Search

Underground personnel behavior recognition is an important measure to ensure safe production in coal mines. The existing research on behavior recognition of underground personnel lacks research and analysis on the perception mechanism, and the feature extraction method is simple. In order to solve the above problems, a behavior recognition method for underground personnel based on fusion networks is proposed. The method mainly includes three parts: data preprocessing, feature construction, and recognition network construction. Data preprocessing: the collected channel status information (CSI) data is processed through CSI quotient models, subcarrier denoising, and discrete wavelet denoising to reduce the impact of environmental noise and equipment noise. Feature construction: the processed data is transformed into images using the Gramian angular summation/difference fields (GASF/GADF) to preserve the spatial and temporal features of the data. Recognition network construction: according to the features of personnel actions, a fusion network composed of a gate recurrent unit (GRU) based encoding and decoding network and a multiscale convolutional neural network (CNN) is proposed. GRU is used to preserve the correlation between pre and post data. The weight allocation strategy of the attention mechanism is used to effectively extract key features to improve the accuracy of behavior recognition. The experimental results show that the average recognition accuracy of this method for eight movements, namely walking, taking off a hat, throwing things, sitting, smoking, waving, running, and sleeping, is 97.37%. The recognition accuracy for sleeping and sitting is the highest, and the most prone to misjudgment are walking and running. Using accuracy, precision, recall, and F1 score as evaluation indicators, it is concluded that the performance of the fusion network is superior to CNN and GRU. The accuracy of personnel behavior recognition is higher than the HAR system, WiWave system and Wi-Sense system. The average recognition accuracy of walking and taking off a hat at normal speed is 95.6%, which is higher than 93.6% for fast motion and 92.7% for slow motion. When the distance between transceiver devices is 2 m and 2.5 m, the recognition accuracy is higher.

In the application of mobile edge computing(MEC) in intelligent mine, the mobile users unload tasks to non-optimal edge servers due to unreasonable resource allocation, which leads to extra transmission time and execution delay, thus resulting in the decrease of the total task completion rate. In order to solve the above problem, a two-dimensional dynamic matching algorithm based on preference is proposed to optimize the resource allocation decision in MEC system. The data of the position of a mobile user in MEC system and the calculation amount required by a task in one time slot is sent to the edge server. The preference table of the edge server for the mobile user is formed according to the set preference value. At the same time, the preference table for all the edge servers is formed by the mobile user according to different physical distances. The two preference tables are combined to form a two-dimensional dynamic preference table, which is abstracted into a two-dimensional matrix. The two-dimensional matrix is processed by a two-dimensional dynamic matching algorithm based on preference, and the matching optimization results of mobile users and edge servers are obtained. The simulation results show that compared with the conventional MEC scene unloading algorithm, the preference-based two-dimensional dynamic matching algorithm can effectively alleviate the problem of the decrease of the total task completion rate in a large number of sudden task scenes, and can achieve the total task completion rate of more than 60% in extreme cases.

The difficulty in fault diagnosis of mine hoist inverters lies in extracting the features that characterize faults. At present, signal processing methods are mainly used to obtain fault statistical features, or the fault depth features are extracted by neural networks. In the actual working environment, the hoist inverter is affected by factors such as background noise and load changes. Therefore, it is difficult to obtain features that can characterize the faults effectively by using a single feature extraction method, resulting in low fault diagnosis accuracy of the hoist inverter. In order to solve the above problems, a fault diagnosis method of hoist inverter based on the fusion of statistical features and depth features is proposed. Firstly, the Hilbert-Huang transform(HHT) is used to conduct modified ensemble empirical mode decomposition(MEEMD) of the inverter output current signal so as to obtain the fault statistical features. At the same time, the squeeze and excitation with densely connected convolutional network(SE-DenseNet) is used to extract the depth features of the output current signal. Secondly, the local fisher discriminant analysis(LFDA) is used to perform fusion and dimensionality reduction processing on the combination of the two features to obtain low-dimensional fusion features of statistical features and depth features. Finally, the low-dimensional fusion features are input to the extreme learning machine to obtain inverter fault classification. Experiments are conducted for a single IGBT open-circuit fault in the hoist inverter. The results show that the low-dimensional fusion features obtained by this method are more capable of fault characterization than single features, which improves the fault recognition accuracy effectively.

