neural network, thus demonstrating its effectiveness.

IntroductionAlthough child and adolescent health is the core of the global health agenda, the cause of death and its expected contribution to life expectancy (LE) among those aged 5–14 are under-researched across countries, especially in low- and middle-income countries (LMICs).MethodsDeath rates per 10 years age group including a 5–14-year-old group were calculated by the formula, which used the population and the number of deaths segmented by the cause of death and gender from the 2019 Global Burden of Disease (GBD) study. LE and cause-eliminated LE in 10-year intervals were calculated by using life tables.ResultsIn 2019, the global mortality rate for children and adolescents aged 5–14 years was 0.522 (0.476–0.575) per 1,000, and its LF was 71.377 years. In different-income regions, considerable heterogeneity remains in the ranking of cause of death aged 5–14 years. The top three causes of death in low-income countries (LICs) are enteric infections [0.141 (0.098–0.201) per 1,000], other infectious diseases [0.103 (0.073–0.148) per 1,000], and neglected tropical diseases and malaria [0.102 (0.054–0.172) per 1,000]. Eliminating these mortality rates can increase the life expectancy of the 5–14 age group by 0.085, 0.062, and 0.061 years, respectively. The top three causes of death in upper-middle income countries (upper MICs) are unintentional injuries [0.066 (0.061–0.072) per 1,000], neoplasm [0.046 (0.041–0.050) per 1,000], and transport injuries [0.045 (0.041–0.049) per 1,000]. Eliminating these mortality rates can increase the life expectancy of the 5–14 age group by 0.045, 0.031, and 0.030 years, respectively.ConclusionThe mortality rate for children and adolescents aged 5–14 years among LMICs remains high. Considerable heterogeneity was observed in the main causes of death among regions. According to the main causes of death at 5–14 years old in different regions and countries at different economic levels, governments should put their priority in tailoring their own strategies to decrease preventable mortality.

Background Alcohol use disorders (AUD) has long been one of the most disability mental disorders and a major cause of health loss. Methods Based on open access data from the 2019 Global Burden of Disease (GBD 2019) study, we extracted data of years lived with disability (YLD), years of life lost (YLL) and disability-adjusted life years (DALY) to describe the changes of AUD burden over the period of 1990-2019 stratified by sex in globe, high-income countries (HICs), upper-middle income countries (UMCs), lower-middle income countries (LMCs), low-income countries (LICs) and China. We used Joinpoint regression model to fit the changing trend of years. And pairwise comparison was applied to test the coincidence parallelism and judge whether the difference of the trend among different regions is statistically significant. Results LMCs experienced the largest changes in the YLD rate of AUD from 1990 to 2019 (average annual percent change (AAPC) = -0.7, 95% confidence interval (CI) = -0.8, -0.7, P

Pneumococcal diseases (PDs) pose a serious health threat to children. Vaccination is the most cost-effective intervention to prevent PDs, but pneumococcal vaccines coverage among children is low in China.This study investigated the willingness of children's caregivers to have their children vaccinated with pneumococcal vaccines under an innovative policy to offer 1-dose of the 13-valent pneumococcal conjugate vaccines at no charge to families. The research found that 70.51% of caregivers were willing to have their infants receive pneumococcal vaccines and that reducing the cost of vaccines may increase caregivers' willingness.This is the first evaluation in China of acceptance of pneumococcal vaccines among children under a 1-dose, cost-free policy. The results provide scientific evidence for updating local and national pneumococcal immunization strategies to promote the use of the pneumococcal vaccine.

Coal-floor water-inrush incidents account for a large proportion of coal mine disasters in northern China, and accurate risk assessment is crucial for safe coal production. A novel and promising assessment model for water inrush is proposed based on random forest (RF), which is a powerful intelligent machine-learning algorithm. RF has considerable advantages, including high classification accuracy and the capability to evaluate the importance of variables; in particularly, it is robust in dealing with the complicated and non-linear problems inherent in risk assessment. In this study, the proposed model is applied to Panjiayao Coal Mine, northern China. Eight factors were selected as evaluation indices according to systematic analysis of the geological conditions and a field survey of the study area. Risk assessment maps were generated based on RF, and the probabilistic neural network (PNN) model was also used for risk assessment as a comparison. The results demonstrate that the two methods are consistent in the risk assessment of water inrush at the mine, and RF shows a better performance compared to PNN with an overall accuracy higher by 6.67%. It is concluded that RF is more practicable to assess the water-inrush risk than PNN. The presented method will be helpful in avoiding water inrush and also can be extended to various engineering applications.

Coal-floor water-inrush accidents have become increasingly serious with the increase of mining depth, threatening mining safety in China. This study presents a mathematical assessment method for coal-floor water-inrush risk integrating the hierarchy-variable-weight model (HVWM) with collaboration-competition theory (CCT). Compared with the variable-weight model (VWM) which adjusts weights of main controlling factors only once, the HVWM can readjust the weights of the factors according to assessment needs. The CCT embodies the relationship among multiple controlling factors that influence coal-floor water-inrush. The HVWM–CCT model is applied to the eastern mining area of the Liuzhuang Coal Mine in South China. First, variable-weights of bottom-layer main controlling factors are calculated based on the VWM. Second, top-layer main controlling factors are constructed according to the CCT. Third, variable-weights of top-layer main controlling factors are calculated using the HVWM. Subsequently, the HVWM–CCT model is established to classify risk grades, and the No. 1 coal-floor water-inrush risk in the eastern mining area of the Liuzhuang Coal Mine is divided into five grades. Then, the VWM model is contrasted with the HVWM–CCT model, and a differences table is obtained between the two models. Finally, the verification of water-inrush points, revealed points, and observation results indicate that the HVWM–CCT model is more reasonable compared with the VWM model. The HVWM–CCT model has great potential for assessing the risk of coal-floor water-inrush.

Capturing long-range dependencies has proven effective on video understanding tasks. However, previous works address this problem in a pixel pairs manner, which might be inaccurate since pixel pairs contain too limited semantic information. Besides, considerable computations and parameters will be introduced in those methods. Following the pattern of features aggregation in Graph Convolutional Networks (GCNs), we aggregate pixels with their neighbors into semantic units, which contain stronger semantic information than pixel pairs. We designed an efficient, parameter-free, semantic units-based dependencies capturing framework, named as Multi-semantic Long-range Dependencies Capturing (MLDC) block. We verified our methods on large-scale challenging video classification benchmark, such as Kinetics. Experiments demonstrate that our method highly outperforms pixel pairs-based methods and achieves the state-of-the-art performance, without introducing any parameters and much computations.

Evacuation planning that navigates evacuees to escape significantly improves the safety of the miners at the mine water inrush conditions. However, conventional methods do not consider the effect of the water flow on the transit time when determining escape paths for evacuees. In the present study, a novel method is proposed to find the earliest arrival path through a time-varying network for the evacuation planning of the mine water inrush. Firstly, the effect of the water flow on the transit time is formulated in the objective function based on the spatiotemporal results of the numerical simulation. Secondly, the time-varying network and the earliest arrival path are defined accordingly, where the edge weight of the network representing the transit time of the corresponding roadway varies over time. Moreover, the earliest arrival path is a path with the minimum arrival time of the destination for each source with a starting time, in which waiting is allowed to trade off a wait for traversing a roadway in less time. Thirdly, since the conventional shortest path methods have limitations on searching shortest paths through a time-varying network, the time-varying earliest arrival path (TVEAP) algorithm is proposed for the improvement. In order to evaluate the validity and feasibility of the proposed algorithm, mathematical proof and experiments are carried out. The results show that the proposed algorithm can effectively find the earliest arrival path to the destination following the evacuation strategy. Moreover, it is found that the corresponding time efficiency meets the real-time requirement of the evacuation planning with the capability of 250 evacuees on a quad-core desktop.

Water inrush is a kind of serious hazards for underground mining works. Moreover, it is extremely difficult to be predicted. Designing the rescue plan is also difficult being ignorant of the flux or even the exact location of water inrush. In this paper, a recognition procedure is introduced to determine the location and flux of water inrush using its spreading process in underground mine. Four steps are included in this procedure. Firstly, a set of water inrush is created that involves the possible cases obtained according to hydrogeologic conditions and mining works. Secondly, the spreading process of inrush water is computed. Then clusters of water spreading processes and water inrush cases are created on account of similarities. Lastly, recognition algorithms are used to determine which cluster includes the most similar cases with the happened one represented using monitor data. As results, the location and flux can be determined using this recognized cluster. This proposed recognition procedure is illustrated to be efficient in a coal mine in the North China. The similar cases in the set of water inrush can be fast recognized to determine the flux and the location. This recognition process provides a framework for rapid response after water inrush happens.

Water inrush from a karst aquifer threatens mine safety in North China because of the special hydrogeological conditions. This paper systematically analyzes the regional hydrogeological conditions of the Cambrian–Ordovician limestone karst aquifer in the Yuxian Basin. Conceptual and mathematical models of karst water flow system are established, and the karst water flow system is simulated by the finite-difference method. The numerical model is checked using hydraulic head from karst water observation holes, and the sensitivity of hydrogeological parameters is analyzed. Further, the risk of water inrush from the karst aquifer underlying the coal seam in the second mining area of the Beiyangzhuang Mine in the basin is evaluated by the water inrush coefficient method. Based on the critical hydraulic head for mining safety, the rate of mine dewatering is simulated for the three stages of the mining plan with the numerical model. The results show that the risk of water inrush in the second mining area is high. The rate of dewatering in the three stages without grouting is 1300, 1860, and 2050 m3/h, respectively, and with local grouting is 1020, 1550, and 1700 m3/h, respectively. Dewatering combined with grouting not only ensures coal mine safety, but also significantly reduces the rate of dewatering. The prediction method of mine dewatering has practical significance in engineering applications.