In coal mine working face, there are problems such as uneven energy distribution of wireless monitoring network, the difficulty of manual maintenance and node battery replacement and energy limitation of some monitoring nodes. In order to solve the above problems, an energy optimization method of wireless monitoring network in coal mine working face based on mobile nodes is proposed by using the characteristics of periodic movement of the shearer. This method installs mobile nodes on the shearer to forward data that collected by monitoring nodes in the network to reduce multi-hop data transmission, and reduces redundant data transmission by setting transmission thresholds to reduce node energy consumption. The mobile nodes are used to radiate radio frequency energy to the monitoring nodes, which provides targeted energy supplement to the monitoring nodes so as to achieve energy balance of the network nodes. The experimental results show that compared with LEACH protocol, this method can make the wireless monitoring network energy distribution more balanced, reduce the network energy consumption and monitoring node mortality rate effectively under the same operation round, and extend the effective life of the network by 2.8 times.

Simple Summary: This review highlights the intricate interactions between two major constituents of marine ecosystems: dinoflagellates and heterotrophic bacteria. Dinoflagellates typically associate with a diverse array of heterotrophic bacteria, which are not limited to but predominantly consist of phyla such as Alphaproteobacteria, Gammaproteobacteria, and the Cytophaga–Flavobacterium–Bacteroides group. These bacterial communities engage with dinoflagellates in a multifaceted manner, including nutrient exchange, the secretion of pathogenic substances, and involvement in the synthesis of chemical entities. Crucially, the process of horizontal gene transfer from bacteria is a significant force sculpting the genomic architecture of dinoflagellates. The objective of this review is to shed light on the dynamic interactions between dinoflagellates and their bacterial partners, examining the relationship through a physiological, ecological, and evolutionary perspective. Dinoflagellates and heterotrophic bacteria are two major micro-organism groups within marine ecosystems. Their coexistence has led to a co-evolutionary relationship characterized by intricate interactions that not only alter their individual behaviors but also exert a significant influence on the broader biogeochemical cycles. Our review commenced with an analysis of bacterial populations, both free-living and adherent to dinoflagellate surfaces. Members of Alphaproteobacteria, Gammaproteobacteria, and the Cytophaga–Flavobacterium–Bacteroides group are repeatedly found to be associated with dinoflagellates, with representation by relatively few genera, such as Methylophaga, Marinobacter, and Alteromonas. These bacterial taxa engage with dinoflagellates in a limited capacity, involving nutrient exchange, the secretion of pathogenic substances, or participation in chemical production. Furthermore, the genomic evolution of dinoflagellates has been profoundly impacted by the horizontal gene transfer from bacteria. The integration of bacterial genes into dinoflagellates has been instrumental in defining their biological characteristics and nutritional strategies. This review aims to elucidate the nuanced interactions between dinoflagellates and their associated bacteria, offering a detailed perspective on their complex relationship. [ABSTRACT FROM AUTHOR]

Global climate change intensifies the water cycle and makes freshest waters become fresher and vice‐versa. But how this change impacts phytoplankton in coastal, particularly harmful algal blooms (HABs), remains poorly understood. Here, we monitored a coastal bay for a decade and found a significant correlation between salinity decline and the increase of Karenia mikimotoi blooms. To examine the physiological linkage between salinity decreases and K. mikimotoi blooms, we compare chemical, physiological and multi‐omic profiles of this species in laboratory cultures under high (33) and low (25) salinities. Under low salinity, photosynthetic efficiency and capacity as well as growth rate and cellular protein content were significantly higher than that under high salinity. More strikingly, the omics data show that low salinity activated the glyoxylate shunt to bypass the decarboxylation reaction in the tricarboxylic acid cycle, hence redirecting carbon from CO2 release to biosynthesis. Furthermore, the enhanced glyoxylate cycle could promote hydrogen peroxide metabolism, consistent with the detected decrease in reactive oxygen species. These findings suggest that salinity declines can reprogram metabolism to enhance cell proliferation, thus promoting bloom formation in HAB species like K. mikimotoi, which has important ecological implications for future climate‐driven salinity declines in the coastal ocean with respect to HAB outbreaks. [ABSTRACT FROM AUTHOR]